MXPA05013303A

MXPA05013303A - Method and apparatus for processing mail to obtain image data of contents

Info

Publication number: MXPA05013303A
Application number: MXPA/A/2005/013303A
Authority: MX
Inventors: T Dewitt Robert; L Hayduchok George
Original assignee: Dewitt Robert R; L Hayduchok George
Priority date: 2003-06-07
Filing date: 2005-12-07
Publication date: 2006-12-13

Abstract

A method and apparatus for processing mail is provided. Mail is placed into an input bin having a conveyor that conveys the mail towards a feeder. The feeder serially feeds the envelopes by engaging the lead envelope in the stack of mail and displacing the lead envelope transverse the stack of mail. The mail is then cut on a side edge and the top edge to cut open each envelope. A transport conveys the cut envelopes to an extractor. The extractor opens the edge-severed mail and presents the contents of the envelopes to an operator who manually extracts the contents. The operator drops the extracted contents onto a conveyor that conveys the contents to an imaging station. The contents are automatically separated and imaged to obtain image data for the contents. The contents are then sorted into a plurality of output bins.

Description

Published: (15) Information about Correction: - without intemational search report and to be lepubltshed see PCT Gfazette No 07/2005 of 17 February 2005, Secupon receipi ofthat leportion p Fot two-letter codes and other abbreviations, refet to the "Gutd - (48) Date of publication of tliis correctcd version: an e Notes on Codes ndAbbreviatwns "'appeai ing at Úie begin- 17 February 2005 nmg ofeach regular tssue ofthe PCTGazette METHOD AND APPARATUS FOR PROCESSING MAIL TO OBTAIN CONTENT IMAGE DATA Field of the Invention The present invention relates to the field of document processing to obtain image data for documents. More specifically, the present invention relates to an operable workstation for processing documents by dropping the contents onto a conveyor that transports the documents to a scanning device. The present invention also relates to mail processing, such as envelopes having content, where open envelopes are presented to an operator, so that the operator can extract the contents of the envelopes.

BACKGROUND OF THE INVENTION Automated and semi-automated machines have been employed to process documents as bulky mail. Due to the large amount of mail received by many companies, there has long been a need to efficiently classify incoming mail. In addition, in many cases it is desirable to obtain image data of the documents received in the mail.

However, documents are often folded and / or documents received in the mail batch are random in size. To explore these documents it is typically necessary to invest considerable time to manually prepare the documents, so that the documents can be automatically fed into a scanning device. This processing is inefficient, therefore increasing the total cost of processing the mail and obtaining image data of the documents in the mail.

SUMMARY OF THE DESCRIPTION Accordingly, a method and apparatus for processing mail are provided to efficiently manage the mail and obtain image data of the documents. Preferably, the present invention provides an apparatus that easily processes a variety of document types to obtain image data of documents. For example, the apparatus preferably accepts documents of different sizes and transports the documents to an image forming device, such as a scanning device, to obtain image data. Preferably, the apparatus also allows the processing of folded or creased documents, and particularly groups of folded or sharpened documents. Also provided is a method and apparatus that allows to process mixed mail with minimal or no mail preparation. A document or package of documents can simply be dropped on a conveyor, which transports the document or documents to a feeder that feeds the stamp and documents downstream for further processing. One type of additional processing uses a scanning device. The feeder separates and feeds the scanning devices in series, which scans the documents to obtain image data of each document. According to another aspect, the method and apparatus are provided to efficiently process documents according to different information that can be identified for the documents based on work, batch, transaction and document. In addition, an interface is provided that allows the operator to intervene to supplement or amend the information scanned with additional information that may be marked on the documents. Another aspect relates to a feeder that operates to accommodate mail with little or no preprocessing. The feeder is operated to receive packages and wrappings transported in a generally horizontal orientation, separating the documents from each other and feeding in series the documents. The feeder includes a feed belt positioned generally parallel to a transport path along with the pack of documents that are transported. The feeder also includes an angled feed arm that projects up and down over the transport path, forming an acute angle with the input transport path.

DESCRIPTION OF THE DRAWINGS The above summary and the following description will be better understood when read in conjunction with the Figures in which: Figure 1 is a perspective view of a mail processing apparatus. Figure 2 is a fragmentary, perspective view of a section of the feeder of the apparatus illustrated in Figure 1. Figure 3 is a fragmentary plan view of the sections of the gravity feeder and the image forming section of the apparatus. illustrated in the Figure 1. Figure 4 is a fragmentary, cross-sectional view of the sections of the gravity feeder and the imaging station illustrated in Figure 3. Figure 5 is a fragmented, amplified view of the sections of the feeder by gravity and from the image forming station illustrated in Figure 4. Figure 6 is a schematic side view of the sections of the gravity feeder of the image forming section illustrated in Figure 3. Figure 7 is a view schematic drawing of the sections of the gravity feeder and the image forming station illustrated in Figure 3. Figure 8 is a fragmentary perspective view of the sections of the gravity feeder and of the imaging station illustrated in the Figure 3, which illustrates the section of image formation in an open configuration. Figure 9 is a side view, fragmented, amplified, of an extraction station of the apparatus illustrated in Figure 1. Figure 10 is a schematic side view of a scraper element of the imaging station illustrated in Figure 3. Figure 11 is a screen plan of a visual representation of the apparatus illustrated in Figure 1, illustrating a view classified according to the transaction. Figure 12 is a screen plan of a visual representation of the sides of the apparatus illustrated in Figure 1, illustrating a view classified according to the document sequence. Figure 13 is a screen plan of a visual representation of the apparatus illustrated in Figure 1, illustrating a view classified according to the tray. Figure 14 is a screen plan of a visual representation of the apparatus illustrated in Figure 1, illustrating a view classified according to the transaction. Figure 15 is a fragmented, perspective view of the gravity feeder illustrated in Figure 3. Figure 16 is a fragmented, fragmented, perspective view of the feeder of Figure 2.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT Referring now to the Figures in general and to Figure 1 in particular, a semi-automated mail processing workstation 10 is illustrated.

The workstation 10 processes mail by cutting one or more edges of each envelope in a mail stack, and presenting the envelopes with the cut edge one at a time to an operator who removes the documents from the envelope with the hand. The operator can then drop the individually extracted documents or stack them on a conveyor that transports the documents to an image forming station. The image forming station separates the documents, feeding in series the documents to an imager who obtains image data for the documents. The documents are then classified into one or more output trays.

Brief Overview A general view of the mail flow is as follows. Initially, a stack of envelopes containing documents, referred to as work, is placed in an inbox. A feeder 30 removes the front envelope 5 from the front of the stack and transfers the envelope to a feeder tray. The envelope 5 on the feeding tray is justified by the edge by a plurality of opposite rollers. From the feed tray, the envelope 5 falls towards a side cutter, which cuts the side edge of the envelope if desired. Of the lateral cutter, the envelope falls towards a ramp. The ramp moves vertically to adjust the height of the upper edge of the envelope to consider variations in the height of the different envelopes in the job. The ramp moves vertically until the height of the upper edge of the envelope 5 is within an acceptable range to advance the envelope toward the upper cutter. The envelope is then transported to the upper cutter, which cuts the upper edge of the envelope 5. The upper cutter of the envelope is advanced to an extraction station 70. The extraction station 70 separates the front and back sides of the envelope to present the contents of the envelope to remove it. An operator then manually removes the contents of the envelope 5. After the operator removes the documents from the envelope 5, the apparatus 10 automatically advances the envelope to a verifier 90. The verifier 90 verifies that all documents are removed from the envelope before that the envelope is discarded. The verifier 90 the envelope is transported to a waste container. Alternatively, the envelope 5 can be manually removed and subjected to imaging at the image forming station 130. Returning to the flow of the documents, after the documents are extracted at the extraction station, the operator unfolds what is necessary and drops or places the extracted documents on a gravity conveyor 100 that transports the documents to an image forming station 130. An imaging input feeder 110 receives the documents from the gravity conveyor 100 and controls the feeding of the documents to the image forming station 130. The image forming input feeder 110 is configured to receive and feed documents of various sizes and conditions. For example, documents are often folded into an envelope. When the documents are extracted and opened, the documents are folded to fold so that they do not remain flat. The feeder 110 is preferably configured to receive those wrinkled or folded documents and serially feed the folded documents into the image forming station 130 with minimal manual preparation by the operator. The image forming station 130 includes an image former 150 which obtains image data for each document as the document is transported along the device. For example, preferably the imager 150 is a scanning device that obtains gray scale or color image data representing an image of each document. The scanning device scans each document at a plurality of points when the document is transported along the scanning device. The information for each document is stored in a data file for each document, so that the image data can be accessed some time later. Of the image forming device, preferably an image forming conveyor 170 transports the documents to a sorting station 200 that sorts the documents into a plurality of output trays 205. The documents can be classified in a variety of ways. For example, the documents may be classified on a basis of the document information obtained from the image data received at the image forming station 130. Alternatively, the operator may indicate information with respect to the document before it is scanned. so that the document is classified according to the information indicated by the operator. Another alternative is that the documents can be stacked in one or more trays simply on the basis of the order in which the documents are processed. Since many of the documents may be wrinkled, documents will typically not be easily stacked in a compact manner, since relatively fewer wrinkled documents can be downloaded onto a tray before the tray is filled. In consecuense, preferably, the documents are processed by a scrubber 180, which is an element that reduces wrinkles or creases in the documents. The scrubber flattens or stretches the documents so that they are flatter on the output trays, so documents can be downloaded to a tray before the tray is full. A controller controls the processing of the mail in response to signals received from different sensors at various places on the workstation 10 and in response to parameters set for work by the operator. For example, in response to an indication of a sensor in the feed tray that there are no envelopes in the feed tray, the controller sends a signal to the feeder 30 indicating that an envelope of the input tray should be fed to the tray. of food. Similarly, in response to an indication of a sensor on the ramp that there is no over on the ramp, the controller sends a signal to the feed tray indicating that an envelope must be dropped from the feed tray to the ramp . The work station is divided into numerous functionally separated sections, which include: the feeding station 30, a side cutting station, a top cutting station, the extraction station 70, the verification station 90, the training station of images 110, and sorting station 200. In many cases the controller controls the operation of the different sections independently of each other. This independence allows the different operations to proceed simultaneously or asynchronously when required. As a result, a decrease in speed in one section does not necessarily decrease the speed of all the other sections. Furthermore, preferably the operation of the gravity conveyor apparatus through the sorting station is controlled separately from the operation of the other stations. Furthermore, preferably, an operator interface is provided so that the operator can intervene to control the processing of the documents. Specifically, a touch screen display 20 is preferably provided which allows the operator to enter different information with respect to the documents.

Workstation Configuration As can best be seen in Fig. 1, preferably workstation 10 is configured so that an operator working on the workstation has easy access in each work area. A sitting area 15 on the front of the apparatus is located in the center, and the different stations are positioned around the seating area with the paper path traced in such a way that the documents remain within easy access of the operator for the seating area. Specifically, preferably, the feeding station 30 is placed on the right side, however, the feeding station can be located on the left side. Preferably the feeding station is within the reach of the operator's arm from the seating area 15. Consequently, preferably the distance from the seating area to the feeding station is not greater than half the total station width. of work, so that the operator can easily access the feeding station 30 of the seating area. From the feed station 30, the mail pieces are fed along a document path extending through the work station across the width of the work station, intermediate at the front and rear edges of the workstation. work station to the extraction station 70. Preferably, the extraction station is substantially aligned with the seating area 15 in relation to the right and left edges of the work station so that the operator can easily hold the mail in the extractor during the operation. For example, the extraction station is generally centered between the right and side edges of the work station, and preferably the center of the sitting area is also generally centered between the right and left edges of the work station. In other words, preferably the seating area is positioned at a distance from the left edge of the work station that is substantially similar to the distance at which the extraction station is positioned from the right edge of the work station. *** The gravity conveyor 100 is preferably adjacent to the front edge of the work station, and is positioned between the extraction station 70 and the seating area 15 so that the operator reaches over a portion of the gravity conveyor to hold documents in the extraction station. More specifically, preferably the portion of the gravity conveyor 100 is positioned adjacent the seating area 15 at the front edge of the work station. In this way the operator can see, unfold and easily drop the document from the extraction station 70 on the gravity conveyor 100 when pulling the documents towards himself. Preferably the imaging station is positioned adjacent the front edge of the work station on the side of the work station opposite the side on which it is placed to the feed station 30. Consequently, the gravity conveyor 100 carries the documents that were dropped away from the seating area 15, adjacent, along a path that is generally parallel to the front edge of the work station. Preferably, the image forming station 130 is positioned at a distance from the sitting area that is less than half the width of the work station so that the operator can easily access the documents to the image forming station. Preferably, the document path of the image forming station 130 to the output trays 205 returns to the sitting area, so that the output trays are separated from the sitting area at a distance that is less than half the width of the work station. In this way, the exit trays 205 are conveniently positioned close to the operator in the seating area, so that the operator can easily move processed documents from the exit trays while the operator is in the sitting area.

