阅读说明：本技术 纸币识别装置、纸币处理装置和纸币识别方法 (Paper money discriminating apparatus, paper money processing apparatus, and paper money discriminating method ) 是由 番匠谷利彦 池本良 永井千裕 于 2020-02-18 设计创作，主要内容包括：本发明提供纸币识别装置、纸币处理装置和纸币识别方法。用于识别纸币的的纸币识别装置具有：一次判定部,将所获取的红外反射图像、和红外反射图像用基准数据进行比较,判定在所述红外反射图像内是否存在相对于所述红外反射图像用基准数据成为允许范围外的异常区域；二次判定部,在所述异常区域存在的情况下,将与所获取的红外透射图像的所述异常区域对应的区域即判定对象区域、和红外透射图像用基准数据进行比较,判定所述判定对象区域相对于所述红外透射图像用基准数据是否为允许范围内；和主判定部,在通过所述二次判定部而判定为所述判定对象区域为允许范围内的情况下,对至少排除了所述红外反射图像的所述异常区域的区域进行真伪判定处理。(The invention provides a bill discriminating apparatus, a bill processing apparatus, and a bill discriminating method. A bill discriminating apparatus for discriminating a bill includes: a first-order determination unit that compares the acquired infrared reflection image with reference data for an infrared reflection image, and determines whether or not an abnormal region outside an allowable range with respect to the reference data for an infrared reflection image exists in the infrared reflection image; a secondary determination unit that, when the abnormal region exists, compares a determination target region, which is a region corresponding to the abnormal region of the acquired infrared transmission image, with reference data for an infrared transmission image, and determines whether or not the determination target region is within an allowable range with respect to the reference data for an infrared transmission image; and a main determination unit that performs an authentication determination process on an area excluding at least the abnormal area of the infrared reflection image when the secondary determination unit determines that the determination target area is within an allowable range.)

1. A paper money discriminating apparatus for discriminating paper money is characterized by comprising:

an image acquisition unit that acquires an infrared reflection image and an infrared transmission image of a banknote;

a storage unit for storing reference data for infrared reflection images and reference data for infrared transmission images based on genuine banknotes;

a first order determination unit that compares a 1 st infrared image, which is one of the infrared reflection image and the infrared transmission image acquired by the image acquisition unit, with 1 st reference data, which is one of the infrared reflection image reference data and the infrared transmission image reference data corresponding to the 1 st infrared image, and determines whether or not an abnormal region outside an allowable range with respect to the 1 st reference data exists in the 1 st infrared image;

a secondary determination unit that, when the abnormal region is present in the 1 st infrared image, compares a determination target region, which is a region corresponding to the abnormal region in a 2 nd infrared image that is the other of the infrared reflection image and the infrared transmission image acquired by the image acquisition unit, with 2 nd reference data, which is the other of the infrared reflection image reference data and the infrared transmission image reference data corresponding to the 2 nd infrared image, and determines whether or not the determination target region is within an allowable range with respect to the 2 nd reference data; and

and a main determination unit that performs an authenticity determination process on an area excluding at least the abnormal area of the 1 st infrared image when the secondary determination unit determines that the determination target area is within an allowable range with respect to the 2 nd reference data.

2. The banknote recognition apparatus according to claim 1,

the storage section further stores a reference value of the size of the abnormal region,

the main determination unit compares the size of the abnormal region with the reference value stored in the storage unit when the abnormal region is present in the 1 st infrared image and the secondary determination unit determines that the determination target region is not within the allowable range with respect to the 2 nd reference data, and performs the authenticity determination process on a region from which at least the abnormal region of the 1 st infrared image is excluded if the size of the abnormal region is smaller than the reference value.

3. The banknote recognition apparatus according to claim 1 or 2,

the 1 st infrared image is the infrared reflection image,

the primary determination unit compares the infrared reflection image with the reference data for the infrared reflection image,

the 2 nd infrared image is the infrared transmission image,

the secondary determination unit compares a determination target region of the infrared transmission image with the reference data for the infrared transmission image.

4. The banknote recognition apparatus according to claim 3,

the infrared reflection image reference data includes a reference value of a pixel value of each pixel of the infrared reflection image,

the infrared transmission image reference data includes a reference value of a pixel value of each pixel of the infrared transmission image,

the primary determination unit determines, as the abnormal region, a region in the infrared-reflected image that is formed of pixels having pixel values smaller than a corresponding reference value included in the reference data for infrared-reflected image,

the secondary determination unit determines that the determination target region is within an allowable range with respect to the reference data for infrared transmission image when the pixel value of each pixel of the determination target region is larger than the corresponding reference value included in the reference data for infrared transmission image.

5. The banknote recognition apparatus according to claim 1 or 2,

the 1 st infrared image is the infrared transmission image,

the primary determination unit compares the infrared transmission image with the reference data for the infrared transmission image,

the 2 nd infrared image is the infrared reflection image,

the secondary determination unit compares a determination target region of the infrared reflection image with the reference data for the infrared reflection image.

6. A bill handling apparatus is characterized in that,

a banknote recognition device according to any one of claims 1 to 5.

