JPS6252506B2 - - Google Patents

Description

[Detailed description of the invention] The present invention relates to a document reading method.

Recently, a manuscript reading method using a solid-state image sensor has been put into practical use. A solid-state image sensor is one in which a large number of extremely small light-receiving elements are closely arranged in a line in one direction and integrated to provide a self-scanning function. One example is the company's CCD (charge coupled device). Through the use of such solid-state image sensors, document reading devices have been simplified and downsized, and document reading speeds have been dramatically increased.

However, the number of bits that can be read with a single solid-state image sensor is currently limited, and if the document to be read is large, some ingenuity is required to read it with sufficient resolution. It is.

The use of a plurality of solid-state image sensors to read a large document is a common technique as described above.

FIG. 1 shows the principle of a document reading method using two solid-state image sensors. In the figure, numerals 1 and 2 are solid-state image sensors, numerals 3 and 4 are imaging lens systems, and numerals L are main scanning lines, L ₁ and L ₂
indicate the scanning line portions, respectively.

The main scanning line L is a fixed linear area virtually set on the document reading surface, and at each moment of reading, information on the document portion on the main scanning line L is read.

The main scanning line L consists of two scanning line parts.
Equally divided into L ₁ and L ₂ , these scanning line parts
Solid-state image sensors 1 and 2 are associated with L ₁ and L ₂ , respectively. That is, the solid-state image sensors 1 and 2 are fixedly installed at fixed positions corresponding to the scanning line portions L ₁ and L ₂ , and are connected to the imaging lens systems 3 and 4 .
are positioned so that the reduced images of the document portions on the scanning line portions L ₁ and L ₂ are imaged onto the light-receiving area of the corresponding solid-state image sensor.

When reading a document, a long and narrow area including the main scanning line L is illuminated in a slit shape, and the document to be read is oriented so that the document surface faces the solid-state image sensors 1 and 2 and crosses the main scanning line L. , transported in a direction perpendicular to the drawing. The direction in which the document surface moves due to conveyance is referred to as the sub-scanning direction.

As the document is conveyed, the solid-state image sensors 1 and 2 are driven, and information on the document is sequentially read and converted into electrical signals. In this example, the document is conveyed in the sub-scanning direction, but the document is stationary, and the position of the main scanning line may move in the sub-scanning direction with respect to the document surface.

As described above, in the reading method using a plurality of solid-state image sensors, there is a problem in the succession of readings between adjacent solid-state image sensors. If this reading transfer is not carried out properly, when the image signal generated by reading is reproduced as an image, the reproduced image will be distorted.

Conventionally, the following method has been used to handle this handover.

In other words, a minute area is set at the junction of the scanning line parts, and the image of this minute area is focused on either light-receiving area of the adjacent solid-state image sensor. The signals from the solid-state image sensors are compared to recognize the overlapping signal, that is, the area where the image signal of the minute area exists, and one of the overlapping signals is removed.

This method requires a complicated electric circuit, and since the overlapping signal obtained is an image signal of information on the document part in a minute area, when the above information is halftone, it is difficult to detect the overlapping part. There is a drawback that boundaries of other parts cannot be properly recognized. In addition, as in the method disclosed in Japanese Patent Application Laid-open No. 138715/1984, marks on a test chart are read in advance with multiple solid-state image sensors, and reading is taken over according to the mark reading position on each solid-state image sensor. It is also known to adjust. This method is useful if the relationship between the image of each part of the main scanning line and each solid-state image sensor remains constant, but for sub-scanning of the document, it is necessary to realize relative movement between the document image and the solid-state image sensor. Since a mechanical movable part is required for this purpose, as the movable part is repeatedly moved, it is difficult to avoid slight distortion in the position and attitude of the imaging optical system.
When such distortion occurs, the image of each part of the main scanning line is shifted with respect to the corresponding solid-state image sensor. Moreover, images are formed in each solid-state image sensor by mutually separate imaging lens systems, and image shifts related to each imaging lens system occur independently of each other. Therefore, even if the image of the same mark on the test chart is formed on the adjacent solid-state image sensor and the reading succession is adjusted, the above-mentioned error occurs, but as time passes, the position of the image of the mark becomes different from each other. Since they are shifted separately relative to the solid-state image sensor, it is difficult to avoid discontinuity in the handover and the occurrence of pixels that are not read at the handover portion or pixels that are read redundantly.