Station Details Power Stations and Edge Cutting The power station 30 includes an input tray and a feeder. The input tray is configured to receive a stack of mail and carry it to the feeder. The feeder comprises a swivel arm with a suction cup that holds an envelope of the mail stack and transports the piece to a side cutting station. In this way, the feeder 35 feeds mail from the mail stack in series. The side cutting station includes a plurality of opposite tension rollers and drive rolls. When the envelope passes between the rollers, a rotating blade cuts the side edge of the envelope.

The cut edge falls down from a release chute into a waste or debris container. From the side cutting station, the envelopes are justified by the top edge so that the top edge remains at a consistent height. The envelopes can be justified by a pair of rollers to drive the envelopes up against a stop at a predetermined height. However, a justifying roller is typically limited to justify envelopes that are of similar height. If there is too much variation between the envelopes in a mail batch the justifier may not be able to properly justify the envelopes. For example, if an envelope in a lot is unusually high, the top edge of the envelope can be placed too high so that it enters the justifier and causes a jam. If the envelope is unusually low, the top edge of the envelope may not be attached to the justifying rollers so that the envelope is not justified. Accordingly, to accommodate a variety of envelopes preferably the apparatus includes a launcher that moves up and down the top edge of each envelope to approximately the appropriate height. The envelopes then enter the top edge justifier to justify the top edge of the envelopes. The launcher is a tray that receives each envelope and moves up or down as needed to adjust the height of the top edge of each envelope when necessary depending on the height of each envelope. After the envelopes are justified by the top edge, the envelopes are brought to the upper cutting station that cuts the upper edge of the envelopes. In this way the upper and leading edge of each envelope is cut from the two cutting stations. Optionally, the side cutting station can be configured so that both sides of each envelope are cut. Another option is to remove or deactivate the side cutters so that only the top edge of the envelopes is open.

Extraction Station The extraction station 70 operates to separate the faces of the envelopes cut by the edges and present the contents so that an operator can easily remove the documents. After the operator removes the content, a sensor sends a signal to the controller that the content has been extracted. The empty envelope is then transported to the verification station 90 and another envelope is sent to the extraction station 70. Referring now to FIG. 9, the extraction station 70 includes a pair of opposed vacuum cups 74 mounted on two arms. revolving fans 72. The suckers 74 are placed on a vacuum pump. In Fig. 9 the extractor arms 72 are shown in two alternative positions. In the first position, the extractor arms rotate away from each other. In the second position the extractor arms rotate towards each other. As shown in Fig. 1, the extraction station 70 is positioned at the front of the seating area 15 intermediate the front and rear edges of the work station. Before the envelope enters the extraction station, the extractor arms 72 rotate one away from the other. When the envelope enters the extractor, the arms 72 rotate towards each other and negative pressure is supplied to the suckers, so that the suckers are coupled to the faces of the envelope. The arms then rotate one away from the other by separating the faces of the envelope, which has been cut along the top edge, and preferably the side edge (see Fig. 9). The operator can then remove the contents of the envelope.

The document transport presses the envelope between the tension rollers and a band. Therefore, when the extractor arms 72 separate the faces of the envelope, the envelope and its contents remain pressed between the tension rollers and the band. To remove the contents, the operator must pull the contents with enough force to overcome the friction between the envelope and the contents caused by the pressing action of the extraction transport. Furthermore, this friction is maintained until the lower edge of the content is pulled along the pressure point. The extraction station 70 is preferably configured to operate in three different modes to determine if the content has been extracted from the envelope: the removal mode, the differential mode and the content activation mode. The simplest way is the removal mode. An optical sensor 75 is located adjacent the extractor arms 72 at the front of the extractor arms. When the operator removes the contents of the envelope, the content passes over the sensor 75 and the sensor detects the presence of the content. Then a signal was sent to the controller indicating that the content was removed. The controller then controls a document transport to advance the envelope of the extraction station 70 to the verifier 90. In addition, an envelope is advanced towards the extraction station. The envelope is advanced from the extraction station as some of the content of the envelope passes over the sensor 75, even if some of the contents remain in the envelope. In the differential mode, the optical sensor 76 measures the thickness of the envelope immediately after the extractor arms separate the faces of the envelope, so that the thickness of the envelope is measured before the content is extracted. The optical sensor 76 continuously detects the thickness of the envelope and its contents, and compares the thickness with the reading of the initial thickness. If the difference in thickness is greater than the predetermined limit, a signal is sent to the controller indicating that the content was removed from the envelope. The controlled one then advances the envelope towards the verifier 90 and advances the envelope towards the extractor. Preferably, the work station includes a second optical sensor similar to the first sensor 76. The second sensor verifies the thickness of the envelope in the same manner as the first sensor 76. When two sensors are employed, the measurements of the two sensors are averaged and compared against the predetermined limit to determine if the content was extracted. If the operator removes all the contents of the envelope, but the differential thickness is not greater than the predetermined limit, the envelope will not be advanced. In those cases, the operator can advance the empty envelope by pressing an override button (not shown). Press the opera button to take the empty envelope to the verifier 90 and carry an envelope to the extractor. The activation mode of the content is similar to the differential mode in which the sensor 76: continuously detects the thickness of the envelope and its contents. However, in the content activation mode the thickness detected by the sensor 76 is compared to a standard thickness based on the thickness of an envelope and a variation tolerance. If the sensor 76 detects a thickness that is less than the standard thickness, a signal is sent to the controller indicating that the contents of the envelope were removed. The envelope is then advanced to the verifier 90 and an envelope is transported to the extraction station 70. Preferably, two sensors 76 are employed, one of which verifies the thickness of the envelope as described above. When two sensors are used, the measurements of. the two sensors are averaged and the average is compared against the standard thickness.

If the operator removes the contents of the envelope, the thickness detected by the sensor is not below the standard thickness and the envelope is not advanced. In those cases, the operator can advance the empty envelope by pressing an override button. In response to pressing the button, the empty envelope is taken to the verifier 90 and an envelope is taken to the extraction station 70. The standard thickness used in the content activation mode can be determined in various ways. For example, the standard thickness can be based on the first envelope in a job. However, preferably, the standard thickness is calculated on the basis of the average thickness of the envelopes processed in a job. In other words, the standard thickness is continuously updated based on an average thickness of the empty envelopes processed in a job. The standard thickness is then calculated on the basis of the thickness of the empty envelopes and a predetermined variation tolerance.

Verification Station Verification station 90 verifies the thickness of each envelope to ensure that all contents have been removed from the envelope before the envelope has been discarded into waste container 25. Verifier 90 may use an optical sensor to verify the thickness of the envelope, similar to the sensor or optical sensors used by the extraction station 70. However, the verifier preferably checks the envelope thickness by measuring the distance between the outer surfaces of the envelope faces. To measure this distance, the tester 90 includes a rotating variable inductive transducer (RVIT). To determine a reference value of zero thickness, an empty envelope is fed to the tester, and the sensor takes a measurement corresponding to the thickness of the empty envelope. The state of the subsequent envelopes is determined on the basis of the zero thickness reference value. Alternatively, the reference value used by the verifier 90 to verify the envelopes is calculated based on the average thickness of the previous envelopes and a variation tolerance, similar to the method described above to determine a thickness standard for the envelope. extraction step in the content activation mode. However, in the present case, the calculation of the reference value differs from the calculation of the standard used in the extraction step. When the reference value is calculated by the verifier 90, if the measured thickness of an empty envelope is greater than the current reference value, the thickness measurement for the envelope is not factored into the average of operation. For example, when calculating the reference thickness for the 100th envelope in a job, if the thickness of the 90th envelope was greater than the reference value based on the sixteen previous envelopes, the thickness of the 90th envelope will not be included in the the average used to calculate a reference value for the 100th envelope. If the tester 90 measures a thickness that is greater than the reference value, then a signal is sent to the controller indicating that the envelope in the tester 90 is not empty. An indicator light (not shown) illuminates indicating to the operator that the envelope in the verifier should be removed and verified to ensure that all content was removed. A verifier sensor adjacent to the RVIT sensor detects the presence of the envelope in the verifier 90. Until the operator removes the envelope from the verifier, the transport of documents will not advance any over vacuum, regardless of whether the envelope of the extraction station 70 is empty. If the verifier 90 detects a thickness that is less than the reference value, a signal is sent to the controller indicating that the envelope in the verifier is empty. The controller then activates the transport of documents to advance the envelope out of the extractor and towards a garbage ramp that discards the envelope towards the waste container behind the verifier 90. The operation of the feeding station 30, the cutting stations lateral and upper 40, 50 and extraction station 70 are similar to the operation of the apparatus described in US Patent No. 6,230,471, which is owned by Opex Corporation, who is also the beneficiary of the present application. U.S. Patent No. 6,230,471 is hereby incorporated by reference. In addition, alternative feeding and cutting stations could be incorporated into the apparatus of the present invention. For example, pending U.S. Patent Application No. 10 / 348,358 describes an apparatus for processing mail having an alternative feeding station and an edge cutting area that could be incorporated in the present apparatus in place of the Feeding and edge cutting stations described above. Application No. 10 / 348,358 is hereby incorporated by reference. The following description discusses the processing and formation of images of documents that have been extracted from open envelopes in the manner discussed above. Nevertheless, in Certain applications, the device can be operated to process documents and use the extraction characteristics of the device. For example, the apparatus can be used to process a batch of documents that have been previously extracted, such as documents that are ejected by high-speed automated processing devices. For these documents it is advantageous to use the feeding and scanning characteristics as discussed below. Similarly, a batch of pre-cut mail can be processed, so the operator manually opens the cut envelopes and then processes the documents as discussed later. Accordingly, unless noted otherwise, the following discussion of the document imaging process is applicable to a variety of applications in which a batch of documents needs to be subjected to image formation, no matter how the documents were obtained (ie the documents are provided in a stack as opposed to the documents to be removed from envelopes.) The features of the present invention are not limited to applications in which the opening and removal characteristics of the envelopes are used. envelopes of the devices.

Gravity Transporter Referring to Figures 1-8 and 15, the gravity conveyor 100 is configured to receive documents extracted from the envelopes. The conveyor 100 is positioned along the front edge of the work station 10, so that the conveyor is operable to convey documents adjacent to and parallel to the front edge of the work station. In addition, the transporter preferably transports the documents that fell to the left side of the workstation. More specifically, preferably the gravity conveyor is located so that it has a dropping zone centered on the seating area and aligned with the extraction station. Preferably, the conveyor 100 is configured to easily receive documents that the operator manually removes from an envelope in the extractor. More specifically, the conveyor is configured to receive documents that simply fall on the conveyor and to transport the documents that fell to the imaging station 130.

In this way, the operator can easily extract and, if necessary, unfold the documents and simply drop a document or package of documents on the conveyor with minimal preprocessing of the documents to prepare the documents for exploration. Although the operator preferably drops the documents on the drop zone of the conveyor, the drop zone is a substantial area that is much larger than that of the documents. Consequently, the operator does not need to be precise with the location and orientation so that the documents are dropped on the conveyor. However, preferably the operator drops the documents so that the documents are turned up on the conveyor. However, since preferably both sides of the documents are scanned at the imaging station, if the documents are turned down, the apparatus can analyze the documents to determine which face is the front face. Up to this point, preferably, the conveyor 100 is a roller bed conveyor. The roller bed provides a generally horizontal surface onto which documents can be dropped. The roller bed comprises a plurality of horizontally placed cylindrical rollers 102 driven by a band 103 which engage in the lower part of the rollers, which in turn are driven by a motor controlled by the system controller. The rollers 102 can be parallel to each other and perpendicular to the direction of travel, so that the documents move directly along the roller bed 100. However, preferably, the rollers are skewed so that the rollers move the documents in front of the bed of the roller and laterally to a justification rail 105. In this way, the slanted rollers 102 direct the documents against the rail 105 for edge alignment or justification of an edge of the documents against the rail. Each of the rollers 102 comprises a plurality of grooves sized to receive O-rings. The O-rings have a coefficient of friction higher than the surface of the rollers, to provide an increased friction area between the roller bed and the rollers. documents, thereby improving the justification of the documents. As mentioned above, the document rests on the rollers. Therefore, when the rollers 102 rotate, the rollers move the documents forward.

