JP2005309743A - Leaf-like article discrimination device and method thereof - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a leaf-like article discrimination device and a method thereof that can achieve a constant density of a read characteristic image of a leaf-like article without being affected by a variation in density of a characteristic image of a leaf-like article by its smear, a flaw, etc. when a characteristic image of the inserted leaf-like article is read. <P>SOLUTION: Light is applied to the leaf-like article being conveyed, and an image for adjustment of the leaf-like article is obtained from the leaf-like article to which light is applied. The amount of emission of the light applied to the leaf-like article is adjusted based on the obtained image for adjustment. The light in the adjusted amount is applied to the leaf-like article, and an image for discrimination including a characteristic image is obtained from the leaf-like article. The leaf-like article is discriminated based on the obtained image for discrimination. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a paper sheet identification apparatus and method, and in particular, when reading a feature image of a paper sheet that can identify the type and authenticity of a paper sheet inserted into the paper sheet identification apparatus. The present invention relates to a paper sheet identification apparatus and method in which the density of a feature image of a read paper sheet is constant without being influenced by the density variation of the feature image such as a watermark region due to dirt or scratches.

  In general, in a paper sheet identification device that identifies the type and authenticity of paper sheets such as banknotes, checks, gift certificates, etc., the paper sheets inserted in the apparatus are put into the apparatus using a magnetic sensor or an optical sensor. The characteristics of the detected paper sheet are detected magnetically or optically, and the type and authenticity of the paper sheet are identified by comparing the detected characteristics of the paper sheet with the characteristics of the genuine note.

  For example, when an optical means is used as a means for detecting the characteristics of a paper sheet, a characteristic image such as a pattern of the paper sheet is optically detected using a transmitted light type optical sensor or a reflective optical sensor, The type and authenticity of the paper sheet are identified by collating the detected feature image with a standard image corresponding to the pre-registered paper sheet feature image.

  Therefore, in order to be able to more accurately identify the type and authenticity of the paper sheet, it depends on the density variation of the feature image due to dirt or scratches on the paper sheet inserted into the paper sheet identification device. In addition, it is indispensable to make the density of the characteristic image of the read paper sheet constant.

  For example, Patent Document 1 discloses a light-emitting element current value adjustment circuit that automatically sets the light emission amount of a light-emitting element to a predetermined value suitable for banknote identification and automatically adjusts the light emission intensity of the identification light-emitting element or the like during a banknote identification operation. Proposals have been made.

Japanese Patent Application Laid-Open No. 2004-228688 discloses an adjustment method for an image reading sensor capable of obtaining an image output that is not affected by changes in temperature characteristics of the image reading sensor and changes with time. And devices have been proposed.
JP 05-226079 A JP 2000-222614 A

  By the way, the proposed Patent Document 1 changes the resistance value while detecting the light emission amount of the light emitting element, stores the resistance value when the light emission amount matches a predetermined value suitable for banknote identification, and stores the banknote identification operation. In some cases, the light emitting element is caused to emit light by setting the resistance value stored in the memory to automatically adjust the light emission intensity of the identifying light emitting element of the bill validator.

  Further, Patent Document 2 stores a peak position when the light source is individually turned on for each block and an output value when the light source is fully lit. When adjusting for a change with time, the output difference in the vicinity of the peak position when the light source is fully lit is the smallest. In this way, the light emission amount of each block is adjusted so that a stable image can be acquired with respect to a change with time.

  However, in Patent Document 1 and Patent Document 2 described above, even if the brightness variation of the light emitting element or the light source can be dealt with, it does not correspond to the density variation of the feature image such as the banknote watermark area or the image of the object. When banknotes with varying density of feature images such as watermark areas due to dirt, scratches, etc. are inserted, the density of captured images obtained by imaging the banknotes also varies, causing inconvenience in identifying banknotes Have

  Therefore, the present invention keeps the density of the feature image of the read paper sheet constant without being affected by the density variation of the feature image such as a watermark area due to illumination variation, paper sheet dirt, scratches, and the like. An object of the present invention is to provide a paper sheet identification apparatus and method that can stably identify paper sheets.