An object of the present invention is to form images of main scanning lines divided into a plurality of parts onto corresponding solid-state image sensors using separate imaging lens systems, and to move two reflecting mirrors in the sub-scanning direction at a ratio of 2:1. A document reading method that performs sub-scanning by moving at a speed ratio, and also allows reading to be taken over continuously without causing pixels that are not read or pixels that are read redundantly in the takeover section. On offer.

The present invention will be described below with reference to the drawings.

FIG. 2 shows one of the apparatuses for carrying out this invention.
Only the main parts of the example are shown. In order to avoid confusion, the same reference numerals are used from FIG. 1 onwards for items that are unlikely to be confused.

In FIG. 2, reference numeral 10 denotes a document placement glass;
Reference numerals 11, 12 and 13 indicate reflecting mirrors, respectively.

The reflecting mirrors 11 and 12 are movable, and when reading a document, each moves in the direction of the arrow from the position shown by the solid line to the position shown by the broken line. That is, the reflecting mirror 11 moves along the document placement glass 10 in the sub-scanning direction together with an illumination lamp (not shown) to sub-scan the document O placed on the document placement glass 10. , moves at 1/2 the moving speed of the reflecting mirror 11, and determines the optical path length from the illuminated part to the light receiving area of the solid-state image sensors 1 and 2, and the light flux to the imaging lens systems 3 and 4. Keep the angle of incidence constant.

That is, in this apparatus, the main scanning line moves with respect to the document surface of the document O, which is kept stationary. The above illumination lamp, reflecting mirrors 11, 12, 1
3. The imaging lens systems 3 and 4 constitute a reading optical system.

As shown in FIG. 3, the main scanning line L is equally divided into scanning line portions L ₁ and L 2 , and solid-state image sensors 1 and ₂ correspond to each of the scanning line portions L ₁ and L 2 . A minute area L ₁₂ is set at the junction of _{the two} , and the images of this area L ₁₂ are formed on the light receiving areas of the solid-state image sensors 1 and 2.

Now, in FIG. 2, the reflecting mirror 11 shown by the solid line
The position indicates the home position of the reflecting mirror 11, and in the upper part of the reflecting mirror 11 located at this home position, the document placement glass 1
0, the document reading start position 10a is set in a narrow shape, and the document reading start position 10a is set to have a narrow width.
The above scanning line area is on the surface of the document placement glass.
As shown in FIG. 4, a mark M is marked in a portion corresponding to the minute region L ₁₂ at the junction of L ₁ and L ₂ . FIG. 5 is an enlarged view of mark M. The width of the hatched part of the mark M and the interval between the hatched parts are determined so that when the mark M is read by the solid-state image sensors 1 and 2, each mark M is read with 1 bit. Mark M and
The positional relationship with the solid-state image sensors 1 and 2 is adjusted so that the reduced images of the shaded areas are respectively formed on the aperture of a single light-receiving element.

Then, in this state, the solid-state image sensors 1 and 2 are simultaneously driven, and the document reading start position 10a is
If this is converted into a signal, the signals obtained from the solid-state image sensors 1 and 2 should be as shown in FIGS. 6 and 6, respectively.

Now, in the manuscript reading method according to the present invention,
The reading sequence is performed as follows.

When the document O to be read is placed on the document placement glass 10 as shown in FIG. 2 and the device is operated, an illumination lamp (not shown) lights up and the document is set at the document reading start position 10a. A slit-shaped area including the main scanning line is illuminated. When the light emission of the illumination lamp becomes stable, the illumination lamp and the reflecting mirror 11 move in the direction of the arrow to start sub-scanning, and at the same time the reflecting mirror 12 moves in the direction of the arrow.

During this period, the control circuit 7-1 in FIG. 7 clears the counter 7-12 and sends the drive circuit 7-2,
Activate 7-3. Each time the drive circuits 7-2 and 7-3 receive a clock pulse, they drive the solid-state image sensors 1 and 2, respectively, and each time the solid-state image sensors 1 and 2 are driven, they output the accumulated image signal one bit at a time. do. The output image signals are amplified by amplifiers 7-4 and 7-5, respectively, and then white information is set to a logic value "1" and black information is set to a logic value "0" by comparators 7-6 and 7-7. to 2
Valued. During this time, the counter 7-12 counts the clock pulses, in other words, the prime number of image signals generated from the solid-state image sensor, and stores the counted number in the latch circuits 7-10, 7-1.
11.

When the applied binary signal is "0", "1", and "0" successively, the encoders 7-8 and 7-9 calculate the count number at the second "0" using a latch circuit. 7-
10, 7-11. Assume now that the number of counts stored in the latch circuit 7-10 is _N1 and the number of sounds stored in the latch circuit 7-11 is _N2 , as shown in FIG.