If a single document or package of folded documents is placed on the conveyor, wrinkles in the documents may tend to cause portions of the documents to project up or down, which could cause feeding problems for the documents when they enter. the image formation station. In consecuense, preferably the justification rail 105 is configured as shown in Figure 4. Specifically, the rail 105 is preferably formed of a low friction material, such as smooth aluminum or low friction plastic, and is formed in a generally U-shaped. In this way, the rail forms an upper lip 106 and a lower lip 107. The upper lip 106 prevents documents from sliding up and over the rail when the documents move forward and towards the rail. The lower lip prevents documents from being forced down between the rail and the edge of the rollers when the documents are moved forward and towards the rail. Although, the gravity conveyor 100 has been described as a roller bed conveyor, alternative types of conveyors can be used as the gravity conveyor. For example, the gravity conveyor may comprise a horizontal conveyor belt. If a conveyor belt is used, preferably the web is biased towards the rail 105 so that the web justifies the documents against the rail. Alternatively, instead of a single conveyor belt, the gravity conveyor may comprise a plurality of smaller conveyor belts onto which the documents may be dropped. Another alternative gravity conveyor comprises a plurality of wheels projecting upwards from a base. In that configuration, the conveyor includes a sufficient number of transport wheels that are separated from each other, so that the upper parts of the wheel form a supporting surface to support the documents. In this way, the documents are dropped on the wheels and the wheels are arranged so that when the wheels turn, the wheels direct the documents forward. Additionally, preferably the wheels are angled towards the rail 105, so that the wheels justify the documents against the rail when the wheels direct the documents forward. Although the conveyor 100 is referred to as a 'horizontal conveyor, preferably the gravity conveyor is angled downward, so that gravity pushes the documents towards the guide rail 105. This can be seen mixed in Figure 4. Preferably the conveyor 100 is angled by approximately five degrees, however, the angle may be larger, and indeed, the angle of the conveyor may be increased to a point at which the conveyor is vertical rather than horizontal. In addition, preferably the image forming station and the sorting station are angled downwards, similarly to the gravity conveyor.

Image Input Feeder Referring to Figures 1-8, 15 and 16, of the gravity conveyor 100, the documents enter an image input feeder 110 at the end of the conveyor. When the documents are conveyed to the image former, the documents are generally placed horizontally, rigid on top of the gravity conveyor 100 are aligned by the edge against the justification rail 105. On the conveyor, the documents are only held down by gravity, and the upper area of the justification lane depending on how the document wrinkles. Since documents can be wrinkled, gravity is generally not sufficient to flatten documents, so that documents can be fed to the imaging station. Therefore, the image input feeder 110 preferably provides at least two functions: (1) positively coupling the documents to pull or push the documents down; and (2) control the feeding of the documents to the imaging station. As shown in Figures 6-7, preferably, the input feeder comprises one or more parallel feed strips placed substantially horizontal and parallel to the base plate of the feeder. Although the input feeder 110 may comprise a single broad band, preferably the input feeder comprises a plurality of parallel bands that are progressively longer, so that the documents are progressively pressed over a wider area when the documents enter the feeder, as discussed later. Preferably, the feeder belt comprises a first band 112 that operates as the front feed belt. The front band may be longer than the other bands in the feeder, so that the front band projects outward toward the gravity conveyor more than the remaining feeder bands. In addition, preferably the front band is positioned adjacent the rail 105, so that the front band is closer to the rail. A tension roll 113 below the base plate is biased upward towards the coupling with the end of the front feeder band 112 to form an inlet contact line. This contact line is the initial contact line that serves to control the documents when they enter the feeder 110. Since the contact line is close to the lane 105, the initial control point is adjacent to the lane, which is the border of the lane. justification of the documents. The front band 112 pulls the documents forward, so that the documents are attached to the secondary power strips 116. At this point, the documents are coupled to the front band and the sub-bands, so that the coupling area with the documents, (and therefore the control over the documents) is wider. Furthermore, preferably, the input feeder further comprises an arm 114 projecting forward and upwardly from the leading edge of the forward feed belt 112, parallel to the front band. The advance arm 114 comprises a rotating band which forms an acute angle in relation to the gravity conveyor 100. The arm operates to push down any document that may have portions adhered upwards as described below. Referring to Figures 2, 6-7, folded documents are illustrated on the gravity conveyor. The document is exemplary of a document that was folded into three sections (that is, into three folds) and placed in an envelope. When the document was extracted from the envelope and unfolded, the document retained the wrinkles that caused the leading and trailing edges of the document to be folded up (or downwards if the documents were released above in relation to Figures 2)., 6-7). Similarly, the document could be a three-fold document in which the fold lines are parallel to the flow direction of the conveyor, instead of perpendicular as shown in Figures 2, 6-7. A document having a leading edge that is wrinkled so that the leading edge is bent upward will tend to bend upwardly when the feed contact line is fed between the forward feed belt 112 and the tension roller 113. If the document is folded up, the image will not be formed properly. The operator could drop the document on top, so that the leading edge does not bend upward, which would limit the potential problems associated with document folding. However, it is desirable to configure the input feeder so that it can accommodate the documents regardless of the orientation of the documents and the folds in the documents. Accordingly, preferably the inlet feeder 110 includes the angled feeder 114, which rotates clockwise from the perspective of Figure 6. The angled feeder arm, or removable arm, is a feed belt that operates to direct the documents down if the documents are wrinkled and stick up. The position of the angled feeder arm tends to push the folded portion of the document down to flatten the document without bending the document up. More specifically, the removable feeder projects forward and upwardly from the front horizontal feed belt 112 so that the leading edge of the removable web is positioned higher than the front web and angled downward where the removable arm 114 it is located approximately at the same distance from the gravity conveyor 100 as the front feed belt 112. In this way, the leading edge of the detachable feed arm 114 extends upwards over the height of most of the folded documents, so that the feeder flattens the documents without bending the documents up. At the same time, the gravity conveyor and the removable feed arm 114 push the documents towards the front feed belt. Once a document is pressed between the front arm 112 and the tension roller 113, the feeder directs the document forward. The shorter feed bands 114 are then coupled to the document, so that the document is pressed (i.e., controlled) by a larger portion of the document. Additionally, the portion of the document pressed by the feed bands 112, 114 is held flat. In other words, when the document enters the feeder, the document is progressively flattened. The contact line formed under the shorter feed bands 114 are formed between the feed belts and a retarder 120. The retarder is formed with a high friction material so that the friction force between the document and the retarder is greater than the frictional force between two documents. The retarder can be formed in any of a number of configurations. However, the present case, preferably the retarder comprises a pad having a plurality of spaced ridges 121 which are positioned between the bands forming the feeder as shown in Figure 5. In this way, the documents are attached to the retarder between the bands. The input feeder 110 operates to feed the documents to the imaging station through the feed slot that is formed between the base plate 108 and an input guide 132. Preferably, the input guide 132 is spaced from the base plate approximately 1/4"(6.35 mm) Since the documents are pressed by the feeder, the portion of the document on or adjacent to the feeder will easily fit into the feed slot., the end of the document away from the feeder can still adhere to wrinkles or folds. For this reason the leading edge of the guide is angled backwards (ie, in the direction of the paper path, as shown in Figure 7). Further, preferably the leading edge of the entry guide is curved upward to provide a guide for forcing the leading edge of the document downwardly if the document has a portion adhered upwardly of the folds or wrinkles. The above discussion described the feeding of a single piece of the conveyor to the feed slot downstream of the feeder 110. However, the feeder can also operate to receive and separate document packages, so that the documents are fed in series to the image formation station. To accommodate document packages, preferably the front feed belt 112 is separated from the base plate approximately 3/16"(4.57 mm). When the package documents are fed, the lowermost document in the package may tend to curl in the package. space between the feed belt 112 and the base plate Specifically, the document pack is pressed between the retarder 120 and the feeder 110, with the trailing edge of the documents pressed between the front band 112 and the tension roller 113. When the second to the last document is fed through the feeder, the trailing edge of the second to the last document is advanced out of the contact line of the tension roller 113 and the front band 112 so that the last document in the stack is coupled by both the feed belt and the tension roller so that the contact line directs - the trailing edge of the last doc forward. However, a portion of the second to the last document is still pressed between the feeder and the retarder, so that the last document is held back by the retarder while the back end of the last document is directed forward, which can cause the last document is rolled between the base plate and the bottom of the rear web 112. Accordingly, preferably, a flexible, thin, elongated plastic finger 117 extends across the front contact line to support the document on the line contact. Preferably the finger 117 is long enough to extend to the retarder 120. By supporting the document, the finger prevents the document from curling from the feed force forward when the retarder is holding the part behind. When a package of two or more documents is received on the conveyor, the feeder is designed to control the documents, so that they progress to the imaging station one at a time. If more than one document is fed at a time then one or more of the documents will not be subjected to image formation. Feeding more than one document at a time is commonly referred to as double feeding. Double feeding occurs when two documents enter the image former together because the frictional force between the two documents tends to pull the second document together with the first document when the first document is fed to the image former. The retarder 120 is provided to hold back the second document (as well as any documents that may be in the packet) so that only the first document in the packet is fed to the image former. Although it is typically desirable to use the retarder so that the document packages can be separated from each other, in certain cases it may be desirable to feed the documents without the retarder. Accordingly, the device includes a rotary delay control button 119 that raises or lowers the retarder. In a first position (which is shown in Fig. 5), the retarder is raised so that the inlet contact line between the feeder and the retarder is formed. By rotating the control button 119, the retarder is lowered so that the upper surface of the retarder remains below the base surface of the imaging section, so that the retarder does not project upwards in contact with the documents. Instead, the contact line or inlet pressure is formed between the feeder and the feeder base plate.