7. A bill discriminating method for discriminating a bill, comprising:

an image acquisition step, namely acquiring an infrared reflection image and an infrared transmission image of the paper money;

a primary determination step of comparing a 1 st infrared image, which is one of the infrared reflection image and the infrared transmission image acquired in the image acquisition step, with 1 st reference data corresponding to the 1 st infrared image among reference data for an infrared reflection image and reference data for an infrared transmission image based on a genuine bill, and determining whether or not an abnormal region outside an allowable range with respect to the 1 st reference data exists in the 1 st infrared image;

a second determination step of, when the abnormal region exists in the 1 st infrared image, comparing a determination target region, which is a region corresponding to the abnormal region in a 2 nd infrared image that is the other of the infrared reflection image and the infrared transmission image acquired in the image acquisition step, with 2 nd reference data, which is one of the reference data for the infrared reflection image and the reference data for the infrared transmission image and corresponds to the 2 nd infrared image, and determining whether or not the determination target region is within an allowable range with respect to the 2 nd reference data; and

a main determination step of performing an authenticity determination process on an area excluding at least the abnormal area of the 1 st infrared image when the secondary determination step determines that the determination target area is within an allowable range with respect to the 2 nd reference data.

Technical Field

The invention relates to a bill discriminating apparatus, a bill processing apparatus, and a bill discriminating method. More specifically, the present invention relates to a banknote recognition apparatus and a banknote recognition method suitable for determining authenticity of a transported banknote, and a banknote processing apparatus including such a banknote recognition apparatus.

Background

Conventionally, in a banknote handling apparatus having a banknote recognition device, during conveyance of banknotes, denomination recognition, authenticity determination, damage determination, serial number recognition, and the like of the banknotes are performed. For example, by analyzing an image of a bill read by an optical linear sensor, the denomination of the bill, the authenticity of the bill, the wear of the bill, the character recognition of a serial number written on the bill, and the like are performed.

For example, patent document 1 discloses a paper contamination discriminating apparatus that receives reflected light or transmitted light of light irradiated to a paper such as a bill with a light receiving sensor, and discriminates whether or not the paper is contaminated based on an output signal of the light receiving sensor. The apparatus arbitrarily sets a detection region on a scan line for detecting contamination of a sheet, adds up a binarized signal obtained by binarizing an output signal of a light receiving sensor in the detection region for each reference timing, and adds up an output signal of a light receiving sensor in the detection region for each reference timing, and discriminates whether contamination is present, whether partial contamination is present, and whether contamination is present as a whole based on the 2 addition results.

Further, patent document 2 discloses a plurality of banknote recognition methods for recognizing a banknote by comparing an output pattern of an electric signal output from a light receiving portion that receives reflected light reflected by the banknote or transmitted light transmitted through the banknote with a standard output pattern stored in advance. For example, a method is disclosed in which a region in which the output of an electric signal from a light-receiving portion is larger than a predetermined value is excluded from comparison targets for identifying banknotes. Further, a method of correcting the standard output pattern in accordance with the output of the electric signal from the light receiving section is disclosed.

Disclosure of Invention

Problems to be solved by the invention

In order to prevent forgery, infrared non-absorbing ink is sometimes used in printing of banknotes. Since the printed portion based on the infrared non-absorbing ink absorbs visible light and does not absorb infrared light, the pattern printed by the ink can be visually recognized by a visible light source, but cannot be imaged by an optical line sensor under an infrared light source, and the banknote can be imaged only in a state where the pattern is lost. In a printed portion based on the infrared non-absorbing ink, an output of an optical line sensor based on light reflected by a bill when infrared light is irradiated to the bill (hereinafter, also referred to as a reflected infrared output) is relatively large, and an output of an optical line sensor based on light transmitted through the bill when infrared light is irradiated to the bill (hereinafter, also referred to as a transmitted infrared output) is relatively small. In contrast, in a forged portion printed by an infrared-absorbing ink without using an infrared-non-absorbing ink, since infrared light is absorbed, both reflected infrared output and transmitted infrared output are small.

Therefore, a portion having a low reflected infrared output can be determined to be likely to be counterfeited with respect to the reference template prepared in advance by learning from the genuine bill. Specific examples of such a determination method include the following methods. That is, first, reflected infrared output (infrared reflection image) is collected in all the print areas of the banknotes to be judged. The collected reflected infrared output is then compared to a reference template, and pixels outside the threshold permit are marked as spurious pixels, the number of which is counted. Then, the total count of the forged pixels is compared with a threshold value set separately, whereby the authenticity is determined.

However, in the above-described determination method, since the authenticity determination is performed based only on the number of pixels other than the threshold allowance using only the reflected infrared output, the medium in which the oil is immersed, which is hard to be visually recognized as the dirt but affects the reflected infrared output and the transmitted infrared output, may be erroneously determined as a counterfeit.

Further, if the bill is dirty, the infrared light is absorbed at that portion, and therefore the reflected infrared output is small. However, in the above-described determination method, since the authenticity determination is performed based only on the number of pixels other than the threshold allowance using only the reflected infrared output, in the case where a medium having a large number of local stains is found in a plurality of locations, the medium may be erroneously determined as a counterfeit note when the total area of the locations is large.

