JPH06301784A - Image processor - Google Patents

Abstract

PURPOSE:To easily separate the top side and reverse side when an image of paper, etc., meets the requirements of plural dictionary data. CONSTITUTION:The image processor is equipped with an optical sensor 1 which gathers the transmitted light signal and reflected light of the paper, etc., in pixel units and a 1st decision part 2 which decides the kind of the paper, etc., and whether it is genuine or not by matching the transmission image of the paper, etc., obtained from the signals with dictionary data. This processor has a top/reverse difference arithmetic part 3 which calculates the difference between the density addition value of the respective pixels of the top-side reflected image of the paper, etc., obtained from the reflected light signal and the density addition value of the respective pixels of the reverse-side reflected image when the decision part 2 decides that the transmission image meets the requirements of plural dictionary data and a 2nd decision part 4 which decides whether the obtained transmission image is the top or reverse side image by comparing the difference value obtained by the top/reverse difference arithmetic part 3 with a specific value.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an improvement of an image processing apparatus for judging the authenticity of paper sheets. In a paper sheet processing apparatus such as an ATM (Automatic Deposit and Payment Machine), the whole paper sheet (banknote) to be inspected is divided into minute areas, and each minute area is patterned by an optical sensor or a thickness sensor. The lightness and shade are measured, and the measured values are converted into gradation signals and stored. Then, a predetermined image processing is performed on this gradation signal, and the authenticity of the bill is determined by comparing with the dictionary data created from the genuine bill in advance.

A light transmission image is one of the authenticity determination items of such a sheet processing apparatus. The light transmission image is used as a powerful determination item for distinguishing a copy ticket because it cannot be determined as a genuine note unless the front and back sides of the paper sheets are correctly overlapped.

However, when viewed with such a light transmission image, there are cases where the image viewed from the front side and the image viewed from the back side are quite similar depending on the type of paper sheet, and a plurality of (front and back)
There was a problem that I passed the dictionary data of. In this case, since the inserted paper sheets go against the general principle that the authenticity is judged only after passing a single dictionary, a judgment method for solving this is required.

[0004]

2. Description of the Related Art FIG. 6 is a block diagram of a conventional example, and FIG. 7 is a flow chart of authenticity determination processing. In bill discrimination in ATM etc.,
Optical patterns, thickness sensors, etc. are used to measure the shades and outlines of patterns, and the measured values are converted into gradation signals, etc.
After the skewed paper sheets are rotated and aligned, and paper sheets with different densities are subjected to various image processing such as density correction,
The authenticity of the bill is determined by comparing and collating with the image of a dictionary prepared in advance.

FIG. 6 is a block diagram showing a part of a bill discriminating apparatus in ATM. Here, the optical sensor 11 is, for example, one in which two light emitting units and light receiving units (not shown) are alternately arranged in a line and are arranged above and below to constitute a light transmitting sensor. By transporting the banknote 10 between them, a light transmission image of each minute area of the banknote 10 is obtained.

That is, while the banknote 10 is being conveyed, the processor 15 controls the driver 12 of the light emitting unit to scan in the line direction while sampling in the conveying direction and to transmit the light transmission signal by the corresponding light receiving unit. Is received, the obtained light transmission signal is amplified by the amplifier 13, and then converted into digital data by the A / D converter 16.

This digital data represents the gradation of each sampled micro area (pixel), and the image processing section 17
After being stored as a banknote image in the image storage unit 18, various image processing is performed, and the dictionary comparison unit 19 compares with the dictionary data to determine the authenticity.

FIG. 7 shows an example of a procedure for discriminating authenticity based on this image. First, in the case of a banknote conveyed obliquely, the image is aligned by rotational movement, and density correction and the like are performed. After that, the dictionary data prepared in advance in the dictionary data section 20 is compared and compared in pixel units corresponding to each pixel to be uniquely specified. Regarding the denomination and direction determined by the above basic judgment, further details, for example, comparison of density addition values in the vertical and horizontal directions, discrimination based on data obtained by thickness sensor, magnetic sensor, etc. are performed, and authenticity is discriminated. To be done.

[0009]

When collating with dictionary data, a plurality of dictionary data may be passed. ATM
In this way, it is possible to discriminate whether the banknote is inserted from the front side or the back side.For example, when a used 1,000-yen bill is inserted, the obtained images are the front side transparent images of the dictionary data. It may pass both backside transmission images.

In this case, the paper sheet is rejected because it violates the principle that it will not be accepted unless it can be uniquely determined, even though the paper sheet is a genuine note. The present invention has been made in view of the above problems, and an object of the present invention is to provide a paper sheet processing apparatus that easily distinguishes the front and back of paper sheets that have passed a plurality of dictionaries.

[0011]

In order to achieve the above object, an image processing apparatus of the present invention is provided with an optical sensor (1) for collecting light transmission signals and light reflection signals of paper sheets in pixel units, and the signal. An image processing apparatus comprising: a first determination unit (2) for determining the type and authenticity of the paper sheet by collating the transparent image of the paper sheet obtained from When the image passes the plurality of dictionary data in the first determination unit, the density addition value of each pixel in the front side reflection image of the paper sheet obtained from the reflection signal and the density of each pixel in the back side reflection image Front and back difference calculator that calculates the difference from the added value
(3) and the difference value calculated by the front / back difference calculation unit is compared with a predetermined value to determine the front / back of the obtained transparent image.
And a determination unit (4).