Image Formation Station From the input feeder, the documents enter in series to a line in contact or pressure formed between a pair of crushing rollers 140. Although the input feeder keeps the documents below, it does not flatten the documents; generally it only keeps one edge of the document flat against the base plate of the feeder. In contrast, the crusher tries to flatten the crumpled documents. The crushing rollers 140 are elongated cylindrical aluminum rollers 142 having a smooth surface. A plurality of elastomeric clamping rings 144 are formed around the circumference of the roller 142, and spaced from each other. Preferably, a first clamping ring is placed at the end of the roller 144 at or near the input feeder 110, and a second clamping ring is placed on the roller a couple of inches further. More specifically, preferably the second clamping ring is spaced apart less than the width of the feeder 110. In addition, preferably a third clamping ring is positioned adjacent the opposite end of the roller. The first and second clamping rings 144 provide contact or pressure lines directing the paper from the feeder inlet to the image former 150. The third clamping rings are positioned so that they are not in the paper path (i.e. the third clamping rings do not engage the documents, instead the third clamping rings provide spacing to keep the rollers parallel with a constant spacing Preferably, the first two clamping rings 144 on the rollers 142 are positioned so that both rolls are coupled to a single fold for three fold documents with the fold lines placed parallel to the paper path (ie the fold line would be perpendicular to the fold line of document 8 in Figure 2). In this way, the clamping rings are coupled to the third justified edge of the document of three bends, while the rest of the The document can slide across the width of the crushing roller since the remaining width of the crushing roller in the paper path is made of aluminum. In this way, the crushing roller flattens the documents without curling the documents. When the documents are processed through the imaging station 130 and through the imaging transport 170 and the sorting station 200, the documents will be appropriately separated to avoid jams, to ensure an appropriately scanned image of each document, and to allow proper restoration of documents during subsequent processing. If the operator separates each document in a transaction and drops the documents in series on the gravity conveyor, the operator has provided the space between the documents. However, if the operator drops a package of documents on the conveyor, the documents need to be separated so that appropriate spaces are provided. The spaces can be provided in one of at least two ways. First, the imaging input feeder 110 can control the feeding of the documents to provide spaces. For example, after the feeder 110 feeds the first document in the package, the feeder may stop or slow down for a predetermined period of time before feeding the next document in the package. This delay will create a space between the documents. Alternatively, the spaces can be provided by operating the feeder 110 at a lower speed of the transport path through the image forming station. This difference between speed will create a space between the documents. As can be seen in Figure 2, a plurality of biased sensors 135 are placed in the feeder between the input feeder 110 and the crushed roller 140. The biased sensors detect the leading edge of each document to determine whether the document is biased relative to the crushing roller (which is parallel to the imager). The biased sensors 135 are preferably a plurality of optical sensors aligned with one another along a line that is parallel to the image former 150. In addition, preferably, the first sensor is positioned adjacent the front edge of the work station, and preferably the second biased sensor is separated from the first biased sensor by a short distance, so that the biased sensors can detect the bias of relatively narrow documents. In addition, a third sensor is aligned with the first two sensors and further apart (ie preferably at least 7.5-15 centimeters (3-6 inches)) away from the second sensor. The central controller receives signals from the segmented sensors. If the documents are slightly skewed, the documents can continue through additional processing. However, if the document is more skewed, the central controller electronically marks the document to be rejected, so that the document is transported to a reject tray without forming that image. Alternatively, if the document is extremely skewed, it may not pass through the image forming station and / or the sorting station without binding. Accordingly, the central controller stops the operation of the apparatus and provides an indicator (such as a warning light or notification on the control screen) that the document needs to be removed from the imaging station. After passing the biased sensors 135 and the squeezing roller 140, the document passes through a thickness detector 147 which measures the document at a plurality of points throughout the length of the document. The thickness detector can be any of a variety of sensors, such as an LVDT sensor. However, preferably the thickness sensor is a Hall effect sensor. The Hall effect sensor. 147 includes a sensor board positioned adjacent to a magnet that is mounted on a brace that biases the magnet towards the sensor. The magnetic field created by the magnet is measured by the sensor board as a function of the distance between the magnet and the sensor. The magnet and the sensor are linked to a pair of rollers between which the documents are squeezed when the document enters the thickness detector 147. When an envelope enters the thickness detector 147, the arms are forced to separate, thereby separating the magnet and the sensor board accordingly, changing the intensity of the magnetic field. The thickness data is communicated with the system computer and stored in a data file associated with the document. From the thickness detector 147, the document enters the image former 150. Preferably the image former comprises a pair of scanning devices for scanning both sides of the document. Specifically, preferably the imager 150 includes a lower plate in which lower scanning devices 150 are located, and an upper plate in which the upper scanning device is located. The lower scanning device 150 scans the lower face of the document, and the upper scanning device scans the upper face of the document. As shown in Figure 8 preferably the top plate can rotate upwardly away from the bottom plate to allow access to the image forming station 130 in the event of a jam at the image forming station. Although the scanning devices may be black and white or preferably gray scale, the scanning devices 150 are color scanning devices. More specifically, preferably the scanning devices 150 are contact image sensor modules (CIS) formed of arrays of photodiodes that operate as scanning elements and LED light sources. The scanning device 150 scans the documents in three light colors, preferably red, green and blue. The scanning device scans the documents and acquires data that represents the intensity of the light of each of the three colors in discrete points of each document. For each color, the light intensity of each point, or pixel, represented by a number that fluctuates from zero to 255. The intensity of light for each pixel is communicated to the computer and stored in a data file. In addition, instead of being scanned in color, the documents can be scanned in gray scale. When the document passes between the scanning devices, the scanning devices scan the faces of the document to obtain image data that represents a color image of the faces of the document. The image is communicated with the system computer and the image data is stored in a data file associated with the document. From the scanning device, the document is transported to a MICR detector which attempts to read any MICR marks on the document. Specifically, the MICR marks are printed on a magnetizable tape. The MICR detector includes a magnet that exposes the document to a magnetic field. The MICR detector also includes a MICR reader that scans the document for magnetic fluctuations indicative of MICR characters. If the device detects the presence of a MICR line, the MICR detector attempts to read the MICR line. The data representing the MICR information is then communicated with the system computer, which stores the MICR data in a data file associated with the document.

Acquisition and Processing of Image Data When the image input sensor detects the presence of a document, the image input sensor sends a signal to the system controller indicating the presence of a document. The system controller then sends a signal to the scanning devices with respect to the document and data about how the document will be processed. For example, the system controller sends a signal to the scan station indicating what type of document is being scanned and whether the default scan parameters are to be modified for the document (for example, the document will be scanned in a scenario darker) . The scanning devices scan the document to acquire image data. The system computer receives the image data from the scanning devices and then stores the data in memory. Once the image data is transferred to the system computer, the image data is processed. Although the scanning device preferably scans the color documents, preferably the image data is processed to provide a grayscale image that can also be binarized, as discussed further below. First, preferably the gray scale data is binarized to create a black and white representation of the document image. By binarizing the data, the data of each pixel is converted from an eight-bit gray scale representation to a black and white representation of the bit, which significantly reduces the space required to store the image data. Furthermore, the binarization of the image data operates to highlight the textual portions of the image, which is advantageous for further processing of the image data. To binarize an image, the data in the gray scale of each pixel of the image are compared with the threshold. If the number of the gray scale for a pixel is greater than the threshold, the gray scale is converted to white. Conversely, if the number of the gray scale is less than the threshold, the gray scale is converted to black. To consider variations between the different documents that are being processed, an adaptive threshold can be used, so that the binarization threshold is variable for each document. An adaptive threshold uses a different threshold for each particular document based on a sampling of the image data for the particular document. Using a threshold that is specific to each particular document, the threshold for a document having a generally dark image will typically be different from the threshold for a document having a generally clear image. In this way, the resulting binarized image will reflect more closely the image of the document than if a fixed threshold is used for all documents. In addition to the binarization of the image data, the image data is filtered to reduce noise, which eliminates the unnecessary background and parasitic markings of the document image. Noise filtration can be carried out in series or in parallel with binarization. To filter the data, the image-forming computer examines the image data to detect any black pixels that are surrounded by white pixels. If a black pixel is surrounded by white pixels, the image-forming computer converts the pixel from black to white. Similarly, if a pair of attached white pixels are surrounded by white pixels, both pixels are converted from black to white. After the image data is binarized and filtered, the image-forming computer tries to read information from the document. For example, the image-forming computer may attempt to read an OCR line if the document is an invoice or its MICR line if the document is a check. OCR line data is useful for final processing because the OCR line for documents, such as invoices, includes information about the customer's account and the amount of the invoice. During further processing, it is useful to know the customer's account number. In addition, it is useful to have other information about the document to process the document. For example, it is useful to know the amount of the invoice. On the basis of the data received from the system controller, the image-forming computer knows what type of document is being scanned. Specifically, as discussed further below, the type of document can be determined automatically by each document based on the information that is acquired for the document during processing (for example, the length and thickness of the document). In addition, as discussed below, the operator can manually identify the type of document by selecting the appropriate document type when the document is dropped on the conveyor. If the image represents a pay stub, the image-forming computer processes the image data for the document to determine the OCR line of the document, which typically appears at the bottom of the pay stubs. The OCR line is a series of characters printed in a predefined uniform type of predefined size. Commonly, the type is a type referred to as OCRA, however, other types of types can be read, such as OCR B, E13B and others. In addition, the image forming computer may operate to process the image data to read the MICR line of checks, so that the MICR line is read both optically and magnetically by the MICR character reader. As described above, the MICR character reader magnetically reads the MICR line over checks. However, the character reader in MICR may be unable to read one or more characters in a MICR line due to imperfections in the magnetic characteristics of the ink of the MICR line. These magnetic imperfections, however, may not affect the ability of the image-forming computer to read the MICR line of the optical image data, so that a character that may not be read magnetically may be optically readable. Therefore, if the MICR character reader is unable to read a character in a MICR line, the optically obtained data is used to supplement the data obtained from the MICR character reader in an attempt to complete the data of the MICR line. Alternatively, it may be desirable to use the image-forming computer to verify the results of the MICR character reader. Verifying the results, the possibility that the checks are processed with inappropriate MICR data is reduced. For this process, the data of optically obtained MICR lines can be compared with the data of the MICR line of the MICR character reader. If there is any difference between the MICR line read optically and the results of the MICR character reader, the image forming computer indicates that the MICR line was not determined. The system controller then marks the document as if it had an undetermined MICR line and the document together with the remaining documents in the same transaction are directed to a reject bin or sorted by the stacker accordingly. After the data from the OCR or MICR line for a document is extracted from the image data, the image data for the document is compressed using a compression algorithm, such as a Group 4 fax compression commonly referred to as "CCITT Group 4". The compressed image is combined with the other data for the document. For example, the document type can be combined with the data representing the MICR line or the OCR line, together with the data from the system controller to form a data record for the document. The system controller data includes information about the envelope from which the particular document was extracted, such as a change of address indication, the presence of the postal network barcode, and the presence of a brand that indicates a response from the customer. The data from the system controller can also include an indication of whether the MICR line and the OCR line were completely determined during the formation of the image. Accordingly, the data record for a document includes the image data (color image data, gray scale image data and / or binarized data), the MICR or OCR line, an indication of whether the OCR line o MICR is complete, a barcode, and miscellaneous information obtained during the processing of the document, such as the length and / or thickness profile of the document, as well as customer response data in the form of a change of address, or a mark of check in a response mark. Alternatively, it is often desirable to store color or gray scale images of the documents in the data records instead of binarized and filtered images. To store the color or grayscale images, a copy of the color image data or the gray scale of the document is binarized and filtered as described above, so that the image computer can determine the MICR line. OCR, and other information of the image data. The color image data is then compressed using data compression, such as JPEG, and the image data is combined with other information to create a data record as described above for the binarized and filtered images. The binarized and filtered images can then be discarded from the RAM of the image-forming computer. In addition to the functions described above, the apparatus is operable to modify the image in any of several ways. For example, the image may be unselected, and the black border that frequently occurs around the scanned image may be removed during image processing. The image can be rotated to correct the orientation of the image. More specifically, the document may be fed to the image forming station in an inappropriate orientation, so that the orientation of the image does not reflect the correct orientation of the image. For example, a document can be printed in what is commonly referred to as a panoramic orientation, in which each printing line extends across the long side of the page (ie the length), rather than through the side short of the page (ie the width). If the short side of the page is fed to the image forming station, the image can be presented in a normal vertical portrait orientation, commonly referred to as a portrait orientation. In that case, the impression on the image would appear rotated (for example instead of being read from left to right, the impression would be seen from top to bottom). By rotating the image 90 ° the image would appear correctly. If the image needs to be rotated it can be determined in one of several ways. For example, the operator can intervene and feed information that indicates that the image needs to be rotated, and how much. Alternatively, the image forming station can automatically determine whether the image should be rotated. In an application, the image forming station can determine the correct orientation of the image on the base and the type of document. As discussed below, the type of document can be determined automatically on the basis of various document characteristics (eg length, width, thickness profile, position of magnetic characters). Based on the type of document, the image forming station can automatically rotate the image when necessary. For example, checks are printed in a panoramic orientation. Determining the length, width and location of the MICR characters on the document, the device can automatically determine that a document is a check. Since the document is a check, the image forming station can rotate the image data when necessary, so that the image data appears in a panoramic orientation. As described above, the type of document is determined automatically based on the characteristics of the document. However, as described below, the type of document can be determined manually or semi-automatically. Regardless of how the type of document was determined, the image data can be automatically rotated to correspond to the appropriate orientation for the type of document. Moreover, instead of being automatic, the rotation of the image can be semi-automatic based on the type of document. For example, based on the type of document, the system can determine that a document should be rotated, but the operator can be suggested to confirm whether the document should be rotated. Depending on the response (or possibly the absence of response) the image data may or may not be rotated. As described above, the image is rotated based on the type of document. In certain cases, it may be desirable to prevent the image from being rotated based on the type of document. Specifically, the system can be configured so that the document images of a certain type are rotated, when necessary, on the basis of the configuration assumed by the type of document. If the operator recognizes that the image of a document should be oriented in a specific way, the operator can force the system to explore the system in a certain configuration. For example, if a document of the type "A" is assumed to be in panoramic orientation, the system will automatically rotate the image to be viewed in panoramic orientation. However, if the operator recognizes a document of type A "is really in portrait orientation, the operator can press a button to perform the rotation.In other words, the operator can manually force the system to recognize the document in a certain orientation , and rotate the document or avoid rotation of the document as appropriate.Image data can also be electronically stamped to identify audit information, such as lot number, part number, document type and / or date When the document was processed, to electronically mark the images of the document, the image data is modified so that the audit information is "printed" on the image file for the document. They will show audit information as if the information were printed on the document before the image of the document was formed. Audit can be stamped on the image of the document in one of several ways. For example, information can be stamped on each document in a predefined area. Alternatively, a border may be added around the image data and the information may be electronically stamped on the edge to ensure that the stamped information does not overlap and thus alter the image data. In yet another alternative, the audit information is stamped on the document so that each pixel constituting the audit information is exactly opposite to the actual image data for the pixels where the information is being stamped. For example, if the image is a binary image, and the information is to be printed in an area that has black and white pixels, each audit pixel that is to be stamped on an area that is black should be stamped as white, while the pixels that are white will be printed as black. Although the previous discussion of the printing of the audit information describes situations in which the image is binarized, the apparatus also operates to stamp portions of the document image that are color or grayscale. Specifically, the electronic stamping is carried out so that each pixel comprising the electronic stamping is the number in gray scale or color that corresponds better to the opposite light intensity of the pixels that constitute the area to be overstaged. Preferably, the documents are classified and maintained in groups referred to as lots, which are identified by a unique lot number. The image data for a batch of documents are organized and maintained in batch data record files referred to as batch files. The batch files are organized so that the organization of the images in a particular batch file corresponds directly to the organization of the documents in the batch. A complete batch file includes a batch header and data records for each of the documents in the corresponding batch. The batch header includes information that is common to all documents in the batch, such as the batch number, the dates the documents were processed, and the number of documents in the batch. Once all the records have been appended to a batch file, the batch file can be exported to an image file server in the form of a personal computer, PC, via an ethernet connection. Difficulties may arise when document images are formed. For example, the MICR 220 module and the image forming computer may not be able to read the MICR line of a check or the OCR line of a pay stub. If that problem arises, the image, together with the data that was determined from the image can be exported to the respective batch file for the document and the document can be classified with other documents in the batch. In this way, the document and the image data for the document are available for further processing, at which time the missing information can be read manually and fed to the data record for the document. Alternatively, the screen may suggest to the operator to examine the scanned image displayed on the screen and enter the missing information. Alternatively, if some information, such as the MICR line or the OCR line is not completely read from a document, the document along with other documents in the same transaction can be directed to a rejection bin. If the documents are directed to a reject tray, then the image computer discards the image data for the documents in the transaction. Although the description has been focused on the processing of checks and invoices, the apparatus is not limited to the formation of images of checks and documents. More specifically, the apparatus can operate to explore and process a variety of documents that can include a variety of information that can be explored and analyzed. For example, the image forming station is configured to accommodate at least as large documents as 8 x 14. In addition, the image data exported to the data file may include more data than just the image file for the entire document. For example, portions of the documents can also be included in the data file. For example, for a check, an amplified view of the cut chart can be included in a check in the data file. In addition, the portion of an image in which a bar code is located can be separated as a file and included in the data file. Portions of scanned images that are appended to the data file are not limited to a single portion. Preferably, up to 10 different image data sets may be included in the image data file. In this way, a variety of portions can be stored in the data file along with the image data for the entire image, if desired. Additionally, a variety of information can be stored in the data file with the image data, such as the length and / or height of the document, the thickness profile, the MICR line, etc. In addition, in certain cases it may be desirable to feed a batch ticket before the batch of documents is fed. The batch ticket includes a variety of information specific to the batch of the document. For example, the batch ticket can identify the batch number and the date the image was processed.