  In order to achieve the above object, the invention according to claim 1 is a paper sheet identification device that reads an image of a paper sheet and identifies the paper sheet based on the read image. A light emitting means for irradiating light on the paper sheet conveyed by the conveying means, an image reading means for reading an image from the paper sheet irradiated with light by the light emitting means, and An adjustment image acquisition unit that acquires an adjustment image of the paper sheet based on the output, and a light emission amount of the light emission unit and an output of the image reading unit based on the adjustment image acquired by the adjustment image acquisition unit An adjustment unit that adjusts at least one of the light sources, and the adjustment image acquisition unit acquires the adjustment image after adjusting at least one of the light emission amount of the light emission unit and the output of the image reading unit. Based on the identification image acquisition means for acquiring the identification image including the characteristic image of the paper sheet based on the output of the image reading means in the same conveyance process, and the identification image acquired by the identification image acquisition means And a paper sheet identifying means for identifying the paper sheet.

  According to a second aspect of the present invention, in the first aspect of the invention, the adjustment unit includes a partial image extraction unit that extracts a part of the feature image included in the adjustment image acquired by the adjustment image acquisition unit. Density detecting means for detecting the density of the partial image extracted by the partial image extracting means; and density histogram calculating means for calculating a density histogram based on each density value of the partial image detected by the density detecting means. The light emission amount of the light emitting unit and the output of the image reading unit are adjusted based on the density histogram calculated by the density histogram calculating unit.

  According to a third aspect of the present invention, in the second aspect of the present invention, the adjusting unit is configured to calculate a light emission amount of the light emitting unit and an output of the image reading unit based on an area of the density histogram calculated by the density histogram calculating unit. It is characterized by adjusting at least one.

  According to a fourth aspect of the present invention, in the third aspect of the present invention, the density histogram calculation means calculates by thinning out the density values of the partial images detected by the density detection means.

  According to a fifth aspect of the present invention, in any one of the first to fourth aspects, the feature image is an image of a watermark area including a watermark pattern.

  The invention according to claim 6 is the invention according to any one of claims 1 to 5, wherein the adjustment image acquisition unit acquires the adjustment image and the identification image acquisition unit acquires the identification image. And the conveyance speed of the paper sheets by the conveyance means is constant. According to a seventh aspect of the present invention, in any one of the first to sixth aspects, the adjustment image acquired by the adjustment image acquisition means is used for identifying the paper sheet.

  The invention according to claim 8 is a paper sheet identification method for reading an image of a paper sheet by an image reading unit, and identifying the paper sheet based on the read image, conveying the paper sheet, The conveyed paper sheet is irradiated with light by the light emitting means, an adjustment image is acquired from the paper sheet irradiated with the light based on the output of the image reading means, and the light emission is performed based on the acquired adjustment image. Adjusting at least one of the light emission amount of the light emitting means and the output of the image reading means by the adjusting means, and adjusting the light emission amount of the light emitting means and at least one of the output of the image reading means by the adjusting means; An identification image is acquired based on the output of the image reading means in the same conveyance process as the acquisition, and the paper sheet is identified based on the acquired identification image.

  The invention according to claim 9 is the invention according to claim 8, wherein a part of the feature image included in the adjustment image is extracted, the density of the extracted partial image is detected by a density detector, and the density A density histogram is calculated by the density histogram calculation means based on each density value of the partial image detected by the detection means, and the light emission amount of the light emission means and the output of the image reading means are calculated based on the density histogram calculated by the density histogram calculation means. It is characterized by adjusting at least one of these.

  According to a tenth aspect of the invention, in the ninth aspect of the invention, at least one of the light emission amount of the light emitting means and the output of the image reading means is adjusted based on the area of the density histogram calculated by the density histogram calculating means. It is characterized by that.

  The invention according to claim 11 is the invention according to claim 10, wherein the density histogram calculation means calculates by thinning out each density value of the partial image detected by the density detection means.

