JP2021163231A - Paper sheet identification device and paper sheet identification method - Google Patents

Abstract

To provide a paper sheet identification device capable of correctly identifying specific numbers even if the specific numbers using different font patterns are printed on paper sheets.SOLUTION: A paper sheet identification device 1 includes: an identification unit 12 that identifies a kind of a paper sheet; a determination unit 131 that determines a photo pattern of the paper sheet whose kind is identified by the identification unit 12 from at least one photo pattern stored by a storage unit 14 on the basis of a second region set outside a first region including at least a part of a specific number in an image obtained by imaging the paper sheet; and a character detection unit 132 that reads a template associated with the photo pattern determined by the determination unit 131 from the storage unit 14 to detect a plurality of characters included in the specific number.SELECTED DRAWING: Figure 1

Description

The present invention relates to a paper leaf identification device and a paper leaf species identification method.

Patent Document 1 discloses a paper leaf processing apparatus that reads a unique number (also referred to as a serial number) printed on paper leaves such as banknotes, securities, and gift certificates to determine the authenticity of the paper leaves. ing.

Japanese Unexamined Patent Publication No. 2017-004153

A conventional paper leaf processing apparatus such as Patent Document 1 is based on the premise that paper sheets of the same type are printed with a unique number in the same font pattern. Therefore, when unique numbers are printed in different font patterns, those unique numbers may not be read correctly.

Some aspects of the present invention are a paper leaf identification device and a paper leaf identification device capable of correctly identifying the unique numbers even if the unique numbers are printed using different font patterns. Provide a method.

The paper leaf identification device according to one aspect of the present invention has at least one font pattern indicating how the unique number of the paper leaf is printed on the paper leaf, and the unique number in at least one font pattern. A storage unit that stores in advance for each type of paper leaf, an identification unit that identifies the type of paper leaf, and an image of the paper leaf are imaged in association with a template that indicates the position where a plurality of contained characters are described. In the image, the identification unit identifies the type from at least one font pattern stored by the storage unit based on the second area set outside the first area including at least a part of the unique number. A determination unit that determines the font pattern of paper sheets, a character detection unit that reads a template associated with the font pattern determined by the determination unit from the storage unit, and detects a plurality of characters included in the unique number. To be equipped.

According to one aspect of the present invention, even paper sheets on which unique numbers are printed using different font patterns can be correctly identified.

The block diagram which shows the structure of the paper leaf identification apparatus which concerns on 1st Embodiment. The figure which shows an example of the banknote handled by the paper leaf identification apparatus which concerns on 1st Embodiment. The figure which shows an example of the font pattern data which the storage part which concerns on 1st Embodiment stores. The figure which shows an example of the template data which the storage part which concerns on 1st Embodiment stores. The figure which shows an example of the relationship between the serial number which concerns on 1st Embodiment, and a template. The flowchart which shows the processing of the paper leaf identification apparatus which concerns on 1st Embodiment. The flowchart which shows the detail of the font pattern determination process which concerns on 1st Embodiment. The figure which shows an example of the process of FIG. The figure which shows an example of the process of FIG. The figure which shows an example of the banknote handled by the paper leaf identification apparatus which concerns on 2nd Embodiment. The figure which shows an example of the font pattern data which the storage part which concerns on 2nd Embodiment stores. The figure which shows an example of the template data which the storage part which concerns on 2nd Embodiment stores. The figure which shows another example of the relationship between the serial number which concerns on 2nd Embodiment, and a template. The flowchart which shows the detail of the font pattern determination process which concerns on 2nd Embodiment. The figure which shows an example of the process of FIG. The figure which shows an example of the process of FIG. The figure which shows an example of the banknote handled by the paper leaf identification apparatus which concerns on 3rd Embodiment. The figure which shows an example of the font pattern data which the storage part which concerns on 3rd Embodiment stores. The figure which shows an example of the template data which the storage part which concerns on 3rd Embodiment stores. The figure which shows another example of the relationship between the serial number which concerns on 3rd Embodiment, and a template. The flowchart which shows the detail of the font pattern determination process which concerns on 3rd Embodiment. The figure which shows an example of the process of FIG. The figure which shows an example of the process of FIG. The figure which shows an example of the banknote handled by the paper leaf identification apparatus which concerns on 4th Embodiment. The figure which shows an example of the font pattern data which the storage part which concerns on 4th Embodiment stores. The figure which shows an example of the template data which the storage part which concerns on 4th Embodiment stores. The figure which shows another example of the relationship between the serial number which concerns on 4th Embodiment, and a template. The flowchart which shows the detail of the font pattern determination process which concerns on 4th Embodiment. The figure which shows an example of the process of FIG. 28. The figure which shows an example of the process of FIG. 28. The figure which shows an example of the serial number and font pattern determination processing which concerns on the 1st modification. The figure which shows an example of the serial number and font pattern determination processing which concerns on the 2nd modification. The figure which shows an example of the serial number and font pattern determination processing which concerns on 3rd modification.

Hereinafter, the first to fourth embodiments and the like will be described in detail with reference to the drawings.
[First Embodiment]
FIG. 1 is a block diagram showing a configuration of a paper leaf identification device 1 according to the first embodiment. The paper leaf identification device 1 identifies the type of paper leaf from an image of the paper leaf such as banknotes, securities, gift certificates, etc., and reads a unique number printed on a predetermined area of the paper leaf. Here, a case where the paper leaf identification device 1 reads a serial number which is a unique number printed on a predetermined area of a banknote (for example, an area on the upper right of the banknote) will be described. Here, the case where the paper leaves are banknotes will be described, but the present invention is not limited to this, and the paper leaves are of the same type (for example, paper leaves of the same size). The disclosure contents of the present application can be applied to paper sheets on which unique numbers are printed in different font patterns.

The paper leaf identification device 1 includes a control unit 10, an image acquisition unit 11, a denomination identification unit 12, a serial number identification unit 13, and a storage unit 14. The control unit 10 is a CPU (Central Processing Unit) or the like, and controls an image acquisition unit 11, a denomination identification unit 12, a serial number identification unit 13, and a storage unit 14 that constitute the paper leaf identification device 1. The image acquisition unit 11 is a camera, an image sensor, or the like, and captures an image of the entire banknote to be identified. The denomination identification unit 12 identifies the denomination of the banknote for which the serial number is to be identified.

2 (a) to 2 (c) are diagrams showing an example of banknotes handled by the paper leaf identification device 1. FIG. 2A shows a banknote B11 having a banknote value of 100 currency units. In the central area of the bill B11, the number B12 "100" indicating that the bill B11 has a value of 100 currency units is printed. A serial number B14 for making the bill B11 unique is printed in two areas on the upper right and the lower left of the bill B11. The serial number B14 is composed of a combination of alphanumeric characters, and here, the serial number B14 of the bill B11 shows a case where the serial number B14 includes 9 characters of "1AB234567". In the bill B11, the size of each of the plurality of characters included in the serial number B14 does not change in the lateral direction of the bill B11 and is constant.

FIG. 2B shows a banknote B21 having a banknote value of 500 currency units. In the central area of the bill B21, the number B22 indicating that the bill B21 has a value of 500 currency units is printed. The serial number B24 printed on the upper right and lower left of the bill B21 is composed of a combination of alphanumeric characters, and here, the serial number B24 of the bill B21 includes a case where nine characters of "1AB234567" are included. In the bill B21, the size of each of the plurality of characters included in the serial number B24 does not change in the lateral direction of the bill B21 and is constant. Note that the banknote B11 shown in FIG. 2A and the banknote B21 shown in FIG. 2B are different denominations because the banknote values are different.