Image Formation Transport As shown in Figure 8, the image formation transport 170 extends between the image forming station 130 and the sorting station 200. Preferably, the image forming transport is formed of two halves, and the upper half 172 rotates away from the lower half 174 to provide access to the transport path to remove any jammed paper in a transport, or perform service on the inner element. As shown in Figure 8, the path of the document between the image forming station 130 and the sorting station 200 is preferably not a straight horizontal path. Instead, preferably, the imaging transport 170 rotates upward and curves downward towards the sitting area 15. Specifically, the imaging transport 170 transports the document away from the scanning devices 150 at a time. horizontal generally parallel direction to the bottom plate of the lower scanner plate. The transport 170 is then curved upwards, thereby rotating the document upwards along a generally vertical path. The transport 170 is then bent backwards, generally horizontally, towards the sitting area. In other words, transport essentially provides a U-turn, which rotates from the image-forming station backward, towards the operator. The rear leg of the U-turn also extends upwards at an angle relative to the base plate of the scanning device. At the end of the rear leg of the U-turn, the transport 170 is again curved upwards, thereby rotating the document upwards along a generally vertical path towards the sorting station 200. Between the sorting station 130 and sorting station 200, preferably an optional dewrinkler station and a printer are placed along the transport path. Referring to Figure 10, a schematic view of a wringer 180 is illustrated. As discussed above, it is desirable to remove wrinkles from folded documents, so that documents can be stacked more efficiently on the exit trays 205. Dewrinkling station 180 is a guide having a sharp edge through which documents pass over when documents revolve around the U-turn. More specifically, the transport path includes a pair of guides that direct documents upward. The guide on the outside is a generally uniformly curved guide 182. The guide 184 on the inside of the turn includes an edge. For example, by comparing the outer guide with the inner guide, the outer guide is a uniformly curved metal rail, while the lower track includes two generally flat, correct legs formed at an angle to each other. The intersection of the two straight rails forms an edge that operates as a cutting edge. Preferably the documents are pulled strongly against the cutting edge when the documents are conveyed through the cutting edge. To do this, preferably the documents are tensioned by increasing the transport speed on the downstream side of the cutting edge. Specifically, a line of contact or pressure is formed downstream of the cutting edge and the conveying speed of the line of contact or forward pressure is greater than the transport speed upstream of the scraper. In this way, when the document enters the line of contact or pressure, the higher transport speed pulls on the document, which stretches the document when it is pulled over the cutting edge. Pulling documents over the cutting edge removes or reduces wrinkles in the documents.

Although a wringing device has been described as a guide having a cutting edge, various other types of wringing elements can be employed. For example, the cutting edge can be mounted on a solenoid adjacent to the document path through transport 170. The solenoid can be controlled so that it moves the sharp edge towards the paper path, so that the paper is pulled over the cutting edge when the solenoid is actuated. The solenoid can then be de-energized to retract the cutting edge out of the paper path over selected documents or over selected portions of the document (ie if only the second half of a document needs to be dewritten, the solenoid could be synchronized to be driven to move the cutting edge towards the path when the second half of the document passes through there). Another alternative scrubber is a heater. The heater includes a pair of rollers that form a line of contact or pressure so the documents pass through it. One of the rollers is heated to a high enough temperature to reduce wrinkles in the documents. Additionally, the apparatus may include a second de-wringer to de-rip the opposite side of the document. Specifically, the second scrubber can be configured in a manner similar to the first scrubber, except that the second scrubber can be formed by the outer guide, instead of the internal guide as described above. Although the debris has been described as being located between the MICR reader and the printer, it may be desirable to locate the debris at a different location. For example, the scrubber may be located before any of the scanning devices 150 or before the MICR reader. Moreover, the scrubber can be located downstream of the printer if desired. Furthermore, preferably, a printer is positioned along the transport 170, so that the printer can print marks on the documents that are transported to the sorting station 200. After the documents are subjected to imaging, the transport transports the documents to a printer module 190 which is controlled by the image-forming computer. The printer module 190 includes at least one ink jet printer. The 190 printers placed behind cover the imaging transport. More specifically, a first printer is preferably placed behind a plate in the upper portion 172 and preferably the second printer is placed behind a plate in the lower portion 174. In response to computer signals, the printer module 190 Print audit data on each document. The information printed on a document includes particular data for the document, such as the type of document for each document, the lot number for the document, the document number, the transaction number for the transaction of which the document is a member, and the date on which the document was processed. The audit information can be used to later locate a particular document within a stack of documents. Generally, it is desirable to only print information on the back side of each document. However, the printer module may also include a second inkjet printer to print additional information on the front face of the documents. If a second printer is used, the image-forming computer controls the printers to selectively print on either or both sides of a document.

Sorting Station The sorting station 200 is located at the end of the imaging conveyance 170, and the sorting station includes a plurality of gates 204 which operate to sort the documents on one of a plurality of trays 205. Preferably, the documents are processed so that the documents are downloaded in trays 205 turned down, so that the stack of documents in the output trays are in the proper order, so that the stack does not need to be reordered after the documents are processed. As shown in Figure 8, the sorting station includes a plurality of gates that are operable to direct the documents to the appropriate tray 205. The sorting can be based on a number of criteria. For example, documents can be classified according to certain information of the image data. One of those examples is related to shapes that have a box change of. address (ie the customer ticks the box if the consumer changes address). The image data can be analyzed to determine if a change of direction is indicated. If so, the classifier 200 can classify those documents to a particular tray 205 by keeping those documents together. Similarly, documents can be classified on the basis of other criteria, such as the thickness profile of the document, which can indicate what type of document the document is, or the documents can be classified based on the length of the documents. documents, which can be measured by one of the sensors at the entrance "of the image-forming station.In addition, the documents can be classified on the basis of a combination of information that is determined for a document (ie the thickness profile, length, MICR reading, OCR reading, bar code reading or an identification mark). In addition, documents that are marked electronically so as not to be classified or rejected can be classified to a particular tray. In addition to identifying parts automatically, the operator can manually identify the type of document before the document is scanned. The piece is then classified based on the type of document indicated. For example, there are five trays in the sorting station, and there may be five or more different types of documents defined in a particular batch of documents. Each type of document is classified to a different one of the trays 205, unless there are more types of trays documents, step in which, more than one type of document will be classified to some or all of the trays. Each type of document is represented by a different button on the touch-sensitive screen 12. When the operator places a document on the conveyor 100, the operator presses the button that identifies the type of document, and the document is classified accordingly. Alternatively, when the operator presses the button to identify the type of document, the apparatus can assume that all subsequent documents are of the same type of document until the operator presses a different button. In this way, the operator does not need to manually identify each type of document if a batch of documents contains groups of the same type of documents. Moreover, the manual indication described above can be combined with the automatic determination described above. Specifically, the apparatus can automatically identify the type of document unless the operator manually instructs the system to identify the type of document by a document when it is placed on the conveyor. In the previous discussion, the documents were classified according to several classification criteria based on information about each document. However, in some applications it may be undesirable to simply sort the documents in the order in which the documents are processed. In that application the documents are classified to the first tray 205 until the first tray is full. The documents are then directed to a second tray until it is full, and / or until all trays are full or all documents are processed. This process of changing from one tray to the next when the tray is full is commonly referred to as cascade, and the trays operate in cascade in various sorting applications. As discussed above, documents may have wrinkles that often affect the processing of documents. In sorting station 200, wrinkles can affect the stacking of documents in the trays. Specifically, if the documents are creased along lines perpendicular to the path of the document, the documents will have to fold down on their own when they are downloaded to the output trays 205. Consequently, preferably the sorting station includes corrugating rollers for corrugating the documents so that the documents are sufficiently rigid to be properly unloaded, so that the documents are properly stacked to the trays. Preferably, the corrugating rollers are configured as follows. The rollers are placed on two parallel axes with one of the axes being diverted towards the other axis. The opposite discharge rollers are located on the outer ends of the axes. The discharge rollers provide contact lines or discharge pressure through which the documents pass through. Intermediate in the discharge rolls are three corrugation rolls which are larger than the discharge rolls, preferably at least about 25%. Two of the corrugating rollers are placed on a first axis, with the third roller being placed on the second axis at an axial location between the two corrugating rollers on the first axis. In this way, the third corrugating roller projects towards the paper path creating a deep corrugation that provides sufficient rigidity to properly unload and efficiently stack the documents.