In contrast, the paper contamination discriminating device described in patent document 1 discriminates whether or not contamination is present on paper, and there is no disclosure of a problem of erroneously discriminating a medium having contamination as a counterfeit note and a solution therefor.

In the banknote recognition method described in patent document 2, since the comparison target for recognizing the banknote is limited based on the output value of the electrical signal using the reflected light or the transmitted light, there is a high possibility that the counterfeit portion is excluded from the comparison target. Further, if a part of the banknote having a plurality of portions is dirty, it may be erroneously determined as a counterfeit banknote.

The present invention has been made in view of the above-described situation, and an object thereof is to provide a banknote recognition apparatus, a banknote processing apparatus, and a banknote recognition method that can reduce erroneous judgment of a dirty/damaged banknote as a counterfeit banknote.

Means for solving the problems

In order to solve the above problems and achieve the object, the present invention is a banknote recognition apparatus for recognizing a banknote, comprising: an image acquisition unit that acquires an infrared reflection image and an infrared transmission image of a banknote; a storage unit for storing reference data for infrared reflection images and reference data for infrared transmission images based on genuine banknotes; a first order determination unit that compares a 1 st infrared image, which is one of the infrared reflection image and the infrared transmission image acquired by the image acquisition unit, with 1 st reference data, which is one of the infrared reflection image reference data and the infrared transmission image reference data corresponding to the 1 st infrared image, and determines whether or not an abnormal region outside an allowable range with respect to the 1 st reference data exists in the 1 st infrared image; a secondary determination unit that, when the abnormal region is present in the 1 st infrared image, compares a determination target region, which is a region corresponding to the abnormal region in a 2 nd infrared image that is the other of the infrared reflection image and the infrared transmission image acquired by the image acquisition unit, with 2 nd reference data, which is the other of the infrared reflection image reference data and the infrared transmission image reference data corresponding to the 2 nd infrared image, and determines whether or not the determination target region is within an allowable range with respect to the 2 nd reference data; and a main determination unit that performs an authenticity determination process on an area excluding at least the abnormal area of the 1 st infrared image when the secondary determination unit determines that the determination target area is within an allowable range with respect to the 2 nd reference data.

In the above-described invention, the storage unit further stores a reference value of a size of the abnormal region, and the main determination unit compares the size of the abnormal region with the reference value stored in the storage unit when the abnormal region exists in the 1 st infrared image and the secondary determination unit determines that the determination target region is not within an allowable range with respect to the 2 nd reference data, and performs the authenticity determination process on a region from which at least the abnormal region of the 1 st infrared image is excluded if the size of the abnormal region is smaller than the reference value.

In the above invention, the 1 st infrared image is the infrared reflection image, the primary determination unit compares the infrared reflection image with the reference data for the infrared reflection image, the 2 nd infrared image is the infrared transmission image, and the secondary determination unit compares a determination target region of the infrared transmission image with the reference data for the infrared transmission image.

In the above-described invention, the reference data for infrared reflection image includes a reference value of a pixel value of each pixel of the infrared reflection image, the reference data for infrared transmission image includes a reference value of a pixel value of each pixel of the infrared transmission image, the primary determination unit determines a region composed of pixels having a pixel value smaller than a corresponding reference value included in the reference data for infrared reflection image in the infrared reflection image as the abnormal region, and the secondary determination unit determines the determination target region as being within an allowable range with respect to the reference data for infrared transmission image when the pixel value of each pixel of the determination target region is larger than the corresponding reference value included in the reference data for infrared transmission image.

In the above invention, the 1 st infrared image is the infrared transmission image, the primary determination unit compares the infrared transmission image with the infrared transmission image reference data, the 2 nd infrared image is the infrared reflection image, and the secondary determination unit compares a determination target region of the infrared reflection image with the infrared reflection image reference data.

The present invention is a banknote handling apparatus including the banknote recognition device.

Further, the present invention is a banknote recognition method for recognizing a banknote, comprising: an image acquisition step, namely acquiring an infrared reflection image and an infrared transmission image of the paper money; a primary determination step of comparing a 1 st infrared image, which is one of the infrared reflection image and the infrared transmission image acquired in the image acquisition step, with 1 st reference data corresponding to the 1 st infrared image among reference data for an infrared reflection image and reference data for an infrared transmission image based on a genuine bill, and determining whether or not an abnormal region outside an allowable range with respect to the 1 st reference data exists in the 1 st infrared image; a second determination step of, when the abnormal region exists in the 1 st infrared image, comparing a determination target region, which is a region corresponding to the abnormal region in a 2 nd infrared image that is the other of the infrared reflection image and the infrared transmission image acquired in the image acquisition step, with 2 nd reference data, which is one of the reference data for the infrared reflection image and the reference data for the infrared transmission image and corresponds to the 2 nd infrared image, and determining whether or not the determination target region is within an allowable range with respect to the 2 nd reference data; and a main determination step of performing an authenticity determination process on a region of the abnormal region excluding at least the 1 st infrared image when the secondary determination step determines that the determination target region is within an allowable range with respect to the 2 nd reference data.

Effects of the invention

According to the banknote recognition apparatus, the banknote processing apparatus, and the banknote recognition method of the present invention, erroneous judgment that a dirty or damaged banknote is judged as a counterfeit banknote can be reduced.