FIG. 2C shows a banknote B31 having a banknote value of 100 currency units. In the central area of the bill B31, the number B32 "100" indicating that the bill B31 has a value of 100 currency units is printed. Here, the serial number B34 of the bill B31 includes nine characters of "1AB234567". In the bill B31, the line connecting the lower ends of the plurality of characters included in the serial number B34 is parallel to the long side of the bill B31, but the size of each of the plurality of characters included in the serial number B34 is It increases from the left end (that is, "1") to the right end (that is, "7"). Note that the banknote B11 shown in FIG. 2A and the banknote B31 shown in FIG. 2C have the same banknote value and therefore have the same denomination.

Returning to FIG. 1, the denomination identification unit 12 identifies the denomination of the banknote imaged by the image acquisition unit 11 based on the image of the central region of the entire image of the banknote captured by the image acquisition unit 11, for example. do. For example, when the denomination identification unit 12 detects that the character "100" is printed in the central portion of the entire image of the banknote captured by the image acquisition unit 11, the denomination identification unit 12 Determines that the banknote is a banknote B11 (see FIG. 2A) or a banknote B31 (see FIG. 2C) having a value of 100 currency units. On the other hand, when the denomination identification unit 12 detects that the character "500" is printed in the central portion of the entire image of the banknote captured by the image acquisition unit 11, the denomination identification unit 12 may perform the denomination identification unit 12. It is determined that the banknote is a banknote B21 (see FIG. 2B) having a value of 500 currency units.

The serial number identification unit 13 of FIG. 1 includes a determination unit 131 and a character detection unit 132. When the serial number of the banknote imaged by the image acquisition unit 11 is a banknote having a plurality of font patterns, the determination unit 131 determines which font pattern the serial number of the banknote is among the plurality of font patterns. To judge. The character detection unit 132 detects each character included in the serial number of the banknote captured by the image acquisition unit 11 by using the template corresponding to the font pattern determined by the determination unit 131.

The storage unit 14 of FIG. 1 is a memory or the like, and stores various data used in the processing of the paper leaf identification device 1. For example, the storage unit 14 stores font pattern data D11, template data D12, character recognition data D13, and the like.

FIG. 3 is a diagram showing an example of font pattern data D11 stored in the storage unit 14 according to the first embodiment. In the font pattern data D11, the denomination and the font pattern are associated and stored in advance. FIG. 3 shows a case where the denomination C11 and the font pattern F11 are associated with each other. Further, FIG. 3 shows a case where the denomination C12 and the font patterns F11 and F12 are associated with each other.

Here, the denomination C11 is the denomination of the banknote B21 having a value of 500 currency units (see FIG. 2B). The denomination C12 is a denomination of banknotes B11 and B31 having a value of 100 currency units (see FIGS. 2 (a) and 2 (c)). In the font pattern F11, the size of each of the plurality of characters included in the serial number B14 does not change in the lateral direction of the bill B11 and is constant. In the font pattern F12, the line connecting the lower ends of the plurality of characters included in the serial number B34 is parallel to the long side of the bill B31, but the size of each of the plurality of characters included in the serial number B34 is the left end. It is assumed that the size increases from (that is, "1") to the right end (that is, "7").

FIG. 4 is a diagram showing an example of template data D12 stored in the storage unit 14 according to the first embodiment. In the template data D12, the font pattern and the template are stored in association with each other. FIG. 4 shows a case where the font pattern F11 and the template T11 are associated with each other. Further, FIG. 4 shows a case where the font pattern F12 and the template T12 are associated with each other.

5 (a) and 5 (b) are diagrams showing an example of the relationship between the serial number according to the first embodiment and the template. FIG. 5A shows a serial number B14 printed in the upper right area of the banknote B11 of FIG. 2A. The serial number B14 contains a plurality of characters (that is, 1, A, B, 2, 3, 4, 5, 6, 7), and a frame in which each of the plurality of characters is included internally. t11, t12, t13, t14, t15, t16, t17, t18, t19 are preset. The template T11 is configured by the frames t11 to t19. In the font pattern F11, for example, when the area set to surround the outer circumference of the template T11 is set as the standard serial number area, the serial number is not detected in the area outside the standard serial number area. Is determined to be the font pattern F11.

FIG. 5B shows the serial number B34 printed in the upper right area of the banknote B31 of FIG. 2C. The serial number B34 contains a plurality of characters (that is, 1, A, B, 2, 3, 4, 5, 6, 7), and a frame in which each of the plurality of characters is included internally. t21, t22, t23, t24, t25, t26, t27, t28, t29 are preset. The template T12 is composed of the frames t21 to t29. The font pattern F12 is determined to be the font pattern F12, for example, when a serial number is detected in the upper right area outside the standard serial number area.

The character recognition data D13 (not shown) stored in the storage unit 14 of FIG. 1 is included in the serial numbers B14, B24, and B34 of the bills B11, B21, and B31 (see FIGS. 2A to 2C). It is data used to recognize each of a plurality of characters. For example, in the character detection unit 132 of FIG. 1, the character recognition data D13 stores which character is included in each of the frames t11 to t19 and t21 to t29 of FIGS. 5 (a) and 5 (b). By collating with the image data of the font of each character to be used, each character included in the serial numbers B14, B24, and B34 is identified.

FIG. 6 is a flowchart showing the processing of the paper leaf identification device 1 according to the first embodiment. First, the image acquisition unit 11 acquires an image of the banknote by imaging the banknote to be identified (step S11).

Next, the denomination identification unit 12 identifies the denomination of the banknote based on the image acquired in step S11 (step S12). For example, the denomination identification unit 12 identifies the denomination by identifying the number printed in the center of the acquired image. For example, when the number 100 is printed in the central portion of the banknote, the denomination identification unit 12 determines that the denomination of the banknote whose image is captured is a banknote whose currency value is 100 currency units (that is, FIG. 2 (that is, FIG. 2). It is determined that the denomination is the bill B11 of a) or the denomination C12 of the bill B31) of FIG. 2C. On the other hand, when the number 500 is printed in the central portion of the banknote, the denomination identification unit 12 determines that the denomination of the banknote whose image is captured is a banknote having a currency value of 500 currency units (that is, FIG. 2 (that is, FIG. 2). It is determined that the denomination is C11 of the banknote B21) of b).

Next, the determination unit 131 determines whether or not a plurality of font patterns exist in the banknote whose denomination is identified in step S12 based on the font pattern data D11 (see FIG. 3) stored in the storage unit 14. (Step S13). For example, when the denomination of the banknote whose image is captured in step S11 is the denomination C11, the determination unit 131 refers to the font pattern data D11, so that the denomination C11 has a plurality of font patterns. Instead, it is determined that only one font pattern F11 exists. That is, when the denomination of the banknote for which the image is captured is the denomination C11, the determination unit 131 determines that the font pattern is the font pattern F11. On the other hand, when the denomination of the banknote for which the image is captured is the denomination C12, the determination unit 131 refers to the font pattern data D11, so that the denomination C12 has a plurality of font patterns F11 and F12. Then it is determined.

If the determination unit 131 determines in step S13 that the banknote for which the denomination has been identified does not have a plurality of font patterns (NO in step S13), the process of step S15, which will be described later, is performed. On the other hand, if the determination unit 131 determines that the banknote for which the denomination has been identified in step S13 has a plurality of font patterns (YES in step S13), the determination unit 131 has taken an image in step S11. The font pattern of the serial number of the banknote is determined (step S14). The process of step S14 will be described later with reference to FIG. 7.