  According to a twelfth aspect of the present invention, in any one of the eighth to eleventh aspects, the feature image is an image of a watermark region including a watermark pattern.

  The invention of claim 13 is the invention according to any one of claims 8 to 12, wherein the conveyance speed of the paper sheets is made constant at the time of acquisition of the adjustment image and at the time of acquisition of the identification image. It is characterized by.

  According to a fourteenth aspect of the present invention, in any one of the eighth to thirteenth aspects, the adjustment image is used for identifying the paper sheet.

  According to the paper sheet identifying apparatus and method of the present invention, the transport means for transporting the paper sheet, the light emitting means for irradiating the paper transported by the transport means, and the light emitted by the light emitting means. Acquired by an image reading unit that reads an image from the irradiated paper sheet, an adjustment image acquisition unit that acquires an adjustment image of the paper sheet based on an output of the image reading unit, and the adjustment image acquisition unit Adjusting means for adjusting at least one of the light emission amount of the light emitting means and the output of the image reading means based on the adjustment image, and at least one of the light emission amount of the light emitting means by the adjusting means and the output of the image reading means. The image for identification including the characteristic image of the paper sheet based on the output of the image reading means in the same conveyance process as the acquisition of the adjustment image by the adjustment image acquisition means. Since it is configured to include an identification image acquisition unit to be obtained and a paper sheet identification unit to identify the paper sheet based on the identification image acquired by the identification image acquisition unit, the paper sheet is imaged. The density of the feature image of the read paper sheet is constant, without being affected by the deterioration of the lighting, and without being affected by variations in the density of the feature image of the paper sheet due to dirt or scratches on the paper sheet. As a result, the feature image of the paper sheet can be read.

  The paper sheet identification apparatus and method according to the present invention reads, for example, a feature image formed on a paper sheet, compares the read feature image with a standard image corresponding to the feature image, and determines the type and authenticity of the paper sheet. In a paper sheet identification device, for example, when identifying securities or banknotes in which a watermark area image including a watermark pattern or the like is formed as a feature image, dirt or scratches on the paper sheet watermark area, etc. Paper sheet identification device capable of stably identifying paper sheets by reading feature images of the paper sheets so that the density of the read image is constant even if there is image density variation due to It is used as

  In the following embodiments, description will be made on the assumption that an image of a watermark area including a watermark pattern formed on a paper sheet is read as a feature image to identify the type and authenticity of the paper sheet.

  An embodiment of a paper sheet identification apparatus and method according to the present invention will be described in detail with reference to the accompanying drawings.

  FIG. 1 is a schematic configuration diagram showing a configuration example of a paper sheet identification apparatus according to the present invention.

  As shown in FIG. 1, the paper sheet identification device 1 includes a control unit 2 that controls the entire paper sheet identification device 1, a paper sheet insertion unit 3 that is a paper sheet insertion port, and an insertion port. A paper sheet transport unit 4 that transports the inserted paper sheet, a paper sheet receiving unit 5 that receives the paper sheet, and a drive unit that drives and controls the paper sheet transport unit 4 according to an instruction from the control unit 2 6, an image reading device 8 that captures and acquires a characteristic image of a paper sheet conveyed by the paper sheet conveying unit 4, and a characteristic image of the paper sheet acquired by the image reading device 8 and registered in advance An identification unit 7 for identifying the type and authenticity of the paper sheet by collating with a standard image corresponding to the characteristic image of the paper sheet is provided.

  When a paper sheet is inserted, the paper sheet insertion unit 3 that is an insertion port for the paper sheet detects that the paper sheet has been inserted by a detection sensor (not shown), and sends a signal to that effect to the control unit. 2 to send.

  The paper sheet transport unit 4 includes a drive motor, a transport belt, and the like so that the paper sheet inserted into the paper sheet insertion unit 3 can be transported in the insertion direction of the paper sheet or in the direction opposite to the insertion direction. The paper sheet is conveyed based on the control instruction of the drive unit 2.

  The drive unit 6 controls the paper sheet transport unit 4 based on the control instruction of the control unit 2 to transport the paper sheet in the insertion direction of the paper sheet or in the direction opposite to the insertion direction, or to control the transport stop, etc. Give instructions.