Drawings

Fig. 1 is a diagram for explaining an outline of a bill identifying apparatus and a bill identifying method according to embodiment 1.

Fig. 2 is a perspective view schematically showing the appearance of the banknote handling apparatus according to embodiment 1.

Fig. 3 is a block diagram illustrating the configuration of the banknote recognition apparatus according to embodiment 1.

Fig. 4 is a schematic cross-sectional view illustrating the configuration of an image acquisition unit included in the banknote recognition apparatus according to embodiment 1.

Fig. 5 is a schematic diagram showing an example of an infrared reflection image acquired by an image acquisition unit included in the banknote recognition apparatus according to embodiment 1.

Fig. 6 is a schematic diagram showing an example of an infrared transmission image acquired by an image acquiring unit included in the banknote recognition apparatus according to embodiment 1.

Fig. 7 is a diagram for explaining an example of a method of calculating the size of an abnormal region in an infrared reflection image by a secondary determination unit included in the banknote recognition apparatus according to embodiment 1.

Fig. 8 is a flowchart showing the processing procedure of the banknote recognition process in the banknote recognition apparatus and the banknote recognition method according to embodiment 1.

Description of the reference symbols

1: paper money discriminating device

10: control unit

11: identification part

12: authentication judging section

12 a: primary judging unit

12 b: secondary determination unit

12 c: main judging part

13: damage completion determination unit

20: detection part

21: image acquisition unit

22: magnetic detection unit

23: UV detection part

30: storage unit

110. 120: optical linear sensor

111. 121: light source for reflection

112. 122: condensing lens

113. 123: light-receiving part

124: light source for transmission

210: infrared reflectance image

211: abnormal area

220: infrared transmission image

221: determination target region

300: paper money processing device

301: paper feeding bucket

302: reject part

303: operation part

305: display unit

306a to 306 d: collecting part

310: shell body

311: conveying path

BN: paper money

Detailed Description

Preferred embodiments of a bill identifying apparatus, a bill handling apparatus, and a bill identifying method according to the present invention will be described below with reference to the drawings.

In the present specification, the reflection image refers to an image based on the intensity distribution of light that is emitted to the bill and reflected by the bill, and a reflection image in which the light emitted to the bill is infrared light is also referred to as an infrared reflection image. The transmission image is an image based on the intensity distribution of light that is irradiated to the bill and transmitted through the bill, and a transmission image in which light irradiated to the bill is infrared light is also referred to as an infrared transmission image.

[ paper money discriminating apparatus and paper money discriminating method ]

First, an outline of the bill identifying apparatus and the bill identifying method according to the present embodiment will be described with reference to fig. 1. The main feature of the present embodiment is as shown in I, II below.

I. By checking both the reflected infrared output and the transmitted infrared output (infrared reflection image and infrared transmission image), contamination due to oil immersion and a forged portion are separated (see the center and right side of each image in fig. 1).

And II, separating local dirt and forged parts (the center and the left side of each image in the figure 1) according to the conditions around the pixels outside the threshold allowance.

According to these features, erroneous judgment of a worn and dirty coin as a counterfeit coin can be reduced. Hereinafter, the respective features will be described in more detail.

(I) When the banknote is impregnated with the oil, irregularities of the fibers of the banknote decrease, and the amount of transmitted light increases, that is, the amount of reflected light decreases. Therefore, the reflected infrared output is relatively small and the transmitted infrared output is relatively large at the dirty portion caused by oil immersion. As described above, since there is a possibility that a portion having a low reflected infrared output is forged with respect to the reference template, a dirty portion due to oil immersion may be erroneously determined as a forged portion only from the reflected infrared output. However, in the present embodiment, the transmitted infrared output of the portion is also confirmed, and when the transmitted infrared output of the portion is large, it is determined that the portion is not a counterfeit portion but a dirty portion due to oil immersion. This makes it possible to exclude the portion determined to be contaminated by oil immersion from the counterfeit pixels, and thus it is possible to reduce erroneous determination of a dirty or worn coin as a counterfeit coin.

(II) empirically, the characteristics of forgery are mostly present in intact sites. Therefore, when there is a region that is likely to be forged around a region that is likely to be forged (the reflected infrared output and/or the transmitted infrared output is not allowed), it is determined that the region is forged, and if not, it is determined that the region is locally soiled. This can reduce erroneous judgment of the local stain as a counterfeit portion.

< Structure of paper money handling apparatus >

The structure of the banknote handling apparatus according to the present embodiment will be described with reference to fig. 2. The banknote handling apparatus according to the present embodiment may have a configuration shown in fig. 2, for example. The banknote handling apparatus 300 shown in fig. 2 is a small-sized banknote handling apparatus installed on a table for use, and includes: a banknote recognition device (not shown in fig. 2) for performing a banknote recognition process, a hopper 301 for placing a plurality of banknotes to be processed in a stacked state, 2 reject units 302 for discharging rejected banknotes such as counterfeit banknotes and unidentified banknotes when the banknotes fed from the hopper 301 into the casing 310 are rejected banknotes such as counterfeit banknotes and unidentified banknotes, an operation unit 303 for an operator to input an instruction, 4 collecting units 306a to 306d for sorting and collecting banknotes of which denomination, authenticity and defective are recognized in the casing 310, and a display unit 305 for displaying information such as a result of the recognition and counting of the banknotes and the collecting conditions of the collecting units 306a to 306 d. Based on the result of the damage determination by the banknote recognition device, among the 4 collection units 306a to 306d, normal banknotes are stored in the collection units 306a to 306c, and dirty and damaged banknotes are stored in the collection unit 306 d. Note that the method of distributing the banknotes to the stacking units 306a to 306d can be arbitrarily set.