Next, the character detection unit 132 reads the template corresponding to the determined font pattern from the template data D12 (see FIG. 4) stored in the storage unit 14 (step S15). For example, when NO is determined in step S13 (that is, when the font pattern of denomination C11 is determined to be font pattern F11), or in step S14, the font pattern of denomination C12 is a font pattern. If it is determined to be F11, the character detection unit 132 reads the template T11 (see FIG. 5A) by referring to the template data D12 (see FIG. 4). On the other hand, when it is determined in step S14 that the font pattern of the denomination C12 is the font pattern F12, the character detection unit 132 refers to the template data D12 (see FIG. 4) to obtain the template T12 (see FIG. 4). (See FIG. 5B) is read.

After that, the character detection unit 132 recognizes each of the plurality of characters included in the serial number of the banknote for which the image was acquired in step S11 by using the template read in step S15 (step S16), and the flowchart of FIG. 6 is shown. End the process. For example, since the serial number B14 of the banknote for which the image was acquired in step S11 is the font pattern F11, the positions of a plurality of characters included in the serial number B14 are specified by the template T11 (see FIG. 5A). , Each of the plurality of characters included in the serial number B14 is identified by collating with the character recognition data D13. On the other hand, since the serial number B34 of the bill whose image was acquired in step S11 is the font pattern F12, the positions of a plurality of characters included in the serial number B34 are specified by the template T12 (see FIG. 5B). , Each of the plurality of characters included in the serial number B34 is identified by collating with the character recognition data D13.

FIG. 7 is a flowchart showing the font pattern determination process of step S14 of FIG. 6 and showing the details of the font pattern determination process according to the first embodiment. First, the determination unit 131 acquires an image of a predetermined area on which the serial number is printed out of the entire image of the banknote acquired by the image acquisition unit 11 in step S11 of FIG. 6 (step S21). For example, when the banknote acquired by the image acquisition unit 11 in step S11 of FIG. 6 is a banknote B11 (see FIG. 2A), in step S21 of FIG. 7, as shown in FIG. 8A. , The image G11 including the serial number B14 is acquired. On the other hand, when the banknote acquired by the image acquisition unit 11 in step S11 of FIG. 6 is a banknote B31 (see FIG. 2C), in step S21 of FIG. 7, as shown in FIG. 9A. , The image G21 including the serial number B34 is acquired.

Here, the case where only the image of the upper right region of the entire image of the banknote acquired by the image acquisition unit 11 is acquired in step S21 is not limited to this, and the image of the entire image of the banknote is not limited to this. The image of the upper right area and the lower left area may be acquired, or only the image of the lower left area of the entire image of the banknote may be acquired. Further, in order to reduce the possibility that the serial number is read incorrectly or the font pattern is determined incorrectly, the background such as the pattern printed in the area where the serial number is printed is removed by image processing. You may do so.

Next, the determination unit 131 scans the image acquired in step S21 in the lateral direction and searches for the center position of the serial number from the pixel distribution corresponding to the serial number (step S22). For example, when the image acquired in step S21 of FIG. 7 is that of banknote B11 (see FIG. 2 (a)), in step S22 of FIG. 7, as shown in FIG. 8 (b), the image G11 The central position L11 in is searched for. On the other hand, when the image acquired in step S21 of FIG. 7 is that of the bill B31 (see FIG. 2C), in step S22 of FIG. 7, as shown in FIG. 9B, the image G21 The central position L11 in is searched for.

Next, the determination unit 131 sets a first region, which is a standard serial number region, based on the central position L11 of the serial number searched in step S22 (step S24). For example, when the banknote acquired in step S21 of FIG. 7 is that of banknote B11 (see FIG. 2A), in step S24 of FIG. 7, it has a rectangular shape as shown in FIG. 8C. The first region R11 of is set. On the other hand, when the banknote acquired in step S21 of FIG. 7 is that of banknote B31 (see FIG. 2C), in step S24 of FIG. 7, it has a rectangular shape as shown in FIG. 9C. The first region R11 of is set.

Next, the determination unit 131 sets a second area for determining the font pattern outside the upper right of the first area set in step S24 (step S25). For example, when the image acquired in step S21 of FIG. 7 is that of banknote B11 (see FIG. 2 (a)), in step S25 of FIG. 7, as shown in FIG. 8 (d), the first A second region R12 is set outside the upper right corner of the region R11. On the other hand, when the banknote acquired in step S21 of FIG. 7 is that of banknote B31 (see FIG. 2C), in step S25 of FIG. 7, as shown in FIG. 9D, the first A second region R12 is set outside the upper right corner of the region R11.

Next, the determination unit 131 determines whether or not the serial number is detected in the second region R12 set in step S25 based on the pixel distribution corresponding to the serial number (step S26). For example, in the determination unit 131, when the area of the pixel corresponding to the serial number is present in a predetermined area (for example, 1% of the area of the second region R12) in the second region R12, the serial number is changed. It can be determined that it has been detected.

When the banknote acquired in step S21 of FIG. 7 is that of banknote B11 (see FIG. 2A), as shown in FIG. 8D, the pixels corresponding to the serial numbers in the second region R12. Does not exist in excess of a predetermined value, so the determination unit 131 determines that the serial number is not detected in the second region R12 (NO in step S26 of FIG. 7). Then, the determination unit 131 determines that the font pattern of the serial number of the banknote for which the image was acquired in step S21 of FIG. 7 (that is, the banknote for which the image was acquired in step S11 of FIG. 6) is the font pattern F11 (that is, the font pattern is the font pattern F11). Step S27). After that, the process of step S15 of FIG. 6 is performed.

On the other hand, when the image acquired in step S21 of FIG. 7 is that of banknote B31 (see FIG. 2C), as shown in FIG. 9D, the second region R12 corresponds to the serial number. Since the number of pixels to be used is equal to or greater than a predetermined value, the determination unit 131 determines that the serial number has been detected in the second region R12 (YES in step S26 of FIG. 7). Then, it is determined that the font pattern of the serial number of the banknote for which the image was acquired in step S21 of FIG. 7 (that is, the banknote for which the image was acquired in step S11 of FIG. 6) is the font pattern F12 (step S28). After that, the process of step S15 of FIG. 6 is performed.

According to the first embodiment described above, the serial numbers of banknotes of the same denomination are the heights of a plurality of characters included in the first font pattern F11 (that is, as shown in FIG. 5A). The second font pattern F12 (that is, as shown in FIG. 5B, the heights of the plurality of characters included in the serial number B34 increase toward the end). However, it is possible to correctly determine whether the serial number of the banknote is the first font pattern F11 or the second font pattern F12.

[Second Embodiment]
Next, the second embodiment will be described. The description of the portion where the configuration and processing of the paper leaf identification device 1 in the second embodiment is the same as the configuration and processing of the paper leaf identification device 1 according to the first embodiment will be omitted.

FIG. 10 is a diagram showing an example of banknotes handled by the paper leaf identification device according to the second embodiment. In the first embodiment, a case where the banknote B11 (see FIG. 2A) and the banknote B31 (see FIG. 2C) are distinguished has been described. However, in the second embodiment, a case where the serial numbers of the banknote B11 (see FIG. 2A) and the banknote B41 (see FIG. 10) are read will be described.

FIG. 10 shows a banknote B41 having a banknote value of 100 currency units. In the central area of the bill B41, the number B42 of "100" indicating that the bill B41 has a value of 100 currency units is printed. Here, the serial number B44 of the bill B31 includes nine characters of "1AB234567". In the bill B41, the line connecting the upper ends of the plurality of characters included in the serial number B44 is parallel to the long side of the bill B41, but the size of each of the plurality of characters included in the serial number B44 is It increases from the left end (that is, "1") to the right end (that is, "7"). Note that the banknote B11 shown in FIG. 2A and the banknote B41 shown in FIG. 10 have the same banknote value and therefore have the same denomination.