Deviation As discussed above, documents leaving the scanning devices 152 are transported along the imaging transport, which extends upward towards the exit trays. Alternatively, the sorting trays can be diverted by directing the documents through a deflection having an opening through the image forming station and under the image forming conveyance 170. The documents can be transported horizontally from the image forming station 130 and discharged to a tray or exit area 166 on the rear side of the image forming station (see Figure 3). The deviation can be used in several applications to improve the flexibility of the device. For example, if a batch of documents simply needs to be subjected to imaging without classifying the documents after processing, then the sorting station is unnecessary. The documents can be dropped on the conveyor 100 and then fed to the imaging station 130. From the imaging station the documents are transported directly backwards and discharged through the deviation 165 in the exit area 166. In this way the scanned documents are simply stacked in a single stack in the output area. The deviation also allows thick documents to be scanned. Specifically, documents that are thicker than twice the thickness of the paper may tend to get stuck in the image-forming transport after the documents are subjected to imaging. However, documents typically do not get stuck if they are downloaded out of the deviation. Accordingly, thick documents can be processed together, so that all documents in the batch are scanned and then directed out of the deviation 165, or optionally, the thick documents can be a defined type of document that can be analyzed automatically on the basis of the output of the thickness detector 147. If the thickness detector 147 detects a document that is larger than a predetermined limit, the document is directed to the deviation 165. Alternatively, the operator can identify the document as a thick document when the document is dropped on the conveyor 100. The document is then directed to the deviation 165 after being submitted to the image formation (there is also a maximum thickness for thick documents.) If the thickness is greater than the thickness maximum, the document is too thick to be processed, even through deviation). Another use of the deviation 165 is related to the placement of envelopes. As discussed below, open envelopes can serve as an indicator of the start or end of a transaction. Typically, it is not necessary to retain the document and waste space in sorting trays 205 by filling them with envelopes. *. Accordingly, when open envelopes are used as transaction markers, preferably the envelopes are discarded through the offset 165, while the documents are scanned and then sorted into the sorting trays 205. The envelope can be identified as an envelope in one of several ways. First, the operator can manually identify the envelope as an envelope. Second and more preferably the apparatus identifies the envelope as an envelope based on the thickness profile detected by the thickness sensor 147. If the thickness profile for a document is correlated with an envelope, then the apparatus assumes that the element is an envelope and the envelope is directed to the deviation and downloaded to the discharge area. For this reason, when envelopes are processed with the documents, preferably a waste container is provided in the discharge area so that the envelopes are taken directly to the garbage. In the previous discussion, it was assumed that you do not want to keep the envelope after processing. In many cases, since the envelope is simply a transaction marker, there is no need to retain the image of the envelopes, so that the images of the envelopes are discarded. However, in certain cases it may be desirable to retain the image of the envelope even when the envelope is actually discarded. In yet another application, the offset 165 is used to accommodate envelopes having content inside. The function of this application is similar to the processing of thick documents described above. In this application the envelopes are scanned to obtain an image of the front and back sides of the envelope. The envelope is then discarded by the deviation 165, - where the envelopes are stacked. In yet another application, deviation is used to accommodate stacks of documents that are processed together, so that only the top document is scanned. The stack of envelopes is then directed through the deviation. An application of this feature is useful for processing certain types of returned mail in which the recipient's address and the post office will no longer send mail. That mail can be returned with certain information about the obscured receiver (such as the recipient's account number) and the new address printed by the post office. Presumably, one of the documents in the envelope, like the first document, will have the recipient's account number. Since the documents in the envelope are returned documents, there is probably no reason to image all the documents. However, it would be desirable to form the envelope image (showing the new address) and one of the documents (showing the customer's account information). Consequently, the operator can feed information that specifies that the documents are going to be processed without separating the documents. The delay controller 119 is turned on so that the delay is decoupled. The document package is then placed on the conveyor 100, as well as the envelope. The document package is then processed through the scanning device together as if it were a single document, so that the upper face of the upper document is scanned. The packet of documents and the envelope are then directed towards the deviation 165. Alternatively, if both directions are visible on the face of the envelope, the envelope can simply be subjected to image formation without removing any of the documents. The envelope and the enclosed content are then directed towards the deviation. The operator can select whether the envelope will be processed alone or with one of the documents enclosed. Before processing the envelope, the operator identifies how the envelope will be processed, feeds the information to the system, such as by means of a button, keyboard, mouse or other, and then drops the envelope (and the document if appropriate) on the conveyor.

Maintenance of the Integrity of the Transaction All the documents in an envelope are referred to as a transaction. In addition, in some applications the envelope can be considered part of the transaction. It is often important to ensure that documents from a first transaction are not mixed with documents from a second transaction. This is known as maintaining the integrity of the transaction. Because each envelope defines the limits for each transaction, and the documents are initially contained within envelopes, the limits for each transaction are known. However, once the documents are removed from an envelope and placed on the gravity conveyor, the documents of different transactions may be mixed. Accordingly, preferably steps are taken to maintain the integrity of the transaction after the documents are removed from the envelopes. In addition, in applications where documents have already been extracted from envelopes, operator intervention may be required to indicate the end of a transaction. One method to maintain the limits of the transaction is to manually indicate the starting point of a transaction. For example, the apparatus may include a push button or the pedal that the operator presses to indicate the start or end of a transaction. Since the operator extracts the documents from the envelope, the operator knows which documents belong to a transaction. Then the operator extracts the documents that are placed on the gravity conveyor one at a time. After the last document is placed on the conveyor, the operator indicates the end of the transaction by pressing a button (or otherwise). The following document will be processed as the first document in the next transaction. A second method to maintain transaction limits is to automatically define the transaction based on the type of document. For example, if the transaction includes two types of documents, the operator places the documents on the gravity conveyor, so that the second type of document is always the back or rear document. Therefore, when the apparatus determines that a document is of the second type of document, the apparatus assumes that the document is the last document in the transaction, and the next processed document is considered in a new transaction. When this method is used, the type of document can be determined on the basis of different characteristics that have been determined for the documents. For example, the device can determine that a document is a check if it has a certain length and a MICR line. Therefore, if the checks are the identified document, the end of each transaction is automatically defined when the device detects a document that has a certain length and a MICR line. Another way to implement the second method is to use the envelope as a back document. An envelope is easily distinguishable from the documents on the basis of its thickness profile since the envelopes generally have numerous bends which produce thickness differences throughout the envelope. Although only the content was submitted to the image formation in the previous discussion, the envelope can also be fed to the image forming station. Once the device identifies a document as an envelope based on the thickness profile, the transaction closes. If you want to keep the envelope, the envelope can be classified, and the image of the envelope can be included in the images of the transaction limits. However, typically the envelope simply serves as a divider, so that the envelope is sorted through the deviation (as discussed above), the envelope image is removed and the document preceding the envelope is considered the last document in the transaction. Alternatively, the envelope can be the front document, so that the envelope marks the start of the transaction instead of the end. In certain cases, this second case may not properly identify the limits of the transaction. For example, using the example of the previous check, if a transaction includes three documents and two of the documents are checks, the transaction will be divided into two transactions. Consequently, instead of automatically defining the limits of the transaction by the type of document, the limits can be defined by the transition from one type of document to another. In other words, the end of the transaction is defined by the point at which documents move from one type of document to another. If the documents of the second type are assumed to be the back documents, then the apparatus assumes that when it identifies a transition from a document that is of the second type to a document, which is of the first type, there is a transaction limit. It is assumed that the document of the first type is the first document in the next transaction. Returning again to the check example, if the checks are supposed to be the type of the back document, then all documents or a series of documents are considered in a single transaction until the device identifies a transition from a check to a different type of check. document. The last check is included in the first transaction and the subsequent documents are considered the first document in a second transaction. A fourth method for maintaining transaction limits is a more automatic method that uses extraction sensors. As discussed above, the apparatus includes several sensors 75, 76 that verify the envelope in the extraction station to determine if the content has been extracted. Once the sensors indicate that the content has been extracted, the envelope is transported automatically forward. This indication that the envelope is empty can be used to mark the limits of the transaction.

For example, once the sensors indicate that the apparatus is empty, the apparatus assumes that all documents that are received in the imaging sensor within the next time interval are considered in the same transaction. Any documents after the period of time are considered in the following transaction. For example, the time period can be 2 seconds, so that after receiving an indication from the sensors that all the documents were extracted from an envelope, all the documents that are received in the image formation section in the following 2 seconds will be considered of the transaction that was extracted from the envelope. In the second, third and fourth methods of defining the transaction, it was assumed that the transactions were automatically determined on the basis of different characteristics. In addition, during any of these methods, the operator can manually perform the determination by pressing a button or otherwise, as described in the first method.

Operator Interface As discussed above, the apparatus includes a touch-sensitive screen to provide an input / output device for the operator to interconnect with the operation of the machine. The interconnection between the operator and the apparatus can be explained more clearly in relation to Figures 11-14, which are images of visual screen representations. Referring first to Figure 11, the visual representation includes three general areas. The lower portion of the screen provides a control button toolbar. The operator can control different aspects of the operation of the apparatus with those buttons, as best described below. The second part of the visual representation is the left side of the screen above the toolbar. The left portion displays tab views of the newly scanned documents. The third portion of the visual representation is the right side of the screen above the toolbar. The right portion is an amplified view of the last document to be scanned, unless a different document is selected by the operator, as discussed below. In addition, under the third area a window is provided in which information is displayed, information regarding the lot of the documents. Referring to Figures 11-13, the tab display section of the visual representation can be displayed in one of three views. The first view (Figure 11), the tabs are organized by transaction. In other words, each row represents a transaction, so that the tab of each document is an observable transaction in the same row. For example, the transaction of the first row includes four documents and the views of the tabs of each document are displayed in the first row. In contrast, the first transaction includes only two documents, as can be observed by the two views of the tabs in the visual representation. In the second view (Fig. 12), the tab views are displayed according to the trays 105. The first row presents the view of the tabs of each document in the upper tray, and the tab views are in the same sequence than the documents in the upper tray. Similarly, the second row presents the views of the document tabs in the tray, under the upper tray and so on through the lower tray. In the third view, the views of the tabs are presented simply in the sequence in which the documents were processed. The first tab in the first row represents the first document scanned in a batch, and the tabs continue in the first row until the end of the first row. The following document was then presented in the leftmost position in the second row. The display of the cross views continues in this pattern until all the documents are displayed. Touching the tab view of any document, an enlarged view of the document is displayed in the right viewing portion of the visual representation. Also, if a document is selected in a view (for example, transaction view) and the view is changed (for example, in the tray view), the document is also selected in the changed view. More specifically, if the screen visual representation is a transaction view a document can be selected in a transaction, as shown in Fig. 11 (the document is selected and is identified by a black border surrounding the selected document) . If the operator wishes to retrieve the selected document, it would be desirable for the operator to know in which tray the document is located, and how far is the document from the top of the stack. Consequently, after selecting the document in the transaction view, the operator can switch to the tray view by touching the "Change" button on the button toolbar. Then the tabs will be displayed according to the tray and the document that was selected from the transaction view will remain selected from the tray view so that the operator sees in which tray the document is and counting the documents in the row in which If the selected document is found, the operator can determine how far the document is from the top or bottom of the tray. Similarly, if a document is selected in the tray or sequence mode, and the view is changed to a different mode, the selected document will remain highlighted after the view is changed. The other buttons in the toolbar are related to other features of the device. For example, as discussed above, during the operation, the operator can identify the type of document when the document is dropped on the conveyor 100. Preferably, the five buttons in the center of the tool belt represent the five different types of tools. documents that are being processed by a batch of documents. Consequently, to identify the type of document the operator can simply touch the appropriate button when a document is dropped on the conveyor. In addition, as discussed above, the operator can manually identify the transaction limits. Consequently, after the operator places the last document in a transaction on the gravity carrier, the operator presses the "new transaction" button, thereby indicating that the next document is a new transaction. In addition, it may be desirable to include controls that allow the operator to specify special scanning criteria for a document. For example, if the operator notices that a document is particularly light, the operator can indicate that the document will be scanned in a darker parameter than usual. Similarly, the operator may wish to change the scan of the color image to a gray scale or vice versa, or the operator may wish to change the resolution of the scan. To provide that control, a special manipulation button may be provided on the screen representation. When the operator wishes to specify some kind of special handling, the operator touches the special handling button when the document is dropped on the conveyor. A view of the different special processing options can then be displayed, allowing the operator to control the scanning characteristics on a piece-by-piece basis if necessary.