  The image reading device 8 adjusts the amount of light emitted from the light emitting means, which will be described later, and captures images of the paper sheets conveyed by the paper sheet conveyance unit 4 at a predetermined position. Acquires and outputs a feature image whose type and authenticity can be identified.

  The identification unit 7 collates the feature image of the paper sheet acquired by the image reading device 8 with a standard image corresponding to the feature image of the paper sheet registered in advance in a database (not shown), and Identify the type and true / false.

  Note that the standard image corresponding to the feature image of the paper sheet is acquired by previously capturing the feature image of the genuine paper sheet by the image reading device 8 for each paper sheet, and stored in a database (not shown).

  The paper sheet receiving unit 5 distinguishes the paper sheets transported by the paper sheet transport unit 4 for each type and stores them in the storage unit.

  The paper sheet identification device 1 configured in this way compares the feature image of the paper sheet acquired by the image reading device 8 with the standard image registered in the database, and identifies the type and authenticity of the paper sheet. The operation to be processed will be briefly described.

  The paper sheet inserted from the paper sheet insertion unit 3 is read by the image reading device 8 while the paper sheet is being conveyed to the paper sheet receiving unit 5 via the paper sheet conveyance unit 4. .

  The image reading apparatus 8 captures an area of a predetermined size including a characteristic image of a paper sheet twice, and performs an adjustment image (hereinafter referred to as “adjustment image”) and an identification image (hereinafter “ The image for identification ”is acquired, and the image for identification acquired by imaging for the second time is output to the identification unit 7.

  The identification unit 7 collates the input identification image with the standard image registered in the database to identify the type and authenticity of the paper sheet.

  When the identification unit 7 identifies the paper sheet as a genuine note by collating the identification image with the standard image, the drive unit 6 performs drive control of the paper sheet transport unit 4 based on a command instruction from the control unit 2. Then, the paper sheet is conveyed to the paper sheet receiving unit 5 and the paper sheet receiving unit 5 receives the conveyed paper sheet.

  When the identification unit 7 identifies the paper sheet as a fake ticket, the drive unit 6 controls the drive of the paper sheet transport unit 4 so that the paper sheet is transported to the paper sheet insertion unit 3 and returned. Do.

  As described above, in the paper sheet identification apparatus 1 that identifies the type and authenticity of the paper sheet by comparing the characteristic image of the paper sheet with the pre-registered standard image, the paper sheet is soiled or damaged. It is possible to stably identify the paper sheet by reading the feature image of the paper sheet so that the density of the read image is constant without being influenced by the density variation of the image of the paper sheet by It becomes possible.

  Therefore, the paper sheet identification device 1 according to the present invention has an image reading device 8 in which the density of the read characteristic image is constant even if the density of the characteristic image varies due to dirt or scratches on the inserted paper sheet. It has.

  FIG. 2 is a schematic configuration diagram showing a configuration example of the image reading device 8 of the paper sheet identification apparatus 1 according to the present invention.

  As shown in FIG. 2, the image reading device 8 includes an image reading control unit 80 that performs overall control of the entire image reading device 8 and a part of the watermark region of the paper sheet by capturing the conveyed paper sheet 20. Alternatively, an imaging unit 81 (a portion surrounded by a broken line) that acquires an adjustment image including all and an identification image, and a binarization processing unit that performs binarization processing on the adjustment image acquired by the imaging unit 81 82, a watermark region extraction unit 83 that extracts a watermark region image from the adjustment image based on the binarized image, and a watermark region density based on the watermark region image extracted by the watermark region extraction unit 83 A histogram calculation unit 84 that calculates a histogram, and a control amount for controlling the amount of light emitted to the paper sheet 20 when an image for identification is captured based on the density histogram calculated by the histogram calculation unit 84. calculate A control amount calculation unit 85, a memory 86 for storing various data acquired or calculated in each unit of the image reading device 8.

  It should be noted that the watermark area which is a characteristic image of the paper sheet 20 includes a watermark pattern.