FIG. 11 is a diagram showing an example of font pattern data D21 stored in the storage unit 14 according to the second embodiment. FIG. 11 shows a case where the denomination C11 and the font pattern F11 are associated with each other, and the denomination C12 and the font patterns F11 and F13 are associated with each other. In the font pattern F13, the line connecting the upper ends of the plurality of characters included in the serial number B44 is parallel to the long side of the bill B41, but the size of each of the plurality of characters included in the serial number B34 is the left end. It increases from (that is, "1") to the right end (that is, "7").

FIG. 12 is a diagram showing an example of template data D22 stored in the storage unit 14 according to the second embodiment. FIG. 12 shows a case where the font pattern F11 and the template T11 are associated with each other, and the font pattern F13 and the template T13 are associated with each other.

FIG. 13 is a diagram showing an example of the relationship between the serial number according to the second embodiment and the template. FIG. 13 shows the serial number B44 printed in the upper right and lower left areas of the banknote B41 of FIG. The serial number B44 contains a plurality of characters (that is, 1, A, B, 2, 3, 4, 5, 6, 7), and a frame in which each of the plurality of characters is included internally. t31, t32, t33, t34, t35, t36, t37, t38, t39 are preset. The template T13 is composed of the frames t31 to t39. In the font pattern F13, for example, when the area set to surround the outer circumference of the template T11 is set as the standard serial number area, the serial number is detected in the lower right area outside the standard serial number area. If so, it is determined that the font pattern is F13.

FIG. 14 is a flowchart showing the font pattern determination process of step S14 of FIG. 6 and showing the details of the font pattern determination process according to the second embodiment. First, the determination unit 131 acquires an image of a predetermined area on which the serial number is printed out of the entire image of the banknote acquired by the image acquisition unit 11 in step S11 of FIG. 6 (step S31). For example, when the banknote acquired by the image acquisition unit 11 in step S11 of FIG. 6 is a banknote B11 (see FIG. 2A), in step S31 of FIG. 14, as shown in FIG. 15A. , The image G31 including the serial number B14 is acquired. On the other hand, when the banknote acquired by the image acquisition unit 11 in step S11 of FIG. 6 is a banknote B41 (see FIG. 10), in step S31 of FIG. 14, the serial number as shown in FIG. 16A. Image G41 including B44 is acquired.

Next, the determination unit 131 scans the image acquired in step S31 in the lateral direction and searches for the center position of the serial number from the pixel distribution corresponding to the serial number (step S32). For example, when the image acquired in step S31 of FIG. 14 is that of banknote B11 (see FIG. 2 (a)), in step S32 of FIG. 14, as shown in FIG. 15 (b), the image G31 The central position L21 in is searched. On the other hand, when the image acquired in step S31 of FIG. 14 is that of the bill B41 (see FIG. 10), in step S32 of FIG. 14, as shown in FIG. 16 (b), the central position in the image G41. L21 is searched.

Next, the determination unit 131 sets a first region, which is a standard serial number region, based on the central position L21 of the serial number searched in step S32 (step S34). For example, when the banknote acquired in step S31 of FIG. 14 is that of banknote B11 (see FIG. 2A), in step S34 of FIG. 14, it has a rectangular shape as shown in FIG. 15C. The first region R21 of is set. On the other hand, when the banknote acquired in step S31 of FIG. 14 is that of banknote B41 (see FIG. 10), in step S34 of FIG. 14, the first rectangular shape as shown in FIG. 16C. The area R21 is set.

Next, the determination unit 131 sets a second area for determining the font pattern outside the lower right of the first area set in step S34 (step S35). For example, when the image acquired in step S31 of FIG. 14 is that of banknote B11 (see FIG. 2 (a)), in step S35 of FIG. 14, as shown in FIG. 15 (d), the first A second region R22 is set outside the lower right corner of the region R21. On the other hand, when the banknote acquired in step S31 of FIG. 14 is that of banknote B41 (see FIG. 10), in step S35 of FIG. 14, as shown in FIG. 16D, the first region R21 The second region R22 is set outside the lower right corner.

Next, the determination unit 131 determines whether or not the serial number is detected in the second region R22 set in step S35 based on the pixel distribution corresponding to the serial number (step S36). For example, in the determination unit 131, when the area of the pixel corresponding to the serial number exists in a predetermined area (for example, 1% of the area of the second region R22) in the second region R22, the serial number is changed. It can be determined that it has been detected.

When the banknote acquired in step S31 of FIG. 14 is that of banknote B11 (see FIG. 2A), as shown in FIG. 15D, the pixels corresponding to the serial numbers in the second region R22. Does not exist in excess of a predetermined value, so the determination unit 131 determines that the serial number is not detected in the second region R22 (NO in step S36 of FIG. 14). Then, the determination unit 131 determines that the font pattern of the serial number of the banknote for which the image was acquired in step S31 of FIG. 14 (that is, the banknote for which the image was acquired in step S11 of FIG. 6) is the font pattern F11 (that is, the font pattern is the font pattern F11). Step S37). After that, the process of step S15 of FIG. 6 is performed.

On the other hand, when the image acquired in step S31 of FIG. 14 is that of the bill B41 (see FIG. 10), as shown in FIG. 16 (d), the pixels corresponding to the serial numbers are provided in the second region R22. Since there are more than a predetermined value, the determination unit 131 determines that the serial number has been detected in the second region R22 (YES in step S36 of FIG. 14). Then, it is determined that the font pattern of the serial number of the banknote for which the image was acquired in step S31 of FIG. 14 (that is, the banknote for which the image was acquired in step S11 of FIG. 6) is the font pattern F13 (step S38). After that, the process of step S15 of FIG. 6 is performed.

According to the second embodiment described above, the serial numbers of banknotes of the same denomination are the heights of a plurality of characters included in the first font pattern F11 (that is, as shown in FIG. 5A). Even if it is the second font pattern F13 (that is, as shown in FIG. 13, the heights of a plurality of characters included in the serial number B44 increase toward the end), the banknotes It is possible to correctly determine which of the first font pattern F11 and the second font pattern F13 is the serial number of.

[Third Embodiment]
Next, the third embodiment will be described. The description of the portion where the configuration and processing of the paper leaf identification device 1 in the third embodiment is the same as the configuration and processing of the paper leaf identification device 1 according to the first embodiment will be omitted.

FIG. 17 is a diagram showing an example of banknotes handled by the paper leaf identification device according to the third embodiment. In the first embodiment, a case where the banknote B11 (see FIG. 2A) and the banknote B31 (see FIG. 2C) are distinguished has been described. However, in the third embodiment, a case where the serial numbers of the banknote B11 (see FIG. 2A) and the banknote B51 (see FIG. 17) are read will be described.

FIG. 17 shows a banknote B51 having a banknote value of 100 currency units. In the central area of the bill B51, the number B52 "100" indicating that the bill B51 has a value of 100 currency units is printed. Here, the serial number B54 of the bill B51 includes nine characters of "1AB234567". In the bill B51, the line connecting the upper ends of the plurality of characters included in the serial number B54 is parallel to the long side of the bill B51, but the size of each of the plurality of characters included in the serial number B54 is It becomes smaller from the left end (that is, "1") to the right end (that is, "7"). Note that the banknote B11 shown in FIG. 2A and the banknote B51 shown in FIG. 17 have the same banknote value and therefore have the same denomination.