< Structure of paper money discriminating apparatus >

The structure of the banknote recognition device according to the present embodiment will be described with reference to fig. 3. As shown in fig. 3, the banknote recognition device 1 according to the present embodiment includes a control unit 10, a detection unit 20, and a storage unit 30.

The control Unit 10 is configured by a program for realizing various processes stored in the storage Unit 30, a CPU (Central Processing Unit) for executing the program, various hardware controlled by the CPU, and a logic device such as an FPGA (field programmable Gate Array). The control unit 10 controls each unit of the banknote recognition apparatus 1 based on signals output from each unit of the banknote recognition apparatus 1 and control signals from the control unit 10 in accordance with a program stored in the storage unit 30. The control unit 10 also functions as a recognition unit 11, an authentication determination unit 12, and a damage completion determination unit 13 by a program stored in the storage unit 30.

The detection unit 20 includes an image acquisition unit 21, a magnetism detection unit 22, and a UV detection unit 23. The image acquiring unit 21 acquires an image of the bill. The magnetic detection unit 22 has a magnetic sensor (not shown) for measuring magnetism, and detects magnetism of magnetic ink printed on the banknotes, security thread, or the like by the magnetic sensor. The magnetic sensor is a magnetic linear sensor in which a plurality of magnetic detection elements are arranged in a straight line. The UV detection unit 23 includes an ultraviolet irradiation unit (not shown) and a light receiving unit (not shown), and detects fluorescence generated when the ultraviolet irradiation unit irradiates the bill with ultraviolet light or ultraviolet light transmitted through the bill.

The storage unit 30 is configured by a nonvolatile storage device such as a semiconductor memory or a hard disk, and stores various programs and various data for controlling the bill discriminating device 1. The storage unit 30 stores reference data for infrared reflection images, reference data for infrared transmission images, and a reference value for the size of an abnormal region (hereinafter also referred to as a size reference value).

The infrared-reflection-image reference data and the infrared-transmission-image reference data are reference templates used as determination references in the authenticity determination process of the authenticity determination section 12, and are parameter sets prepared in advance by machine learning for each denomination of a bill to be recognized based on the pixel values (output values) of each pixel of the infrared-reflection image and the infrared-transmission image of the genuine bill (genuine bill). Therefore, the infrared-reflected-image reference data and the infrared-transmitted-image reference data represent reference values of pixel values of pixels of the infrared reflected image and the infrared transmitted image of the bill to be authenticated, respectively.

The size reference value is a threshold value used as a criterion for determining the size of the abnormal region in the authentication process of the authentication determining section 12.

The recognition unit 11 compares the feature pattern of the image of the bill acquired by the image acquisition unit 21 with the denomination recognition template, which is the feature pattern of each denomination of the bill, and determines the denomination of the bill.

The authenticity judging section 12 judges authenticity of the bill based on an image of the bill obtained by irradiating infrared light, and details thereof will be described later. The authenticity judging section 12 compares the detection signals of the magnetic detecting section 22 and the UV detecting section 23 with reference data for authenticity identification, and judges the authenticity of the bill. In this manner, the authenticity judging section 12 performs a plurality of types of authenticity judgment processing on the banknote, judges that the banknote is a genuine banknote when all judgment results are true, and judges that the banknote is a counterfeit banknote when any one judgment result is counterfeit.

< Structure of image acquisition section >

The configuration of the image acquisition unit 21 will be described with reference to fig. 4. As shown in fig. 4, the image acquisition section 21 has optical linear sensors 110 and 120 arranged to face each other. Between the optical linear sensors 110 and 120, a gap for conveying the bill BN is formed, which constitutes a part of the conveyance path 311 of the bill handling apparatus. The optical linear sensors 110 and 120 are respectively located at the upper and lower sides of the conveying path 311.

The optical linear sensor 110 includes a light source 111 for reflection, a condenser lens 112, and a light receiving unit 113. The reflection light source 111 irradiates the upper surface of the bill BN with light of a predetermined wavelength (invisible light such as infrared light, monochromatic light such as red/green/blue, or visible light such as white light). The condenser lens 112 emits light from the reflection light source 211, and condenses the light reflected by the bill BN. The light receiving section 113 has a plurality of light receiving elements (not shown) arranged in a straight line in a direction (main scanning direction) orthogonal to the conveying direction (sub-scanning direction) of the bill BN, and converts the light condensed by the condenser lens 112 into an electric signal. The optical linear sensor 110 converts the electric signal converted by the light receiving section 113 into a digital signal and outputs the digital signal.