In addition, in the second judging section (4),
A method of calculating a front-back density difference by a density addition value of pixels corresponding to a reflection image of a predetermined portion of the paper sheet, and comparing the front-back density of the paper sheet with a corresponding predetermined value, or
A method is used in which a front-back density difference is calculated at a predetermined pixel in the reflection image of the paper sheet and compared with a corresponding predetermined value to determine the front-back.

[0013]

First, the first judging section 2 compares and collates the transparent image with the dictionary data. Here, when the types of paper sheets pass a plurality of dictionaries and cannot be uniquely determined, the front-back difference calculation unit 3
The difference value between the front and back (upper and lower) of the added value of density of each pixel of the reflection image is calculated, and the second determination unit 4 compares it with a predetermined value corresponding to this difference value, for example, a predetermined value obtained from each front and back genuine bill. Then, the front and back are judged.

FIG. 2 shows an example of the procedure of authenticity determination. For example, when the optical sensor having the structure shown in FIG. 4 is used, the values of the transmitted light amounts on the front and back sides are similar in the transmitted image. Also, (upper reflection added value-lower reflection added value)
(It is called a decision function) utilizes the fact that there is a certain difference.

That is, when a dark image of a person or the like exists on the front surface of a paper sheet and only a thin pattern or number is written on the back surface, a difference in density addition value surely appears, and the difference value separates the front and back sides. It will be possible.

As for the method of obtaining the difference value between the front and back sides by the reflected light, as shown in FIG. 2, the method of obtaining the difference value by the added value of the entire surface of the paper sheet, the existence part of a specific person, etc. are known in advance. In this case, there are a method of obtaining with a specific rectangle in which a sufficient difference can be expected, and a method of obtaining with a specific pixel such as a specific pattern or mark in which a sufficient difference can be expected.

As described above, even if a plurality of dictionary data are passed, it is possible to easily separate the front and back, and the reliability of image processing is improved.

[0018]

FIG. 3 is a block diagram of an embodiment, FIG. 4 is a principle diagram of a light transmission sensor and a light reflection sensor, and FIG. 5 is a process flowchart of the embodiment.

FIG. 3 is a block diagram of the banknote appraisal portion of the ATM shown in the conventional example of FIG. 6, in which reference numeral 21 is a vertical difference calculation unit, which corresponds to the front-back difference calculation unit 3 of FIG. The density addition value of each pixel of the upper reflection image of the banknote 10 obtained by the upper reflection light (upper reflection addition value) and the density addition of each pixel of the lower reflection image of the banknote 10 obtained by the lower reflection light Value (lower reflection addition value) and difference value = (upper reflection addition value−lower reflection addition value) are calculated.

Reference numeral 22 is a front / back determination unit, which is the second determination unit 4 in FIG.
In response to the instruction (corresponding to the dictionary comparison unit 19 (first determination unit 2)) (passes a plurality of dictionaries), the vertical difference calculation unit 21 calculates the above-mentioned difference, and the result, The front and back are judged by comparing with a corresponding predetermined value stored in the difference data section 23.

Reference numeral 23 is a difference data section, which is composed of a non-volatile memory, and stores the difference value (predetermined value) obtained with a genuine note. In addition, the same reference numerals as those in FIG. 6 represent the same objects.

FIG. 4 shows an example of the structure of the optical sensor 11 in FIG. 3, in which N light-emitting portions (upper light-emitting portion 26, lower light-emitting portion are not shown) and light-receiving portions (upper light-emitting portion) are provided in the upper and lower portions, respectively. When the upper light emitting part 26 is made to emit light and the lower light receiving part 28 receives light with an array sensor having a light receiving part 27, a lower light receiving part 28),
When the lower light emitting section emits light and the upper light receiving section 27 receives light, it operates as a light transmission sensor.When the upper light emitting section 26 emits light and the upper light receiving section 27 receives light, the lower light emitting section emits light. When the lower light receiving section 28 receives light, it operates as a light reflection sensor.

In FIG. 3, the light transmission sensor section 24 and the light reflection sensor section 25 include a driver for driving each light emitting section, a selection section for selecting each light receiving section, and the like, and the processor 15 selectively controls the light receiving section and the light emitting section. By doing so, a light transmission sensor and a light reflection sensor that form a pair are set, and scanning is performed in a predetermined order. As a result, the vertical transmission signal and the vertical reflection signal are output to the light receiving unit, the amplification unit 13 and the A / D conversion unit 16 output the gradation data of each pixel over the entire surface of the banknote, and the image processing unit 17 outputs the gradation data of the banknote 10. The upper and lower transmission image data and the upper and lower reflection image data are assembled and stored in the image storage unit 18.