FIG. 18 is a diagram showing an example of font pattern data D31 stored in the storage unit 14 according to the third embodiment. FIG. 18 shows a case where the denomination C11 and the font pattern F11 are associated with each other, and the denomination C12 and the font patterns F11 and F14 are associated with each other. In the font pattern F14, the line connecting the upper ends of the plurality of characters included in the serial number B54 is parallel to the long side of the bill B51, but the size of each of the plurality of characters included in the serial number B54 is the left end. It becomes smaller from (that is, "1") to the right end (that is, "7").

FIG. 19 is a diagram showing an example of template data D32 stored in the storage unit 14 according to the third embodiment. FIG. 19 shows a case where the font pattern F11 and the template T11 are associated with each other, and the font pattern F14 and the template T14 are associated with each other.

FIG. 20 is a diagram showing an example of the relationship between the serial number according to the third embodiment and the template. In FIG. 20, the serial number B54 printed in the upper right and lower left areas of the bill B51 of FIG. 17 is shown. The serial number B54 contains a plurality of characters (that is, 1, A, B, 2, 3, 4, 5, 6, 7), and a frame in which each of the plurality of characters is included internally. t41, t42, t43, t44, t45, t46, t47, t48, t49 are preset. The template T14 is composed of the frames t41 to t49. In the font pattern F14, for example, when the area set to surround the outer circumference of the template T11 is set as the standard serial number area, the serial number is detected in the lower left area outside the standard serial number area. In this case, it is determined that the font pattern is F14.

FIG. 21 is a flowchart showing the font pattern determination process of step S14 of FIG. 6 and showing the details of the font pattern determination process according to the third embodiment. First, the determination unit 131 acquires an image of a predetermined area on which the serial number is printed out of the entire image of the banknote acquired by the image acquisition unit 11 in step S11 of FIG. 6 (step S41). For example, when the banknote acquired by the image acquisition unit 11 in step S11 of FIG. 6 is a banknote B11 (see FIG. 2A), in step S41 of FIG. 21, as shown in FIG. 22A. , The image G51 including the serial number B14 is acquired. On the other hand, when the banknote acquired by the image acquisition unit 11 in step S11 of FIG. 6 is a banknote B51 (see FIG. 17), in step S41 of FIG. 21, the serial number as shown in FIG. 23 (a). Image G61 including B54 is acquired.

Next, the determination unit 131 scans the image acquired in step S41 in the lateral direction and searches for the center position of the serial number from the pixel distribution corresponding to the serial number (step S42). For example, when the image acquired in step S41 of FIG. 21 is that of banknote B11 (see FIG. 2 (a)), in step S42 of FIG. 21, as shown in FIG. 22 (b), the image G51 The central position L31 in is searched. On the other hand, when the image acquired in step S41 of FIG. 21 is that of the bill B51 (see FIG. 17), in step S42 of FIG. 21, as shown in FIG. 23 (b), the central position in the image G61. L31 is searched.

Next, the determination unit 131 sets a first region, which is a standard serial number region, based on the central position L31 of the serial number searched in step S42 (step S44). For example, when the banknote acquired in step S41 of FIG. 21 is that of banknote B11 (see FIG. 2 (a)), in step S44 of FIG. 21, it has a rectangular shape as shown in FIG. 22 (c). The first region R31 of is set. On the other hand, when the banknote acquired in step S41 of FIG. 21 is that of banknote B51 (see FIG. 17), in step S44 of FIG. 21, the first rectangular shape as shown in FIG. 23 (c). The area R31 is set.

Next, the determination unit 131 sets a second area for determining the font pattern outside the lower left of the first area set in step S44 (step S45). For example, when the image acquired in step S41 of FIG. 21 is that of banknote B11 (see FIG. 2 (a)), in step S45 of FIG. 21, as shown in FIG. 22 (d), the first A second region R32 is set outside the lower left of the region R31. On the other hand, when the banknote acquired in step S41 of FIG. 21 is that of banknote B51 (see FIG. 17), in step S45 of FIG. 21, as shown in FIG. 23 (d), the first region R31 The second region R32 is set outside the lower left.

Next, the determination unit 131 determines whether or not the serial number is detected in the second region R32 set in step S45 based on the pixel distribution corresponding to the serial number (step S46). For example, in the determination unit 131, when the area of the pixel corresponding to the serial number exists in a predetermined area (for example, 1% of the area of the second region R32) in the second region R32, the serial number is changed. It can be determined that it has been detected.

When the banknote acquired in step S41 of FIG. 21 is that of banknote B11 (see FIG. 2A), as shown in FIG. 22D, the pixels corresponding to the serial numbers in the second region R32. Does not exist in excess of a predetermined value, so that the determination unit 131 determines that the serial number is not detected in the second region R32 (NO in step S46 of FIG. 21). Then, the determination unit 131 determines that the font pattern of the serial number of the banknote for which the image was acquired in step S41 of FIG. 21 (that is, the banknote for which the image was acquired in step S11 of FIG. 6) is the font pattern F11 (that is, the font pattern is the font pattern F11). Step S47). After that, the process of step S15 of FIG. 6 is performed.

On the other hand, when the image acquired in step S41 of FIG. 21 is that of the bill B51 (see FIG. 17), as shown in FIG. 23 (d), the pixels corresponding to the serial numbers are provided in the second region R32. Since there are more than a predetermined value, the determination unit 131 determines that the serial number has been detected in the second region R32 (YES in step S46 of FIG. 21). Then, it is determined that the font pattern of the serial number of the banknote for which the image was acquired in step S41 of FIG. 21 (that is, the banknote for which the image was acquired in step S11 of FIG. 6) is the font pattern F14 (step S48). After that, the process of step S15 of FIG. 6 is performed.

According to the third embodiment described above, the serial numbers of banknotes of the same denomination are the heights of a plurality of characters included in the first font pattern F11 (that is, as shown in FIG. 5A). Even if it is the second font pattern F14 (that is, as shown in FIG. 20, the heights of a plurality of characters included in the serial number B54 become smaller toward the end), the banknotes It is possible to correctly determine which of the first font pattern F11 and the second font pattern F14 is the serial number of.

[Fourth Embodiment]
Next, the fourth embodiment will be described. The description of the portion where the configuration and processing of the paper leaf identification device 1 in the fourth embodiment is the same as the configuration and processing of the paper leaf identification device 1 according to the first embodiment will be omitted.

FIG. 24 is a diagram showing an example of banknotes handled by the paper leaf identification device according to the fourth embodiment. In the first embodiment, a case where the banknote B11 (see FIG. 2A) and the banknote B31 (see FIG. 2C) are distinguished has been described. However, in the fourth embodiment, a case where the serial numbers of the banknote B11 (see FIG. 2A) and the banknote B61 (see FIG. 24) are read will be described.

FIG. 24 shows a banknote B61 having a banknote value of 100 currency units. In the central area of the banknote B61, the number B62 "100" is printed to indicate that the banknote B61 has a value of 100 currency units. Here, the serial number B64 of the bill B61 includes nine characters of "1AB234567". In the bill B61, the line connecting the vertical centers of the plurality of characters included in the serial number B64 is parallel to the long side of the bill B61, but the size of the plurality of characters included in the serial number B64. S, increases from the left edge (ie, "1") to the center (ie, "3") and decreases from the center (ie, "3") to the right edge (ie, "7"). .. Since the bill B11 shown in FIG. 2A and the bill B61 shown in FIG. 24 have the same bill value, they have the same denomination.

FIG. 25 is a diagram showing an example of font pattern data D41 stored in the storage unit 14 according to the fourth embodiment. FIG. 25 shows a case where the denomination C11 and the font pattern F11 are associated with each other, and the denomination C12 and the font patterns F11 and F15 are associated with each other. In the font pattern F15, the line connecting the vertical centers of the plurality of characters included in the serial number B64 is parallel to the long side of the bill B61, but the size of each of the plurality of characters included in the serial number B64 is large. It increases from the left edge (that is, "1") to the center (that is, "3") and decreases from the center (that is, "3") to the right edge (that is, "7").