The optical linear sensor 120 includes a reflection light source 121, a condenser lens 122, a light receiving unit 123, and a transmission light source 124. The reflection light source 121 and the transmission light source 124 irradiate the lower surface of the bill BN with light of a predetermined wavelength (invisible light such as infrared light, monochromatic light such as red/green/blue, or visible light such as white light). The condenser lens 122 emits light from the reflection light source 121, and condenses the light reflected by the bill BN. The light receiving section 123 has a plurality of light receiving elements (not shown) arranged linearly in a direction perpendicular to the conveyance direction of the bill BN, and converts the light condensed by the condenser lens 122 into an electric signal. The optical linear sensor 120 converts the electric signal converted by the light receiving section 123 into a digital signal and outputs the digital signal.

The transmission light source 124 is disposed on the optical axis of the condenser lens 112 of the optical linear sensor 110, and a part of the light emitted from the transmission light source 124 transmits through the banknote BN, is condensed by the condenser lens 112 of the optical linear sensor 110, and is detected by the light receiving unit 113.

The optical linear sensors 110 and 120 repeatedly take images of the banknotes BN conveyed in the conveying direction at regular time intervals (exposure of the light receiving elements) and output signals, respectively, and the image acquiring unit 21 acquires an image of the entire banknote BN (image acquiring step). Specifically, the image acquiring unit 21 acquires the reflection image of the upper surface of the bill BN and the transmission image of the bill BN based on the output signal of the optical linear sensor 110, and acquires the reflection image of the lower surface of the bill BN based on the output signal of the optical linear sensor 120. Further, the image acquiring section 21 acquires the infrared reflection image of the upper surface of the banknote BN and the infrared reflection image of the lower surface of the banknote BN as the reflection images of the banknote BN. Further, the image acquiring section 21 acquires the infrared transmission image of the banknote BN as the transmission image of the banknote BN.

< Structure of true/false judging section >

The configuration (function) of the authentication determining unit 12 will be described with reference to fig. 5 and 6. As shown in fig. 3, the authentication determining unit 12 includes a primary determining unit 12a, a secondary determining unit 12b, and a main determining unit 12 c.

As shown in fig. 5, the primary determination unit 12a performs a primary determination process (primary determination step) of comparing the infrared-reflected image 210 of the banknote (all printed regions) as the 1 st infrared image with the infrared-reflected-image reference data (reference data for infrared-reflected images corresponding to the denomination recognized by the recognition unit 11) as the 1 st reference data, and determining whether or not there is an abnormal region 211 that is an area outside the allowable range with respect to the infrared-reflected-image reference data in the infrared-reflected image 210. This allows detection of a dirty portion, a partially dirty portion, and/or a counterfeit portion due to oil immersion as the abnormal region 211.

More specifically, the linear judgment unit 12a compares the pixel value of each pixel of the infrared-reflected image 210 with the corresponding reference value included in the reference data for the infrared-reflected image, and if there is a pixel having a pixel value smaller than the reference value, determines a region composed of these pixels as the abnormal region 211.

As shown in fig. 6, when 1 or more abnormal regions 211 are present in the infrared reflection image 210, the secondary determination unit 12b performs a secondary determination process (secondary determination step) of comparing each of 1 or more determination target regions 221 corresponding to 1 or more abnormal regions 211 in the infrared transmission image 220 of the banknote (all printed regions) as the 2 nd infrared image with the reference data for the infrared transmission image (reference data for the infrared transmission image corresponding to the denomination recognized by the recognition unit 11) as the 2 nd reference data, and determining whether or not each determination target region 221 is within an allowable range with respect to the reference data for the infrared transmission image. This makes it possible to determine whether or not the abnormal region 211 and the determination target region 221 are dirty parts due to oil immersion.

More specifically, the secondary determination unit 12b determines 1 or more regions corresponding to 1 or more abnormal regions 211 (that is, having the same portion as each of 1 or more 1 st abnormal regions 211) in the infrared transmission image 220 as the determination target regions 221. The secondary determination unit 12b compares the pixel value of each pixel of each determination target region 221 with a corresponding reference value included in the reference data for infrared transmission image, and determines that the determination target region 221 is within an allowable range with respect to the reference data for infrared transmission image if the pixel value of each pixel is greater than the reference value, and determines that the determination target region 221 is not within the allowable range with respect to the reference data for infrared transmission image if the pixel value of each pixel is equal to or less than the reference value.

When the secondary determination unit 12b determines that at least one determination target region 221 is within the allowable range with respect to the infrared transmission image reference data, the main determination unit 12c performs the authenticity determination process on the infrared reflection image 210 excluding at least 1 or more abnormal regions 211 corresponding to 1 or more determination target regions 221 determined to be within the allowable range (main determination step). Thus, when the abnormal region 211 is a dirty region due to oil immersion, the authenticity determination can be performed on the region excluding the dirty region, and therefore, erroneous determination that a dirty coin due to oil immersion is determined as a counterfeit coin can be reduced.