Here, as shown in FIG. 2, the method for obtaining the difference value is, as a first method, a method for obtaining a difference value by the density addition value of the entire surface of the banknote 10, and as a second method, a specific method. When the presence part of a person or the like is known in advance, a method of obtaining a specific rectangle for which a sufficient difference can be expected, and a third method are:
There is a method of obtaining a specific pixel such as a specific pattern or a mark that can be expected to have a sufficient difference, but the determination operation will be described below with reference to FIG.

In FIG. 5, the normalized transmission data and reflection data in which the inclination, size, direction, density, etc. of the image have been corrected by the operation described in the conventional example are stored in the image storage unit 18.
Shows the operation after being stored in. (S1) The dictionary comparison unit 19 compares and collates the image data of the inserted banknote 10 (paper sheet) with all the dictionary data of the dictionary data unit 20 as a basic determination process.

If the type and direction of the banknote 10 are uniformly determined in this collation, detailed comparison processing is performed by combining pixels. If this comparison process is passed, the bill 10 is confirmed as true, and if it is not accepted, it is rejected as a counterfeit bill. (S2) In the above basic determination, when a plurality of dictionary data are passed, the dictionary comparison unit 19 notifies the front / back determination unit 22 to perform front / back separation processing. (S3) By this notification, the front / back surface determination unit 22 becomes
Let 21 calculate the difference. That is, the up-down difference calculation unit 21 reads the reflection image data (gradation data) on the upper side and the lower side of the image storage unit 18 on a pixel-by-pixel basis, adds up the pixels on the entire surface of the banknote 10, and calculates the difference between the upper and lower sides. .

The front / back determination unit 22 sets two sets of front and back of the difference value and the corresponding denomination stored in the difference data unit 23 (the denomination is discriminated because it has passed a plurality of dictionaries). Compare with the difference value and determine the closer one (front and back) or the one within the specified value. Then, based on the confirmation result, (S
Perform the detailed comparison processing shown in 1). (S4) In the basic judgment, if the corresponding dictionary data does not exist, reject it as reject.

As described above, when a plurality of dictionary data are passed, they can be uniformly determined by the difference between the upper and lower reflection images, and the inconvenience that even a genuine note is judged as a counterfeit ticket is improved.

In the embodiment, the basic judgment and the detailed judgment are carried out. However, the present invention is not limited to this, and it is needless to say that the judgment can be set in advance as one item.

[0030]

As described above, according to the present invention, it is possible to reliably perform the above-described method by introducing the difference value between the upper and lower sides of the reflected light into a sheet for which it is difficult to determine the front and back sides of the sheet by a transmission image. Since it is possible to separate the paper sheets that have been conventionally rejected as fake paper sheets from the conventional paper sheets, it is possible to accurately distinguish old paper sheets.

Further, all the methods of the present invention can be easily realized by software, and no modification of hardware is required, so that the effect is extremely large.

[Brief description of drawings]

FIG. 1 is a principle diagram of the present invention.

FIG. 2 is an explanatory diagram of the present invention.

FIG. 3 is a configuration diagram of an embodiment.

FIG. 4 Principle diagram of light transmission sensor and light reflection sensor

FIG. 5 is a process flowchart of the embodiment.

FIG. 6 is a block diagram of a conventional example

FIG. 7: Flow chart of authenticity determination processing

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 Optical sensor 2 1st determination part 3 Front-back difference calculation part 4 2nd determination part 10 Banknote 11 Optical sensor 12 Driver 13 Amplifier part 14 Operation part 15 Processor 16 A / D conversion part 17 Image processing part 18 Image storage part 19 Dictionary comparison unit 20 Dictionary data unit 21 Vertical difference calculation unit 22 Front / back determination unit 23 Difference data unit 24 Light transmission sensor unit 25 Light reflection sensor unit 26 Upper light emitting unit 27 Upper light receiving unit 28 Lower light receiving unit

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁵ Identification number Office reference number FI technical display location G07D 7/00 E 9340-3E // G06F 15/30 D 7343-5L

Claims (3)

[Claims] 1. An optical sensor (1) for collecting light transmission signals and light reflection signals of a paper sheet on a pixel-by-pixel basis, and a transmission image of the paper sheet obtained from the signal is collated with dictionary data. An image processing apparatus comprising: a first determination unit (2) for determining the type and authenticity of the paper sheet, wherein when the transparent image passes a plurality of dictionary data in the first determination unit, A front-back difference calculation unit (3) for calculating a difference between the density addition value of each pixel in the front-side reflection image of the paper sheet obtained from the reflection signal and the density addition value of each pixel in the back-side reflection image; A second determination unit that compares the difference value calculated by the difference calculation unit with a predetermined value to determine the front and back of the obtained transparent image.
(4) An image processing device comprising: 2. A front and back density difference value is calculated by calculating a density difference value between the front and back sides from a density addition value of pixels corresponding to a reflection image of a predetermined portion of the paper sheet and comparing with a corresponding predetermined value. The image processing apparatus according to claim 1, wherein the image processing apparatus is configured to determine. 3. The front and back density difference values are calculated at predetermined pixels in the reflected image of the paper sheet, and the front and back are judged by comparing with a corresponding predetermined value. The image processing device described.