FIG. 26 is a diagram showing an example of template data D42 stored in the storage unit 14 according to the fourth embodiment. FIG. 26 shows a case where the font pattern F11 and the template T11 are associated with each other, and the font pattern F15 and the template T15 are associated with each other.

FIG. 27 is a diagram showing an example of the relationship between the serial number according to the fourth embodiment and the template. FIG. 27 shows the serial number B64 printed in the upper right and lower left areas of the banknote B61 of FIG. 24. The serial number B64 contains a plurality of characters (that is, 1, A, B, 2, 3, 4, 5, 6, 7), and a frame in which each of the plurality of characters is included internally. t51, t52, t53, t54, t55, t56, t57, t58, t59 are preset. The template T15 is configured by the frames t51 to t59. The font pattern F15 is, for example, an area above and below the center in the left-right direction outside the standard numbering area when the area set to surround the outer circumference of the template T11 is used as the standard numbering area. Is detected, it is determined that the font pattern is F15.

FIG. 28 is a flowchart showing the font pattern determination process of step S14 of FIG. 6 and showing the details of the font pattern determination process according to the fourth embodiment. First, the determination unit 131 acquires an image of a predetermined area on which the serial number is printed out of the entire image of the banknote acquired by the image acquisition unit 11 in step S11 of FIG. 6 (step S51). For example, when the banknote acquired by the image acquisition unit 11 in step S11 of FIG. 6 is a banknote B11 (see FIG. 2A), in step S51 of FIG. 28, as shown in FIG. 29A. , The image G71 including the serial number B14 is acquired. On the other hand, when the banknote acquired by the image acquisition unit 11 in step S11 of FIG. 6 is the banknote B61 (see FIG. 24), in step S51 of FIG. 28, the serial number as shown in FIG. 30A. Image G81 including B64 is acquired.

Next, the determination unit 131 scans the image acquired in step S51 in the lateral direction and searches for the center position of the serial number from the pixel distribution corresponding to the serial number (step S52). For example, when the image acquired in step S51 of FIG. 28 is that of banknote B11 (see FIG. 2 (a)), in step S52 of FIG. 28, as shown in FIG. 29 (b), the image G71 The central position L41 in is searched for. On the other hand, when the image acquired in step S51 of FIG. 28 is that of the bill B61 (see FIG. 24), in step S52 of FIG. 28, as shown in FIG. 30 (b), the central position in the image G81. L41 is searched.

Next, the determination unit 131 sets a first region, which is a standard serial number region, based on the central position L41 of the serial number searched in step S52 (step S54). For example, when the banknote acquired in step S51 of FIG. 28 is that of banknote B11 (see FIG. 2A), in step S54 of FIG. 28, it has a rectangular shape as shown in FIG. 29C. The first region R41 of is set. On the other hand, when the banknote acquired in step S51 of FIG. 28 is that of banknote B61 (see FIG. 24), in step S54 of FIG. 28, the first rectangular shape as shown in FIG. 30 (c). The area R41 is set.

Next, the determination unit 131 sets a second region and a third region for determining the font pattern outside the upper and lower sides of the center in the horizontal direction of the first region set in step S54 (step S55). For example, when the image acquired in step S51 of FIG. 28 is that of banknote B11 (see FIG. 2 (a)), in step S55 of FIG. 28, as shown in FIG. 29 (d), the first The second region R42 is set above and outside the lateral center of the region R41, and the third region R43 is set below and outside the lateral center of the first region R41. On the other hand, when the banknote acquired in step S51 of FIG. 28 is that of banknote B61 (see FIG. 24), in step S55 of FIG. 28, as shown in FIG. 30 (d), the first region R41 The second region R42 is set above and outside the center in the lateral direction, and the third region R43 is set below and outside the center in the lateral direction of the first region R41.

Next, the determination unit 131 determines whether or not the serial number is detected in the second region R42 and the third region R43 set in step S55 based on the pixel distribution corresponding to the serial number (step S56). .. For example, in the determination unit 131, the area of the pixel corresponding to the serial number is present in a predetermined area (for example, 1% of the area of the second region R42) or more in the second region R42, and in the third region R43. In the case where the area printed in the color of the ink used for printing the serial number is equal to or larger than a predetermined area (for example, 1% of the area of the third region R43), the serial number is detected. Can be determined.

When the banknote acquired in step S51 of FIG. 28 is that of banknote B11 (see FIG. 2 (a)), it is described in the second region R42 and the third region R43 as shown in FIG. 29 (d). Since the pixel corresponding to the number does not exist in the predetermined value or more, the determination unit 131 determines that the serial number is not detected in the second region R42 and the third region R43 (NO in step S56 of FIG. 28). Then, the determination unit 131 determines that the font pattern of the serial number of the banknote for which the image was acquired in step S51 of FIG. 28 (that is, the banknote for which the image was acquired in step S11 of FIG. 6) is the font pattern F11 (that is, the font pattern is the font pattern F11). Step S57). After that, the process of step S15 of FIG. 6 is performed.

On the other hand, when the image acquired in step S51 of FIG. 28 is that of the bill B61 (see FIG. 24), the serial numbers are in the second region R42 and the third region R43 as shown in FIG. 30 (d). Since there are more than a predetermined value of pixels corresponding to the above, the determination unit 131 determines that the serial number has been detected in the second region R42 and the third region R43 (YES in step S56 of FIG. 28). Then, it is determined that the font pattern of the serial number of the banknote whose image was acquired in step S51 of FIG. 28 (that is, the banknote whose image was acquired in step S11 of FIG. 6) is the font pattern F15 (step S58). After that, the process of step S15 of FIG. 6 is performed.

According to the fourth embodiment described above, the serial numbers of the banknotes of the same denomination are the first font pattern F11 (that is, as shown in FIG. 5A, each of the plurality of characters included in the serial number B14. And the second font pattern F15 (that is, as shown in FIG. 27, the heights of the plurality of characters included in the serial number B54 increase toward the center and decrease toward the end). However, it can be correctly determined whether the serial number of the banknote is the first font pattern F11 or the second font pattern F15. Further, according to the fourth embodiment, in step S56 of FIG. 28, not only is it determined whether the serial number is included in the second region R42, but also whether the serial number is included in the third region R43. Therefore, the accuracy of the font pattern determination process can be improved.

[First modification]
Next, a first modification will be described. FIG. 31 is a diagram showing an example of serial number and font pattern determination processing according to the first modification. In the first modification, the font pattern F21 having the serial numbers shown in FIGS. 31 (a) and 31 (b) and the font pattern F22 having the serial numbers shown in FIGS. 31 (c) and 31 (d) can be determined. do.