  The imaging unit 81 includes a pair of light emitting means 811 and light receiving means 812, and the light emitting means 811 and the light receiving means 812 face each other and sandwich the conveyance line of the paper sheet conveyance section 4 that conveys the paper sheet 20. In this way, it is disposed at a predetermined position.

  In addition, the imaging unit 81 causes the light emitting means 811 to emit light at a predetermined position where the conveyed paper sheet 20 enters the imaging range 87 of a predetermined size of the imaging unit 81 and irradiates the paper sheet 20 with light. The light receiving means 812 detects the transmitted light that has passed through the paper sheet 20.

  The light emitting means 811 is composed of, for example, a plurality of light emitting elements, and is configured such that the light emission amount of the light emitting elements can be controlled.

  In addition, the light receiving unit 812 includes, for example, a plurality of light receiving elements, and receives light transmitted through the paper sheet 20 by light emitted from the light emitting unit 811 and outputs a signal corresponding to the received light amount. It is configured.

  Specifically, the light emitting element of the light emitting unit 811 emits light to irradiate light on the paper sheet 20, and the light receiving element of the light receiving unit 812 transmits the light transmitted through the paper sheet 20 to a signal level corresponding to the amount of light received. The electrical signal is output.

  Note that any of infrared, ultraviolet, and visible light can be applied to the light emitting means 811 and the light receiving means 812.

  In addition, the image pickup unit 81 has an image area including a part of the watermark area of the paper sheet 20 conveyed by the paper sheet conveyance unit 4 as an image pickup range 87 of the image pickup unit 81 (a part indicated by a mesh line in the drawing). ), The paper sheet 20 is imaged with a predetermined light emission amount (first time) to obtain an adjustment image, and after adjusting the light emission amount of the light emitting means 811 based on the acquired adjustment image, the paper sheet When the image area including all of the 20 watermark areas is entered, the paper sheet 20 is imaged (second time) to obtain an identification image.

  The first and second imaging timings include the entire image area and the watermark area including a part of the watermark area by detecting the transport position of the paper sheet 20 by a paper sheet position detection sensor (not shown). It is configured so that it can be detected that the image area has entered the range of the imaging unit 81.

  Further, the amount of light emitted from the light emitting means 811 at the time of imaging for obtaining an adjustment image (first time) is examined in advance by examining the influence of variations in the density of the image in the watermark area due to dirt or scratches on the paper sheets. Even if the adjustment image output in this way is a bright image, the image is adjusted to a light emission amount that does not saturate.

  In addition, the adjustment of the amount of light emitted when the adjustment image is captured is adjusted by increasing the light emission amount from the dark image state rather than adjusting the light emission amount from the image state in which the output adjustment image is saturated. More accurate adjustment can be performed based on the brightness and darkness state of the image with respect to the known light emission amount.

  The memory 86 stores a digital signal corresponding to the signal level of the electrical signal output from the imaging unit 81 at a predetermined time interval in a predetermined storage area of the memory 86 in correspondence with the output order from the imaging unit 81 and stores the digital signal. The digital signal information is assigned with continuous addresses, and is temporarily stored as image data of the paper sheet 20, and various data acquired or calculated in each unit of the image reading device 8 is stored.

  The binarization processing unit 82 reads out the adjustment image captured by the imaging unit 81 for the first time and stored in the memory 86, performs binarization processing on the adjustment image, and then performs the adjustment image (2 It is stored in the memory 86 as data).

  The watermark area extraction unit 83 specifies the watermark area of the adjustment image data based on the adjustment image (binarized) data stored in the memory 86, and the watermark is extracted from the adjustment image data based on the specified watermark area. The image of the area is extracted and stored in the memory 86 as watermark area image data.

  The histogram calculation unit 84 calculates a density histogram of the watermark region image based on the watermark region image data extracted by the watermark region extraction unit 83.

  Specifically, for example, the number of pixels corresponding to the density of each pixel of the watermark area image (density value is 0 to 255) is calculated using a histogram.