More specifically, when the secondary determination unit 12b determines that at least one determination target region 221 is within the allowable range with respect to the infrared transmission image reference data, the main determination unit 12c sets, as the authenticity determination region to be subjected to the authenticity determination process, a region excluding at least 1 or more abnormal regions 211 corresponding to 1 or more determination target regions 221 determined to be within the allowable range (that is, having the same location as that of the 1 or more determination target regions 221), in the infrared reflection image 210. The main judgment unit 12c performs the authentication judgment process on the authentication judgment area by a normal method. For example, in the example shown in fig. 5 and 6, the abnormal region 211 on the right side of the infrared-reflected image 210 is excluded from the authenticity determination region because the pixel value of each pixel in the determination target region 221 on the right side of the infrared-transmitted image 220 corresponding thereto is larger than the corresponding reference value.

When the secondary determination unit 12b determines that at least one determination target region 221 is not within the allowable range with respect to the infrared transmission image reference data, the main determination unit 12c performs the following processing. That is, the main determination unit 12c compares the respective sizes of 1 or more abnormal regions 211 corresponding to 1 or more determination target regions 221 determined not to be within the allowable range with the size reference value stored in the storage unit 30, and if the sizes are smaller than the size reference value, performs the authenticity determination process on the infrared reflection image 210 from which at least the abnormal regions 211 have been removed (second main determination step). Thus, when the abnormal area 211 is a partially contaminated area having a small size, the authenticity determination can be performed on the area excluding the area, and therefore, erroneous determination that a contaminated coin due to local contamination is determined as a counterfeit coin can be reduced.

More specifically, when the secondary determination unit 12b determines that at least one determination target region 221 is not within the allowable range with respect to the infrared transmission image reference data, the main determination unit 12c identifies 1 or more abnormal regions 211 corresponding to 1 or more determination target regions 221 determined not to be within the allowable range (that is, having the same portion as each of the 1 or more determination target regions 221), and calculates the size of each of the abnormal regions. The main judgment unit 12c compares the calculated size of each abnormal region 211 with the size reference value stored in the storage unit 30, and if the size is smaller than the size reference value, sets a region excluding the abnormal region (abnormal region having a size smaller than the size reference value) 211 from the infrared-reflected image 210 as the authenticity judgment region. The main judgment unit 12c performs the authentication judgment process on the authentication judgment area by a normal method. For example, in the example shown in fig. 5 and 6, the pixel values of the respective pixels in the center and left determination target regions 221 of the infrared transmission image 220 are equal to or less than the corresponding reference values, but in the infrared reflection image 210, the sizes of the respective corresponding left abnormal regions 211 are smaller than the size reference value, and therefore these abnormal regions 211 are excluded from the authenticity determination region. On the other hand, in the infrared reflection image 210, since the size of the corresponding abnormal region 211 in the center is equal to or larger than the size reference value, the abnormal region 211 is not excluded from the authenticity determination region.

As a method of calculating the size of each abnormal region 211 by the secondary determination unit 12b, for example, a method of binarizing the infrared reflection image 210, and then performing filter processing or label processing to calculate the number of pixels constituting each abnormal region 211 is given.

In the case of using the filter process, for example, the infrared reflection image 210 is first binarized, and the binarized bill region is reversed in black and white. As a result, as shown in the upper part of fig. 7, a portion where the infrared reflection output is low and there is a possibility of forgery becomes a white pixel. Subsequently, predetermined filter processing is performed. Then, the pixel value of each pixel is set to a predetermined threshold value, and the number of pixels larger than the threshold value is counted as fake pixels. Here, if the threshold is set to, for example, 3, then in the case of the left image in fig. 7, the central 4 pixels are within the allowable range of the threshold, and no counterfeit pixel is detected. On the other hand, in the case of the right diagram of fig. 7, the central 4 pixels are outside the allowable range of the threshold value, and are detected as forged pixels.

< paper money discriminating processing >

The process performed by the banknote recognition apparatus 1 will be described with reference to fig. 8. This process is repeated for each banknote fed to the banknote recognition device 1.

When the banknotes are fed into the banknote recognition apparatus 1 (S11: YES), the recognition unit 11 recognizes the denominations of the banknotes based on the images of the upper and lower surfaces of the banknotes acquired by the image acquisition unit 21 and the denomination recognition templates as described above (S12).

Next, as described above, the primary judging unit 12a performs the primary judging process of comparing the infrared reflection image (1 st infrared image) 210 of the bill with the infrared reflection image reference data (1 st reference data) corresponding to the denomination recognized in step S12, and judging whether or not there is an abnormal region 211 outside the allowable range with respect to the infrared reflection image reference data in the infrared reflection image 210 (S13).

Next, when 1 or more abnormal regions 211 are present in the infrared reflection image 210, the secondary determination unit 12b performs the secondary determination process of comparing 1 or more determination target regions 221 corresponding to 1 or more abnormal regions 211 in the infrared transmission image (2 nd infrared image) 220 of the bill with the infrared transmission image reference data (2 nd reference data) corresponding to the denomination recognized in step S12, and determining whether or not each determination target region 221 is within the allowable range with respect to the infrared transmission image reference data (S14), as described above.

When it is determined in step S14 that at least one determination target area 221 is within the allowable range with respect to the infrared transmission image reference data, the main determination unit 12c excludes 1 or more abnormal areas 211 corresponding to 1 or more determination target areas 221 determined to be within the allowable range from the authenticity determination area of the infrared reflection image 210 as described above (S15).