As shown in FIG. 31B, the serial number B71 of the banknote identified in the first modification has a plurality of characters (that is, 1, A, B, 2, 3, 4, 5, 6, 7). include. The line L51 connecting the vertical centers of the plurality of characters included in the serial number B71 is parallel to the long side of the bill. The size of the plurality of characters included in the serial number B71 decreases from the left end (that is, "1") to the right end (that is, "7"). In the font pattern F21, for example, serial numbers are detected in the upper and lower regions R51 and R52 of the outer left end portion of the predetermined rectangular region R50 including the serial number, and the upper and lower regions of the outer right end portion of the rectangular region R50 are detected. If the serial number is not detected in R53 and R54, it is determined that the font pattern is F21. Based on this determination, the character detection unit 132 identifies the positions of the plurality of characters included in the serial number B71 by using the template corresponding to the font pattern F21. In FIG. 31B, when the upper left corner of the frame surrounding the character “1” is the upper left apex of the rectangular area R50, the area R52 below the outer left end of the rectangular area R50. When the serial number is detected in, and the serial number is not detected in the region R54 below the right end portion outside the rectangular region R50, it may be determined that the font pattern F21. Further, in FIG. 31B, when the lower left corner of the frame surrounding the character “1” is the lower left apex of the rectangular region R50, the region R51 above the outer right end of the rectangular region R50. When the serial number is detected in the area R53 above the right end portion outside the rectangular region R50, the font pattern F21 may be determined. By doing so, the load of the determination process can be further reduced.

Further, as shown in FIG. 31 (d), the serial number B72 of the banknote identified in the first modification has a plurality of characters (that is, 1, A, B, 2, 3, 4, 5, 6, 7). )It is included. The line L51 connecting the vertical centers of the plurality of characters included in the serial number B72 is parallel to the long side of the bill. The size of the plurality of characters included in the serial number B72 increases from the left end (that is, "1") to the right end (that is, "7"). In the font pattern F22, for example, the serial number is not detected in the upper and lower regions R51 and R52 of the outer left end portion of the predetermined rectangular region R50 including the serial number, and the serial number is not detected in the upper and lower regions of the outer right end portion of the rectangular region R50. When the serial number is detected in the areas R53 and R54, it is determined that the font pattern is F22. Based on this determination, the character detection unit 132 identifies the positions of the plurality of characters included in the serial number B72 by using the template corresponding to the font pattern F22.

In the first modification, in order to determine the font pattern F21 and the font pattern F22, the paper leaf identification device 1 replaces the second region R12 with the regions R51 and R52 in step S25 of the flowchart of FIG. R53 and R54 are set. Then, the paper leaf identification device 1 determines whether or not the serial number is detected in the regions R51, R52, R53, and R54 in step S26 of the flowchart of FIG. 7. When the serial number is detected in the areas R51 and R52 and the serial number is not detected in the areas R53 and R54, the paper sheet identification device 1 displays the serial number of the banknote to be processed in FIG. 31 (a). It is determined that the font pattern F21 is shown. Further, when the serial number is not detected in the areas R51 and R52 and the serial number is detected in the areas R53 and R54, the paper sheet identification device 1 sets the serial number of the banknote to be processed as FIG. 31 (c). ) Is determined to be the font pattern F22.

According to the first modification, regardless of whether the serial number of the banknote is the font pattern F21 or the font pattern F22, both can be correctly identified. According to the first modification, even when the size of a plurality of characters included in the serial number of a banknote gradually changes in a predetermined direction (for example, the longitudinal direction of the banknote), the font pattern is correctly adjusted. Can be identified.

[Second modification]
Next, a second modification will be described. FIG. 32 is a diagram showing an example of serial number and font pattern determination processing according to the second modification. In the second modification, the font pattern F31 having the serial numbers shown in FIGS. 32 (a) and 32 (b) and the font pattern F32 having the serial numbers shown in FIGS. 32 (c) and 32 (d) can be determined. do.

As shown in FIG. 32 (b), the serial number B81 of the banknote identified in the second modification has a plurality of characters (that is, 1, A, B, 2, 3, 4, 5, 6, 7). include. The size of the characters 6 and 7 included in the serial number B81 is smaller than the size of the other characters 1, A, B, 2, 3, 4, and 5. The font pattern F31 is, for example, a serial number in the upper and lower regions R61 and R62 of a predetermined rectangular region R60 including a serial number in a predetermined position in the left-right direction (here, the third and fourth character positions from the left). Is detected, and if no serial number is detected in the upper and lower areas R63 and R64 of the right end portion outside the rectangular area R60, it is determined that the font pattern F31. Based on this determination, the character detection unit 132 identifies the positions of the plurality of characters included in the serial number B81 by using the template corresponding to the font pattern F31.

Further, as shown in FIG. 32 (d), the serial number B82 of the banknote identified in the second modification has a plurality of characters (that is, 1, A, B, 2, 3, 4, 5, 6, 7). )It is included. The size of the letters B and 2 included in the serial number B82 is smaller than the size of the other letters 1, A, 3, 4, 5, 6 and 7. In the font pattern F32, for example, the serial number is not detected in the outer regions R61 and R62 of the predetermined rectangular region R50 including the serial number, and the serial number is detected in the outer regions R63 and R64 of the rectangular region R60. If so, it is determined that the font pattern is F32. Based on this determination, the character detection unit 132 identifies the positions of the plurality of characters included in the serial number B82 by using the template corresponding to the font pattern F32.

In the second modification, in order to determine the font pattern F31 and the font pattern F32, the paper leaf identification device 1 replaces the second region R12 with the regions R61, R62 in step S25 of the flowchart of FIG. Set R63 and R64. Then, the paper leaf identification device 1 determines whether or not the serial number is detected in the regions R61, R62, R63, and R64 in step S26 of the flowchart of FIG. 7. When the serial number is detected in the areas R61 and R62 and the serial number is not detected in the areas R63 and R64, the paper sheet identification device 1 displays the serial number of the banknote to be processed in FIG. 32 (a). It is determined that the font pattern is F31. Further, when the serial number is not detected in the areas R61 and R62 and the serial number is detected in the areas R63 and R64, the paper sheet identification device 1 sets the serial number of the banknote to be processed as FIG. 32 (c). ) Is determined to be the font pattern F32.

According to the second modification, even when a part of the plurality of characters included in the serial number of the banknote is smaller than the other characters, the font pattern can be correctly identified.

[Third variant]
Next, a third modification will be described. FIG. 33 is a diagram showing an example of serial number and font pattern determination processing according to the third modification. In the third modification, the font pattern F41 having the serial numbers shown in FIGS. 33 (a) and 33 (b) and the font pattern F42 having the serial numbers shown in FIGS. 33 (c) and 33 (d) can be determined. do.

As shown in FIG. 33 (b), the serial number B91 of the banknote identified in the third modification has a plurality of characters (that is, 1, A, B, 2, 3, 4, 5, 6, 7). include. The size of the characters 1 and A included in the serial number B91 is larger than the size of the other characters B, 2, 3, 4, 5, 6 and 7. In the font pattern F41, for example, the serial number is detected in the upper and lower regions R71 and R72 of the left end portion outside the predetermined rectangular region R70 including the serial number, and the font pattern F41 is located at a predetermined position in the left-right direction outside the rectangular region R70. In FIG. 33A, when the serial number is not detected in the upper and lower regions R73 and R74 of the “4” and “5” character positions), it is determined to be the font pattern F41. Based on this determination, the character detection unit 132 identifies the positions of the plurality of characters included in the serial number B91 by using the template corresponding to the font pattern F41.

Further, as shown in FIG. 33 (d), the serial number B92 of the banknote identified in the third modification has a plurality of characters (that is, 1, A, B, 2, 3, 4, 5, 6, 7). )It is included. The size of the characters 4 and 5 included in the serial number B92 is larger than the size of the other characters 1, A, B, 3, 6 and 7. In the font pattern F42, for example, the serial number is not detected in the outer regions R71 and R72 of the predetermined rectangular region R70 including the serial number, and the serial number is detected in the outer regions R73 and R74 of the rectangular region R70. If so, it is determined that the font pattern is F42. Based on this determination, the character detection unit 132 identifies the positions of the plurality of characters included in the serial number B92 by using the template corresponding to the font pattern F42.