  The control amount calculation unit 85 picks up the optimal light emission of the light emission unit 811 when capturing the image of the paper sheet 20 being conveyed based on the density histogram of the watermark area calculated by the histogram calculation unit 84. The control amount of the light emitting means 811 corresponding to the amount and the light emission amount is calculated and output to the imaging unit 81.

  The operation of the image reading apparatus 8 configured in this manner to image the paper sheet being conveyed and read the adjustment image and the identification image will be described with reference to the flowchart of FIG.

  As shown in FIG. 3, when a paper sheet is inserted from the paper sheet insertion unit 3 of the paper sheet identification device 1 (YES in step S <b> 301), the inserted paper sheet is transferred to the paper sheet transport unit 4. Is conveyed to the imaging unit 81 of the image reading device 8 (step S302).

  The imaging unit 81 emits light when a part of the watermark region (for example, about half of the watermark region) that is a feature image of the conveyed paper sheet enters the imaging range 87 of the imaging unit 81 (YES in step S303). The means 811 emits light to irradiate the paper 20 with light, and the light receiving means 812 receives the transmitted light transmitted through the paper 20 and obtains it as an adjustment image (step S304).

  The image pickup unit 81 outputs an electric signal having a signal level corresponding to the amount of light received by the light receiving unit 812, and the memory 86 outputs digital signal information corresponding to the signal level of the electric signal output from the image pickup unit 81. The image is temporarily stored as an adjustment image of the paper sheet 20.

  When the adjustment image of the paper sheet 20 is stored in the memory 86, the binarization processing unit 82 reads out the adjustment image stored in the memory 86, and performs binarization processing on the adjustment image (step). In step S305, the watermark area extraction unit 83 extracts an image in the watermark area based on the adjustment image (binarization) data binarized by the binarization processing unit 82 and the adjustment image data (step S306).

  The binarization processing unit 82 binarizes the adjustment image, so that the distinction between the watermark region of the paper sheet and the normal region outside the watermark region becomes clear.

  When the watermark area image data is stored in the memory 86 by the watermark area extraction unit 83, the histogram calculation unit 84 reads the watermark area image data from the memory 86 and calculates a density histogram of the watermark area based on the watermark area image data. (Step S307).

  The density histogram data calculated by the histogram calculation unit 84 is input to the control amount calculation unit 85, and the light emission unit 811 emits light to capture an identification image based on the input density histogram data. The optimum light emission amount to be calculated is calculated, and the control amount of the light emitting means 811 corresponding to the calculated light emission amount is output (step S309).

  The control amount output by the control amount calculation unit 85 is input to the imaging unit 81, and after the light emission amount of the light emitting unit 811 is adjusted and controlled by the input control amount, the entire watermark region of the paper sheet 20 is included. When an image area of a predetermined size is entered (YES in step S310 and step S311), the paper sheet 20 is imaged (second time) and an identification image is acquired (step S312).

  The identification image of the paper sheet 20 captured and read by the imaging unit 81 is output from the image reading device 8 via the image reading control unit 80, and the processing operation of the image reading device 8 ends.

  The identification image of the paper sheet 20 output from the image reading device 8 is input to the identification unit 7 via the control unit 2 of the paper sheet identification device 1 and registered in advance with the identification image by the identification unit 7. A standard image corresponding to the identification image of the paper sheet is collated, and the type and authenticity of the paper sheet are identified.

  Here, the image reading is performed so that the density of the feature image of the read paper sheet is constant without being influenced by the density variation of the feature image due to dirt or scratches on the paper sheet into which the image reading device 8 is inserted. The method will be described in detail with reference to FIGS.

  FIG. 4 is a diagram illustrating an example of a method of acquiring an adjustment image by capturing an image of a paper sheet with the imaging unit 81 and extracting a watermark region from the acquired adjustment image.