In step S14, when it is determined that at least one determination target region 221 is not within the allowable range with respect to the infrared transmission image reference data, the sizes of 1 or more abnormal regions 211 corresponding to 1 or more determination target regions 221 determined not to be within the allowable range are compared with the size reference value, and if the sizes are smaller than the size reference value, the abnormal region 211 is further excluded from the authenticity determination region of the infrared reflection image 210 (S15).

As described above, the main judging unit 12c performs the authentication judgment process on the authentication judgment area specified in step S15 to judge the authenticity of the banknote (S16).

The processing in steps S13 to S16 for the authenticity check is performed for each of the top and bottom surfaces of the banknote. That is, the authenticity determination process described above based on the infrared reflection image 210 and the infrared transmission image 220 of the upper surface of the bill, and the authenticity determination process described above based on the infrared reflection image 210 and the infrared transmission image 220 of the lower surface of the bill are performed separately.

The authentication determining unit 12 compares the detection signals of the magnetic detecting unit 22 and the UV detecting unit 23 with the reference data for authentication, and determines the authenticity of the banknote (S17).

If the authenticity judgment results in steps S16 and S17 are both true (yes in S18), the authenticity judging section 12 judges that the banknote is a genuine banknote (S19), and if at least 1 of the authenticity judgment results in steps S16 and S17 is false (no in S18), the authenticity judging section 12 judges that the banknote is a counterfeit banknote (S20).

When the banknote is determined to be genuine, the end damage determination unit 13 determines the end damage of the banknote (S21).

Then, the control unit 10 sends out the banknotes for which the authenticity and damage determination has been completed from the banknote recognition apparatus 1 (S22), and ends the processing of the banknote recognition apparatus 1.

As described above, in the present embodiment, the primary judging unit 12a detects a portion that is likely to be falsified as the abnormal region 211 based on the infrared reflection image 210 of the banknote, the secondary judging unit 12b judges whether or not the judgment target region 221 corresponding to the abnormal region 211 is a dirty portion due to oil immersion based on the infrared transmission image 220, and when the judgment target region 221 is judged to be a dirty portion due to oil immersion, the main judging unit 12c performs the authenticity judgment process on the authenticity judgment region from which at least the abnormal region 211 is excluded in the infrared reflection image 210, so that erroneous judgment for judging a dirty banknote due to oil immersion as a counterfeit banknote can be reduced.

In the present embodiment, even when the secondary determination unit 12b determines that the determination target area 221 is a counterfeit portion, the main determination unit 12c determines whether or not the abnormal area 211 is a local dirty portion based on the size of the abnormal area 211 corresponding to the determination target area 221. Further, if the abnormal area 211 is a local contamination portion, the main judgment unit 12c also performs the authenticity judgment process by excluding the abnormal area 211 from the authenticity judgment area, and therefore, it is possible to reduce erroneous judgment that a contaminated coin due to local contamination is judged as a counterfeit coin.

In the above embodiment, the case where the determination is performed in the order of the infrared-reflected image 210 and the infrared-transmitted image 220, that is, the case where the 1 st infrared image is the infrared-reflected image 210, the first-order determination unit 12a compares the infrared-reflected image 210 with the reference data for the infrared-reflected image, the 2 nd infrared image is the infrared-transmitted image 220, and the second-order determination unit 12b compares the determination target region 221 of the infrared-transmitted image 220 with the reference data for the infrared-transmitted image has been described, but the determination may be performed in the order of the infrared-transmitted image 220 and the infrared-transmitted image 210. That is, the 1 st infrared image may be the infrared transmission image 220, the primary determination unit 12a may compare the infrared transmission image 220 with the reference data for the infrared transmission image, the 2 nd infrared image may be the infrared reflection image 210, and the secondary determination unit 12b may compare the determination target region of the infrared reflection image 210 with the reference data for the infrared reflection image. In this case, as compared with the case where the determination is performed in the order of the infrared reflection image 210 and the infrared transmission image 220, basically, only the order of the processing is changed before and after, and the processing can be performed in the order of (1) to (3) below, for example.

(1) The determination is performed once, and the infrared transmission image 220 is compared with the reference data for infrared transmission image, and it is determined whether or not an abnormal region outside the allowable range with respect to the reference data for infrared transmission image exists in the infrared transmission image 220.

(2) When an abnormal region exists in the infrared transmission image 220, a secondary determination is performed, in which a determination target region, which is a region corresponding to the abnormal region in the infrared reflection image 210, is compared with the infrared reflection image reference data, and a determination is made as to whether or not the determination target region is within an allowable range with respect to the infrared reflection image reference data.

(3) When it is determined by the secondary determination that the determination target region is within the allowable range with respect to the infrared-reflected-image reference data, the authenticity determination process is performed on the region excluding at least the abnormal region after the infrared transmission image 220 is removed.

The embodiments of the present invention have been described above with reference to the drawings, but the present invention is not limited to the above embodiments. The configurations of the respective embodiments may be appropriately combined or modified within a range not departing from the gist of the present invention.

INDUSTRIAL APPLICABILITY

As described above, the present invention is a technique for reducing erroneous judgment of a worn and dirty banknote as a counterfeit banknote in a banknote recognition apparatus and a banknote recognition method.