Rejection processing During a processing, a document can be rejected for one or several reasons, such as too much bias or insufficient space between the documents' for example. There are two main modes of processing the rejected documents and the documents that are upstream of the documents when they were rejected. In the first reject processing mode, the exact sequence of all documents in the batch is not critical. In this way, the rejected piece is sorted to a rejection tray or directed to the deviation 165, and the subsequent documents in the transaction are processed according to the common procedures discussed above. At the end of the batch, all rejected documents are retrieved and processed again. In the second reject processing mode, • the order of documents within a batch or within a transaction is important. Therefore, the rejected and subsequent documents will be processed in such a way as to allow an appropriate order of the pieces to be maintained. Consequently, if the part is rejected, all parts of the input feeder of subsequent images will be rejected. The apparatus then suggests the operator retrieve all rejected documents and re-feed the documents in the appropriate order to summarize the processing.

Editing / Reexploration During processing, the operator can determine that a document needs to be re-scanned again. For example, the operator may notify on screen 20 that an image is too light. The operator selects the document by touching the image of the tab. If the operator is unsure of which tray the document is located in, the operator can press - the change button and change the tray view, which will show the operator the tray in which the document is located, as discussed above. . The operator can then retrieve a document, and then select the option to reexplore a list of editing functions. The document is then dropped on a conveyor and is subsequently scanned. Then the newly scanned image replaces the previous image.

Similarly, if the operator notices that the document needs to be added, as it will be added, to a transaction that has already been processed, the operator selects the option to insert a list of editing functions. The document is then dropped on the conveyor and is then scanned. The newly scanned image is inserted in the appropriate transaction and / or appropriate sequence in the batch. The operator can then replace the physical document in the appropriate document sequence. If the documents are classified or stored according to a sequence number of documents, then the document is inserted in the row of documents according to the sequence number of documents (ie, the order in which the document was originally processed; regardless of the order in which it was re-explored). Alternatively, documents can be classified and / or maintained according to the document scan number (ie the order in which the documents have been scanned). If the documents are kept according to the document scan number, then the document must be in the proper order after being scanned. In certain cases the operator can determine that a document should be deleted. To do this, the operator selects the images from the document tab, and then selects the option to delete from a list of editing functions so that the image is discarded. Another feature is the ability to move a document from one transaction to another or within the sequence in the batch. The document can be measured in one of two ways. First, the operator can select the document by touching the image in the tab. The operator then selects the option to move from a list of editing functions. It is then suggested to the operator to indicate if the document is going to be moved, as to which transaction and which document in the transaction. Alternatively, preferably the operator can move the document by dragging and dropping, touching the tab and dragging the tab across the screen to the appropriate place.

Document Processing with Removable Portions Documents often include a portion that is designed to be detached. More often, a perforation separates the detachable portion of the body from the document. The removable portion of the document is commonly referred to as a skirt. In some cases, the skirt may include information that is useful for further processing of the document. For example, the check may have a skirt that identifies the invoice numbers that correspond to the payment. When processing a batch of documents, it may be desirable to process the checks with the appended apron, rather than removing the skirt. In addition, in certain cases, the skirt must be removed selectively. As a result, documents with skirts can be processed as follows. The operator treats the documents with skirts like other documents: the documents are simply dropped on the drop zone of the conveyor without detaching the skirt. The document with the appended flap is then subjected to image formation in the image forming station, and classified in one of the sorting trays. After the documents are removed from the output tray, the skirts can be removed if desired. Moreover, if the documents with skirts have an identification feature, then the image data can be explored by the identification feature. The documents that have the property can then be classified to a particular tray. In this way, the skirts can be removed in groups instead of one at a time. Since the skirt is to be discarded, it may be desirable to discard the skirt portion of the image as well, thereby reducing the file size for the image data. Accordingly, if a document is identified as a document having a flap, the image data can be analyzed syntactically to retain only the portion of the data corresponding to the portion of the document without the flap. Alternatively, the image data can be analyzed syntactically so that the image of the skirt is in a file and the image of the remaining portion of the document is in a separate file. False documents can also be classified according to various characteristics on the documents. For example, if the documents have a MICR or OCR line, the documents can be classified according to the information in the MICR or OCR line. In an application, documents can be checks, and the MICR line can be read optically or using a MICR reader, as discussed above. The MICR line includes information that identifies the institution from which the check was withdrawn. Depending on the institution from which you received the check, it may be desirable to retain the skirt. Therefore, checks withdrawn from institutions from which the skirt must be retained are classified in a tray, or one of several trays, while checks withdrawn from an institution from which the skirt must be discarded are classified in different trays. In this way, the checks for which the skirt must be removed are separated from the checks from which the skirts must be peeled off and discarded. By separating the documents, the skirts can be detached in groups without removing the skirts that should not be detached. In addition to classifying documents based on various characteristics of a document, the image data can be analyzed syntactically on the basis of several characteristics. Depending on the account identified in the MICR line, the image data can be analyzed syntactically so that only the image in checks without the skirt is retained. The image data scanned by the skirts are deleted. Similarly, based on the account identified in the MICR line of a check, the image data can be parsed in two files: a file for the image data of the check, a file for the image data of the skirt.

Assignment of Reference Information Another aspect of apparatus 10 is the ability to assign one or more pieces of information to one or more documents. The information can be assigned to a single document, or it can be assigned to each document in a transaction, lot or a whole job that includes multiple lots. An interface is provided between the operator and the apparatus, so that the operator can feed the reference information when necessary or desired. One way in which the reference information can be assigned is according to the lot. Typically, batch information is assigned before processing the batch of documents. Each document in the batch is then marked with the batch information. The batch reference information can be physically or electronically marked on the document. The reference information can be physically marked on a piece by printing the reference information on the piece by the printer modules 190. The reference information can be electronically marked by associating the reference information with each piece in a data file. For example, a data file may be produced for each document, which would include information, the sequence number of the piece, the image data for the piece, the date and time the piece was processed and information that is determined electronically, as a MICR line or an OCR line on the part. The batch reference information can be included in the data file for the parts in the batch. Another way in which the reference information can be electronically marked on a piece is by electronically printing the information on the image data. More specifically, the reference information is added to the image data for a part, so that when the image data is viewed, the reference information also appears. The reference information can be superimposed on the image data, so that when the image is viewed, the information appears printed on the document. However, more typically reference information is placed adjacent to the image when the image is viewed. When marking a batch of documents with the batch reference information, it is desirable to identify the reference information before processing the documents. By identifying the reference information of the batch before processing, the reference information can be easily marked on the documents while the documents are processed through the apparatus. In some cases it may be acceptable to identify the lot information after the lot has been processed. The batch reference information can be identified in any of a number of ways. The operator can type reference information on the keyboard through a keyboard or similar input device. Alternatively, the operator can explore in the reference information. For example, the apparatus may include a manual barcode scanning device that can be used to scan the reference information. It is also possible to identify reference information by processing the document. For example, before the first piece of mail is processed, a document with reference information printed on it can be processed through the machine. The reference information can be a bar code, a MICR line, characters that can be read using OCR or otherwise. The apparatus can scan the first piece, read the reference information and then identify the documents with reference information. In addition to marking the documents with reference information according to the batch, the documents can be marked by transaction. A transaction includes one or more documents that were extracted from a single envelope. In certain cases it may be desirable to mark each document in a transaction with reference information that relates to the transaction. As with the reference information of the batch, the reference information of the transaction can be identified manually by the operator input or automatically by the scan information of a document. After the transaction is finished, the device stops the marking of documents with the reference information of the transaction. In addition to the batch and transaction bases, the reference information can be marked on a document on a document-by-document basis. As with the discussion of the reference information of the lot and transaction, the reference information of the document can be identified manually by the operator or automatically. With the reference information of the document the information is marked only in the specific document. However, the document can have several pages. If the document has several pages, each page is marked with the reference information of the document. As discussed above, reference information can be marked on documents on a work, batch, transaction or document basis. In addition, references can be combined so that a particular document in a job can be marked with the job reference information, batch reference information, transaction reference information and document reference information. Moreover, there may be several types of each reference information. For example, the document may have a single piece of batch reference information, two pieces of transaction reference information, and two pieces of reference information of the document. The following examples illustrate the application of the reference information described above. A first example illustrates the use of batch reference information. One of the benefits of scanning images of each document is that the images of the documents can be used during the subsequent processing instead of the original documents. Therefore, it may be possible to discard the documents and simply use the images. However, it may be necessary to examine an original document in the event of a dispute regarding the document or if the document image does not show all the details of the document. Accordingly, it is desirable to provide an efficient method for scanning and archiving documents to facilitate retrieval of documents later if desired.

The first step is to provide a container, such as a file box, in which a unique identifier, such as a barcode, is attached. Before processing a batch of documents, it is suggested that the operator in the observation screen enter an identification number of the file box. The operator scans the bar code using a manual barcode scanning device connected to the apparatus 10. The bar code scanning device reads the bar code and decodes the bar code to determine the identification number of the box archive. Alternatively, the identification number can be printed in the file box and the operator can manually type in the box number in the system. The identification number for the box is the reference information for the batch of documents that will be stored in the box. Other reference information for the lot can also be identified. For example, the operator can type information about the batch of documents as to what type of documents are in the batch or where they were received. After the operator identifies the batch reference number, the operator processes the documents. For example, if the batch of documents was previously extracted from the envelopes, the documents are processed by dropping the documents on the conveyor, alone or in packages. The documents are then scanned and downloaded into the output trays, as described above. During processing, documents are marked electronically with lot data. After processing, the documents are removed from the tray or trays and placed in the file box. Preferably, a sequence number and the identifier of the box are printed on the documents when the documents are processed, and the documents are placed in the archive box in order according to the sequence. Additional processing for processed documents can be effected using the image data of the documents. However, if the physical document needs to be retrieved, the data file for the document will identify the identification number of the box in which the document is located, and the sequence number of where the document is located in the box. Again, when the document needs to be replaced, the correct box and its location can be easily determined since the document is physically and / or electronically marked with the identification number of the box and the sequence number. In this way, the processed document can be easily stored and retrieved efficiently if necessary. At the same time, the identification number of the box can be identified directly from the box and the documents can go directly to the box after processing, so that there is less chance of placing erroneous documents in the wrong box. Moreover, if the identification number of the box is scanned out of the box, such as by means of a bar code scanning device, there is less likelihood that an incorrect identification number will be entered for a box, which would lead to wrong storage of a batch of documents. A second example of using the reference information is an application that uses the transaction reference information. Frequently, important patterns are sent via a courier that tracks the location of the package from the person making the package shipment to the person receiving the package. More typically, packages are followed using the same type of bar code. When the documents are received and then scanned, it is desirable to retain the package tracking information, but it is usually desirable to dispose of the packaging materials.