In the third modification, in order to determine the font pattern F41 and the font pattern F42, the paper leaf identification device 1 replaces the second region R12 with the regions R71 and R72 in step S25 of the flowchart of FIG. R73 and R74 are set. Then, the paper leaf identification device 1 determines whether or not the serial number is detected in the regions R71, R72, R73, and R74 in step S26 of the flowchart of FIG. 7. When the serial number is detected in the areas R71 and R72 and the serial number is not detected in the areas R73 and R74, the paper sheet identification device 1 displays the serial number of the banknote to be processed in FIG. 33 (a). It is determined that the font pattern is F41. Further, when the serial number is not detected in the areas R71 and R72 and the serial number is detected in the areas R73 and R74, the paper sheet identification device 1 sets the serial number of the banknote to be processed as FIG. 33 (c). ) Is determined to be the font pattern F33.

According to the third modification, even when a part of the plurality of characters included in the serial number of the banknote is larger than the other characters, the font pattern can be correctly identified.

As described above, in the present application,
(1) For example, in the first to fourth embodiments, in the paper leaf identification device 1, how the unique number (for example, serial number) of the paper leaf (for example, a bill) is printed on the paper leaf. In association with at least one font pattern F11, F22 indicating whether or not, and templates T11, T12 indicating the position where a plurality of characters included in the unique number are described in at least one font pattern F11, F12, of paper leaves. The storage unit 14 that stores each type (denomination C11, C12) in advance, the identification unit 12 (also referred to as the denomination identification unit 12) that identifies the type of paper leaves, and the image of the paper leaves are unique. Based on the second area R12 set outside the first area R11 including at least a part of the numbers, the identification unit 12 is of the type from at least one font pattern F11 and F12 stored in the storage unit 14. The determination unit 131 that determines the font pattern of the paper leaf that has identified the above and the template associated with the font pattern determined by the determination unit 131 are read from the storage unit 14 to detect a plurality of characters included in the unique number. The character detection unit 132 is provided. As a result, the font pattern can be correctly identified even when the unique number is printed on the same type of paper leaf with different font patterns.

(2) For example, in the first to fourth embodiments, the determination unit 131 determines whether or not the second region R12 includes at least a part of the unique number, so that the identification unit 12 identifies the type. Judge the font pattern of the paper leaves. As a result, in order to determine the font pattern based on whether or not the second region R12 includes at least a part of the unique number, as compared with the case where the font pattern is determined for the entire serial number region, The load of judgment processing can be reduced.

(3) For example, in the first to fourth embodiments, the determination unit 131 determines the font pattern after removing the backgrounds of the first region R11 and the second region R12. As a result, even when the plurality of characters included in the unique number and the background are printed in the same area, the font pattern can be correctly determined, and the plurality of characters can be correctly recognized.

(4) For example, in the fourth embodiment, the determination unit 131 is stored in the storage unit 14 based on the third region set outside the first region in the image obtained by capturing the paper sheets. From at least one font pattern, the identification unit 12 determines the font pattern of the paper leaf type for which the type has been identified. As a result, the font pattern can be determined based on a plurality of regions of the banknote (for example, the second region R42 and the third region R43 in FIG. 30 (d)), so that the font pattern determination may be erroneous. Can be lowered.

In the first embodiment and the like, the case where the serial number of the banknote is printed in the upper right and lower left areas of the banknote has been described (FIGS. 2 (a) to 2 (c), FIGS. 10, 17, and 24. (See), but is not limited to this. For example, the serial number of a banknote may be printed in a specific area of the banknote (for example, an area such as upper left or lower right of the banknote).

Further, in the first embodiment, a case where it is determined in step S26 of FIG. 7 whether the font pattern used for the banknote to be processed is one of the two font patterns F11 and F12 has been described. It is not limited to. For example, in the first embodiment, in step S25 of FIG. 7, the process of step S35 of FIG. 14 relating to the second embodiment is also performed at the same time, that is, together with the second region R12 shown in FIG. 8 (d), FIG. The second region R22 shown in (d) is set, and the font pattern to be processed is the font pattern F11, F12, F13 based on whether or not the serial number is detected in the second regions R12 and R22. It may be decided which is.

In the above-mentioned example, only the image in the upper right region of the entire image of the banknote is acquired, and the font pattern used for printing the banknote is determined based on the image, but the present invention is limited to this. It's not something. For example, an image in the upper right region and an image in the lower left region of the entire image of the banknote may be acquired, and the font pattern used for printing the banknote may be determined based on those two images. Alternatively, an image of the lower left region of the entire image of the banknote may be acquired, and the font pattern used for printing the banknote may be determined based on the image.

It should be noted that the functions of each part of the paper leaf identification device 1 shown in the block diagram of FIG. 1 and the processing by the paper leaf identification device 1 shown in the flowcharts of FIGS. 6, 7, 14, 21, and 28 are realized. These processes may be performed by recording the program for recording on a computer-readable recording medium, causing the computer system to read the program recorded on the recording medium, and executing the program. The term "computer system" as used herein includes hardware such as an OS and peripheral devices. Further, the "computer system" shall also include a WWW system provided with a homepage providing environment (or display environment). The "computer-readable recording medium" refers to a portable medium such as a flexible disk, a magneto-optical disk, a ROM (Read Only Memory), a CD-ROM, or a storage device such as a hard disk built in a computer system. say. Furthermore, a "computer-readable recording medium" is a volatile memory (RAM) inside a computer system that serves as a server or client when a program is transmitted via a network such as the Internet or a communication line such as a telephone line. Including those that hold the program for a certain period of time.

The present invention is not limited to the above-described embodiment, and includes various changes and improvements that can be made within the scope of the technical idea thereof.

1 Paper leaf identification device 10 Control unit 11 Image acquisition unit 12 Kind identification unit 13 Serial number identification unit 14 Storage unit 131 Judgment unit 132 Character detection unit

Claims (5)

At least one font pattern indicating how the unique number of the paper leaf is printed on the paper leaf, and the position where a plurality of characters included in the unique number are described in the at least one font pattern. A storage unit that stores in advance for each type of paper leaf in association with the indicated template,
An identification unit that identifies the type of paper leaves and
In the image obtained by capturing the paper sheets, the at least one font pattern stored in the storage unit is based on the second region set outside the first region including at least a part of the unique number. From the inside, a determination unit that determines the font pattern of the paper leaf whose type is identified by the identification unit, and
A character detection unit that reads a template associated with the font pattern determined by the determination unit from the storage unit and detects a plurality of characters included in the unique number, and a character detection unit.
Paper leaf identification device equipped with. The determination unit determines whether or not the second region includes at least a part of the unique number, thereby determining the font pattern of the paper leaf whose type is identified by the identification unit. The paper leaf identification device described in 1. The paper leaf identification device according to claim 1, wherein the determination unit determines the font pattern after removing the backgrounds of the first region and the second region. In the image obtained by capturing the paper sheets, the determination unit further selects the font pattern stored in the storage unit from among the at least one font pattern stored in the storage unit based on the third region set outside the first region. The paper leaf identification device according to claim 1, wherein the identification unit determines a font pattern of the paper leaf whose type has been identified. At least one font pattern indicating how the unique number of the paper leaf is printed on the paper leaf, and the position where a plurality of characters included in the unique number are described in the at least one font pattern. By associating with the template shown, it is stored in advance for each type of paper leaf.
Identify the type of paper leaf and
In the image obtained by capturing the paper sheets, from among the at least one font pattern stored in advance based on the second region set outside the first region including at least a part of the unique number. Judge the font pattern of the paper leaf that identified the type,
A paper leaf identification method for reading a template associated with a determined font pattern and detecting a plurality of characters included in the unique number.

Images