  FIG. 4A is a diagram showing an image of the surface portion of the paper sheet, and FIG. 4B is a paper sheet when the adjustment image is acquired from the paper sheet 20 shown in FIG. The figure which shows the imaging position of the leaves 20, FIG.4 (c) is a figure which shows the image for adjustment acquired by imaging at the imaging position shown in FIG.4 (b), FIG.4 (d) is FIG.4. FIG. 4E shows an image obtained by binarizing the adjustment image shown in FIG. 4C, and FIG. 4E shows an image obtained by extracting a watermark area from the adjustment image shown in FIG. 4C. .

  FIG. 4 shows the case where the feature image of the paper sheet is a watermark region including a watermark pattern.

  As shown in FIG. 4A, the paper sheet 20 is composed of a watermark area 21 which is a characteristic image of the paper sheet 20 and a normal area 22 outside the watermark area, and the watermark area 23 has a watermark pattern 23. Is formed.

  The paper sheet 20 configured as described above is inserted from the paper sheet insertion unit 3 and conveyed to the imaging unit 81 of the image reading device 8 by the paper sheet conveyance unit 4 as shown in FIG. Then, when a part (about half) of the watermark area 21 of the paper sheet 20 enters the imaging range 87 of the imaging unit 81, the paper sheet 20 is imaged, and the adjustment image 400 as shown in FIG. Read.

  The adjustment image 400 includes a partial image 211 of the watermark area 21 (hereinafter referred to as “watermark area image (part) 211”) and a partial image 221 of the normal area 22 outside the watermark area (hereinafter referred to as “normal area”). Each image of the area image (part) 221 ") exists, and a part of the watermark pattern 23 (hereinafter referred to as" watermark pattern image (part) 231 ") is included in the watermark area image (part) 211. Existing.

  The binarization processing unit 82 performs binarization processing on the adjustment image 400, and the normal area image (part) 221 as shown in FIG. 4D is black, and the watermark area image (part) 211 is white. An image 410 (hereinafter, referred to as an “adjustment image (binarization) 410”) converted to each value indicating the above is calculated.

  By performing this binarization processing, the adjustment image binarized (binarization) 410 can clearly distinguish the watermark region 212 and the normal region 222, and the adjustment image (binarization) The watermark region 212 can be accurately specified from 410.

  The watermark area extraction unit 83 identifies a watermark area 212 (for example, the pixel value is “white”) from the adjustment image (binarization) 410 calculated by the binarization processing unit 82, and corresponds to the identified watermark area 212. An image area is extracted from the adjustment image 400 shown in FIG. 4C and an image 420 of the watermark area 211 including the watermark pattern 231 as shown in FIG. 420 ").

  When the watermark area image (for adjustment) 420 is extracted, the optimum light emission amount and light emission amount of the light emission means 811 for capturing an identification image based on the image density of the extracted watermark area image (for adjustment) 420 are obtained. A control amount is calculated.

  The control amount of the light emitting means 811 corresponding to the amount of light emitted by the light emitting means 811 is calculated based on the image density of the watermark area image (for adjustment) 420, for example.

  Specifically, for example, the density value (0 to 255) of each pixel of the extracted watermark area image (for adjustment) 420 is detected, and the pixels existing in the watermark area image (for adjustment) 420 corresponding to each density value are detected. A density histogram expressed by a number is calculated, and a control amount corresponding to an optimum light emission amount to be emitted by the light emitting unit 811 is calculated based on the calculated density histogram.

  FIG. 5 is an explanatory diagram showing an example of a method in which the histogram calculation unit 84 calculates a density histogram based on the watermark region image (for adjustment) 420.

  FIG. 5A is a diagram showing a watermark area image (for adjustment) 420, and FIG. 5B is a density calculated based on the watermark area image (for adjustment) 420 shown in FIG. It is a figure which shows an example of a histogram.

  As shown in FIG. 5A, the density value of each pixel 421 constituting the watermark area image (for adjustment) 420 is detected and aggregated, and each value of density value 0 to 255 in the horizontal axis (X) direction, A density histogram as shown in FIG. 5B represented by the value of the number of pixels Y existing in the watermark area image (for adjustment) 420 corresponding to each density value X (0 to 255) in the vertical axis (Y) direction. 500 is calculated.