The count numbers N ₁ and N ₂ stored in the latch circuits 7-10 and 7-11 are immediately transferred to the control circuit 7-1, and the control circuit 7-1 uses the results to
Calculate N ₁ −N ₂ and store N ₁ and N ₁ −N ₂ . The process up to this point takes place in an extremely short time, and in this state, the position of the main scanning line has not yet left the reading start position.

When the above N ₁ and N ₁ −N ₂ are obtained, the control circuit 7 −
1 is a comparator 7-6, 7-7, an encoder 7-8, 7
-9, the operation of the latch circuits 7-10, 7-11 and the counter 7-12 is stopped, and the analog switch 7-13 is operated instead, and this time the drive circuit 7-2
Accordingly, only the solid-state image sensor 1 is activated. When the reading progresses to N ₁ −N ₂ , the control circuit 7
-1 starts driving the solid-state image sensor 2 by the driving circuit 7-3.

As can be easily seen, when the reading by the solid-state image sensor 1 has progressed to N ₁ - _{N 2} , _{the scanning line portion L 1 in} FIG. It means that it has been read.

Thereafter, when the reading by the solid-state image sensor 1 reaches _N1 , the control circuit 7-1 turns off the analog switch 7-13 and turns on the analog switch 7-14 instead. Then, the image signals applied to the buffer 7-15 are the signals read up to _N1 by the solid-state image sensor 1, followed by the signals read by the solid-state image sensor 2. That is, the reading is transferred from the solid-state image sensor 1 to the solid-state image sensor 2 using the mark M as the boundary.

When the scanning by the solid-state image sensor 2 is completed, the control circuit 7-1 turns off the analog switch 7-14.
13 is turned on, and scanning by the solid-state image sensor 1 is restarted.

This reading scan is repeated, and as the main scanning line moves, the original O is read sequentially, and with each scan,
An image signal with no excess or deficiency can be obtained.

In the example described above, the reading is taken over with the mark M at the reading start position 10a as the boundary, but the reading may be taken over anywhere within the minute area _L12 . In other words, after the reading scan by the solid-state image sensor 1 progresses to N ₁ - _{N 2} and the reading scan by the solid-state image sensor 2 starts, at which moment during the reading scan by the solid-state image sensor 1 advances to N ₁ , the reading starts. You may take over.

In the present invention, the marks are read before each document is read, and the reading result is used to set the reading handover timing. Even if the image formation position of the mark is shifted individually for each sensor, the reading handover timing is always set so that the reading handover is continuous.

Further, in the above description, the case where two solid-state image sensors are used has been described, but the present invention is of course applicable to the case where three or more solid-state image sensors are used.

Furthermore, the mark marked at the document reading position is not limited to the one shown in FIG. 5, but may have an appropriate shape.

[Brief explanation of the drawing]

FIG. 1 is a diagram for explaining a reading method using a plurality of solid-state image sensors, FIG. 2 is a side view schematically showing only the main parts of an example of an apparatus for carrying out the present invention, and FIG. 6 to 6 are diagrams for explaining the present invention, and FIG. 7 is a block diagram showing one example of an electric circuit for implementing the present invention. L...Main scanning line, _L1 , _L2 ...Scanning line portion, _L12 ...Minute area, M...Mark, 1,
2...Solid image sensor.

Claims (1)

[Claims] 1 The main scanning line is divided into n in the longitudinal direction (n≧
2. Each of the n scanning line portions is imaged by each of the n imaging lens systems in a corresponding manner to each of the n solid-state image sensors, and the direction perpendicular to the main scanning line is the sub-scanning direction. Then, the two movable reflectors of the reading optical system including the two movable reflectors and the n imaging lens systems are moved in the sub-scanning direction 2:
In a document reading method in which the information pattern on the document is converted into an image signal by sub-scanning the document by moving the document at a speed ratio of 1, n-1 joints of n scanning line portions ( n-1) microscopic areas are set, and the images of these microscopic areas are formed on the light receiving areas of mutually adjacent solid-state image sensors by corresponding imaging lens systems, and the above-mentioned reading optical system At the document reading start position set corresponding to the home position, the above (n
-1) The marks marked in minute areas are read by each solid-state image sensor prior to each document reading, and the mark read by the m-th (1≦m≦n-1) solid-state image sensor is signal and
The signal of the mark read by the (m+1)th solid-state image sensor allows the reading to be taken over from the m-th solid-state image sensor to the m+1-th solid-state image sensor when reading the original. A document reading method characterized by setting the timing of handover so that it is carried out continuously within a document.