As a result, a batch of documents in courier envelopes, such as, Federal Express, Express Mail or UPS envelopes, can be processed as follows. Before processing the documents in the envelope, the operator identifies the reference information of the transaction for the documents in the envelope. The operator can type the information, but preferably, the operator simply scans the tracking barcode on the package, a barcode scanning device. After the reference information of the transaction is identified, each document in the transaction is processed in each document, in the transaction it is marked with the transfer information of the transaction. To ensure that the reference information of the transaction is associated with the appropriate document, it is important to ensure that the beginning and end of a transaction are properly defined. The start of a transaction can be manually identified by the operator by pressing a transaction start button. Alternatively, the start can be defined simply by the act of scanning the tracking bar code. The system then processes each document processed later as part of the transaction until the end of the transaction is indicated in some way. The end of the transaction can be identified in one of several ways. For example, the operator can manually press an end of transaction button after the last document of the transaction is processed. Alternatively, the act of scanning in the following tracking bar code may indicate the end of a transaction. Specifically, after scanning a tracking bar code, the system can assume that all subsequent documents are part of the same transaction until the operator scans the next tracking bar code. In other words, the act of scanning the tracking bar code can operate as an indicator of the end of the previous transaction and as the start of the next transaction. As described above, the reference information of the transaction is typically defined before scanning the documents of the transaction. The method described above includes scanning the bar code of the reference information, processing a transaction, and then scanning the next bar code for the next transaction. It may be desirable to identify the reference information for the next transaction while the previous transaction is still being fed into the image forming station. Accordingly, the operator can indicate whether the reference information is for the current transaction or the next transaction. This option can be displayed on the display screen as a choice to be made before feeding the reference information. In this way, while the machine is processing the transaction, the operator can select the next transaction option and then enter the reference information of the transaction for the next transaction. The system will continue processing the current transaction with the previous transaction reference information. After the current transaction is being processed, the operator drops the next transaction on the gravity conveyor and the system marks the documents with the new transaction reference information that the operator. has already fed. Another application of reference information relates to the use of document reference information. In certain cases, the operator can manually identify some information with respect to a document that can not be easily identified from the scanned image. Before exploring the image, the browser identifies the information regarding the document, such as typing information into a system via a keyboard, or another interface, such as a voice recognition system. For exampleIf a document has unusual impressions that are typically not well scanned, someone who looks for a later scanned image of the document may not know that the image may not actually reflect the document. An example is the document that has red ink, which is not scanned as well as other colors. When the image is observed, it may appear that some information is absent or incorrect. If the person observing the document knows that the document has been printed in red ink, the person can easily recognize the need to recover the original document, or alter the processing of the document. For example, during further processing, the operator can see the image of the document to give credit to a customer's account - with a payment. The amount of the check may not be clear on the check because the printing on the check may not be scanned properly (for example, too light, printed in red ink, etc.). The operator may have recognized the problem with printing, manually read the amount of the check, and type the information, such as reference information of the document. During the subsequent processing, the operator that accredits the payment can use the reference information of the document to identify the amount to be credited to the client's account. Another example of the use of the reference information of the document is a memo on the line of a memo, a check or document. A customer can write information in the memo line, as an account number, or some indication that the document requires special processing. When the documents are scanned, the operator can manually read the information and feed the information as reference information of the document. The reference information of the document can be used to classify the document. For example, the reference information of the document may specify that there is a change of address. All documents can be classified separately. Subsequently, an operator can process all document address changes while viewing the image, and typing in the change of address information in the client's account. Each of the above applications was discussed only using a single type of reference information. However, as described above, different types of reference data can be used in combination. For example, in the first example, all documents were marked with batch reference information, with respect to the identification number for the box in which the documents were archived. If the package includes a package with a tracking barcode, the operator can scan the barcode, so that all documents in the package are identified in the tracking information. At the end of the transaction, the operator presses an end of transaction button to indicate the end of the transaction. Also, when one of the documents is processed in the transaction, the operator may notice an abnormality with one of the documents. Before processing the document, the operator can feed the reference information to the specific document, and only that particular document is marked with the document information. In yet another example, a job can include two batches of documents that are mixed, and the documents will be marked with the reference information of the appropriate batch. In that case, the reference information of the appropriate lot will be identified by each document, which can be done in one of several ways. At another point the operator can identify the batch reference information for a first batch, and the documents are marked with the first batch reference information until a document is identified as the start of the second batch. This can be done by the operator indicating that the document is the beginning of the second batch. The document is then marked as if it were in the second batch. Similarly, subsequent documents can be marked as if they were in the second batch until a document that is in the first batch is identified. Alternatively, the batch to which a document belongs can be defined on a document-by-document or transaction-by-transaction basis. An application of the mixed batch would be in which a job of the documents needs to be classified according to certain criteria and stored in two different boxes. Before processing a document job, the operator can scan the first box and identify the identification number of the box as reference information of the batch of a first batch of documents. The operator can then scan the second box to identify the identification number of the box, as reference information of the batch for a second batch of documents. When the documents have been processed, the documents are identified as being in the first batch or the second batch. The reference information of the batch is marked on the respective documents and the documents are classified in the output trays according to the batch. The documents of the first batch are then removed from the appropriate trays and placed in the first box and the documents of the second batch are removed from the appropriate trays and placed in the second batch. As mentioned above, the lot for which the document is identified can be determined in a number of different ways. In one example, in the mixed batch application it is to classify the documents in batches based on the information in the documents. Specifically, a MICR line includes information regarding the institution from which a check was made. The MICR line can be read from the checks when the documents are processed, the documents can then be classified in batches according to the institution. For example, if a transaction has a check that is from a first institution (or groups of institutions), then all the documents in the transaction are identified with the first reference information of the lot. If a transaction has a check that is from a second institution (or groups of institutions), then all documents are identified with the second reference information of the lot. As can be seen in the previous examples, the apparatus 10 provides flexibility to process documents in automated form, while allowing efficient operator intervention to supplement or modify the information stored for the documents. Several applications have been described, however the flexibility of the characteristics of the apparatus allows it to be used in a wide variety of applications using different combinations of different characteristics. Consequently, although modalities and particular applications of the apparatus have been illustrated and described, it is not intended to be limited to the specific modalities and applications.

Claims

NOVELTY OF THE INVENTION Having described the invention as above, property is claimed as contained in the following: CLAIMS 1. An apparatus for processing documents, characterized in that it comprises: an input tray for receiving a stack of envelopes containing documents; a feeder to feed an envelope of the «stack of envelopes; a cutting machine that operates to cut at least one edge of the envelope; an extractor that tries to open the envelope cut by the edge separating the envelope to expose the documents to be removed manually by the operator; a gravity conveyor for receiving the extracted documents, wherein the conveyor is configured to receive and transport documents that are dropped on the conveyor in a generally horizontal configuration; and a scanning device that operates to scan a face of the extracted document. The apparatus according to claim 1, characterized in that it comprises a sorting station that operates to classify the documents. The apparatus according to claim 1, characterized in that it comprises an area for an operator to sit on, where a portion of the gravity conveyor is positioned between the seating area and the extractor. The apparatus according to claim 1, characterized in that the gravity conveyor comprises a roller bed. . . The apparatus according to claim 1, characterized in that it comprises a second feeder adjacent to one end of the gravity conveyor, so that the second feeder operates to receive documents from the gravity conveyor and seriously feed the documents to the scanning device. . The apparatus according to claim 5, characterized in that the second feeder comprises a feeder arm that extends vertically upwards at an angle in relation to the conveyor. The apparatus according to claim 1, characterized by a scrubber to reduce wrinkles of lines or creases in the documents. The apparatus according to claim 7, characterized in that the scrubber comprises a guide having an edge that is preferably at least as wide as the width of the documents, where the scrubber is positioned so that the documents are pulled over the edge when documents are transported through the device. . 9. A method for processing documents, characterized in that it comprises the steps of: dropping a folded document onto a generally horizontal document transport; transporting the document along the document transport to a scanning device; justify a document edge when the document is transported to the scanning device; flatten the document to reduce folds in the document; Scan the documents with the scanning device to obtain a set of image data for the scanned documents. The method according to claim 9, characterized in that the method comprises the step of unfolding the folded document. The method according to claim 9, characterized in that it comprises the step of classifying the documents into one of a plurality of trays. The method according to claim 9, characterized in that the method comprises the step of analyzing the image data for the document to determine a characteristic of the document. The method according to claim 9, characterized in that the step of 'dropping a document comprises dropping a pack of two or more documents in a generally horizontal orientation on the conveyor. The method according to claim 13, characterized in that it comprises the step of opening an envelope containing the package of two or more documents and extracting the documents. 15. The method according to claim 14, characterized in that it comprises the step of feeding an envelope from a mail stack. 16. A method - for processing documents, characterized in that it comprises the steps of: dropping a document on a document transport; transporting the document along the transport of documents to the scanning device; scanning the document to obtain a set of image data corresponding to the image of the document; and download the document in an output tray. 17. The method according to claim 16, characterized in that the document is folded and the method comprises the step of unfolding the folded document. 18. The method according to claim 16, characterized in that it comprises the step of identifying a characteristic of the document and classifying the document on the basis of the characteristic. The method according to claim 16, characterized in that the method comprises the step of analyzing the image data for the document to determine a characteristic of the document. The method according to claim 16, characterized in that the document is a document of a package of two or more documents, and the step of dropping the document comprises dropping the package of documents on the transport of documents. The method according to claim 16, characterized in that the step of dropping the document comprises dropping the document in a generally horizontal orientation on the transport of documents. 22. The method according to claim 16, characterized in that it comprises the step of opening an envelope containing the document and extracting the document. 23. The method according to claim 22, characterized in that it comprises the step of feeding the envelope from a mail stack. 24. The method according to claim 16, characterized in that it comprises a first flattening step, wherein the first portion of the document is flattened when the document is transported to the scanning device. 25. An apparatus for processing documents, characterized in that it comprises: a conveyor configured so that the documents can be manually dropped on the conveyor in the generally horizontal orientation; an operable scanning device for receiving documents from the conveyor and scanning the documents to obtain image data for the documents; non-volatile image storage medium for storing image data for documents; and an operable reorientation element for reorienting the documents when the documents are transported to the scanning device; where the conveyor is operable to transport the documents in a generally horizontal orientation. 26. The apparatus according to claim 25, characterized in that the documents are contained within an envelope and the apparatus comprises an opener that operates to open the envelope, so that the documents can be removed from the envelope. 27. The apparatus according to claim 25, characterized in that it comprises a flattening element that operates to flatten substantially the entire width of each document, when each document is transported to the scanning device. The apparatus according to claim 27, characterized in that the flattening element has a width and comprises a high friction portion which operates to retain a portion of each document at a fixed point relative to the width of the flattening element, and a low friction portion that operates to allow documents to slide across the width of the flattening element when documents are transported to the scanning device. 29. The apparatus according to claim 25, characterized in that it comprises a feeder having a feeding contact line for pressing the documents, wherein the feeder is configured to receive documents having a leading edge bent upwards, so that the front edge is placed on top of the feed contact line, where the feeder is configured to direct the leading edge of those documents folded downwards, towards the feed contact line. 30. The apparatus according to claim 25, characterized in that the conveyor comprises a substantially horizontal conveyor belt or a plurality of rollers configured to support and transport the documents. 31. The apparatus according to claim 25, characterized in that it comprises a communication link for connecting the apparatus to the file server, so that the image data can be exported to the file server via a communication link. 32. The apparatus according to claim 25, characterized in that it comprises a feeder for receiving the documents from the conveyor and seriously feeding the documents to the scanning device, where the conveyor is configured to transport the documents to the feeder. without pressing the documents. The apparatus according to claim 25, characterized in that the documents are contained within envelopes, and the apparatus comprises: an input tray for receiving a stack of envelopes containing documents, a feeder for feeding an envelope of the stack of envelopes; a cutting machine that operates to cut at least one edge of the envelope; and an extractor that operates to open the envelope cut on the edge separating the envelope to expose the documents for manual removal by the operator. 34. The apparatus according to claim 25, characterized in that it comprises a verifier or monitor for displaying the scanned image data. 35. The apparatus according to claim 25, characterized in that it comprises a flattening element configured to flatten the document widthwise without substantially curling the document. 36. The method according to claim 35, characterized in that it comprises a second flattening step, wherein the second portion of the document is flattened when the document is transported to the scanning device. 37. The method according to claim 36, characterized in that the second portion of the document is larger than the first portion of the document. 38. A method for processing a document contained within an envelope, characterized in that it comprises the steps of: dropping a document in a generally horizontal orientation on a document transport; transporting the document that fell along the transport of documents to an image acquisition device; scanning the document of the image acquisition device to obtain a set of image data for the document; store the image data in a non-volatile storage medium. 3'9. The method according to claim 38, characterized in that it comprises the steps of: feeding the envelope of a stack of envelopes; and open the envelope, so that the document can be removed from the envelope. 40. The method according to claim 39, characterized in that the step of opening the envelope comprises the steps of: feeding the envelope of a stack of envelopes to a cutting station; cut one or more edges of the document in a cutting station; and separate the envelope to present the document for the removal of the envelope. 41. The method according to claim 38, characterized in that the step of storing the image data comprises exporting the image data to a file server. 42. The method according to claim 38, characterized in that it comprises the step of reorienting the document when the document is transported to the scanning device. 43. The method according to claim 42, characterized in that the reorientation step comprises reorienting the document when the document is transported in a generally horizontal orientation. 44. The method according to claim 38, characterized in that it comprises the steps of: transporting the document to a feeder that operates to keep the document in a fixed position; and feeding the document from the feeder to the scanning device. 45. The method according to claim 44, characterized in that it comprises the step of reorienting the document during the step of transporting the document to the feeder. 46. The method according to claim 45, characterized in that the step of reorienting the document comprises reorienting the document while the document is in a generally horizontal orientation. 47. The method according to claim 38, characterized in that it comprises the step of analyzing the image data to determine a characteristic of the document. 48. The method according to claim 46, characterized in that it comprises the step of classifying the document into one of a plurality of trays in response to the determined characteristic.