  When calculating the density histogram 500, the processing time may be shortened by thinning out the image data of the watermark region image (for adjustment) 420.

  As shown in FIG. 5B, the pixel density Y corresponding to the density value X of the extracted watermark area image (for adjustment) 420 can be read from the calculated density histogram 500, and the watermark area image (adjustment) is adjusted. The trend of whether the image 420 is an overall bright image or a dark image can be read.

  For example, the correspondence relationship between the image brightness and darkness will be described with reference to FIG. 6, taking as an example each density histogram of a watermark area image (for adjustment) 420 including a bright image, a dark image, and a noise component including small holes. To do.

  FIG. 6 is a diagram illustrating an example of a correspondence relationship between the brightness and darkness of the watermark region image (for adjustment) 420 and the density histogram.

  FIG. 6A is a diagram showing an example of a density histogram when the watermark area image (for adjustment) 420 of the adjustment image is a bright image, and FIG. 6B is a watermark area image ( FIG. 6C is a diagram illustrating an example of a density histogram when the adjustment 420 is a dark image. FIG. 6C illustrates a small hole in the watermark area of the paper sheet due to mischief, and the watermark area including the small hole is the adjustment image. It is a figure which shows an example of the density histogram at the time of being imaged and acquired as.

  As shown in FIG. 6A, the density histogram 600 when the watermark area image (for adjustment) 420 is bright is the value of the density X of the image compared to the density histogram 610 when dark (see FIG. 6B). However, there is a wide range of pixels, and the maximum value Xmax of the density X of the pixels existing in the watermark area image (for adjustment) 420 is also larger than the density histogram 610 in the case of darkness.

  On the other hand, the density histogram 610 when the watermark area image (for adjustment) 420 is dark, as shown in FIG. 6B, is compared with the density histogram 600 of the bright image (see FIG. 6A). There are many pixels having a small value, and the maximum value Xmax of the density X of the pixels existing in the watermark area image (for adjustment) 420 is also smaller than that of the density histogram 600.

  Further, a small hole exists in the watermark region of the paper sheet 20 due to mischief, and the density histogram 620 when the watermark region including the small hole is captured as an adjustment image has a pixel density X corresponding to the small hole image region. The maximum value Xmax of the density X of the density histogram 620 is present isolated from the range of the density X value of each other pixel.

  As a method for calculating the optimum light emission amount to be emitted by the light emitting means 811 and capturing the control amount corresponding to the light emission amount when capturing an image for identification based on the correspondence between the brightness of the image and the density histogram, for example, The following method can be considered.

  The first method is a method of calculating the light emission amount to be emitted by the light emitting means 811 and the control amount corresponding to the light emission amount based on the calculated maximum value Xmax of the density X of the density histogram.

  When this method is used, as shown in the density histogram 620 of FIG. 6C, if a small hole exists in the watermark area of the paper sheet 20 due to mischief or the like, the density X of the pixel corresponding to the small hole Is isolated and detected as the maximum value Xmax of the density X of the density histogram 620, and may be erroneously recognized as a bright image even though it is originally a dark image.

  The second method is a method of calculating the optimum light emission amount and the control amount corresponding to the light emission amount based on the image density X occupied by the number of pixels equal to or larger than a predetermined value among the pixels constituting the image.

  In this method, for example, the density X value of the density histogram is changed from the smaller side (the density value is closer to 0) to the larger side (the density value is closer to 255), or the image density X value is larger (the value is higher). The cumulative value R is calculated by adding the number of pixels Y at each density X while sequentially moving from the density value closer to 255) to the smaller side (the density value closer to 0). Is determined based on the value X (D) of the density X of the image (hereinafter referred to as “specific density X (D)”), and the image for identification is determined. This is a method of calculating a light emission amount to be emitted by the light emitting means 811 when taking an image and a control amount corresponding to the light emission amount.

  A specific description of the second method will be described in detail with reference to FIG.

  FIG. 7 determines the brightness of the watermark area image (for adjustment) 420 based on the density histogram, and calculates the light emission amount to be emitted by the light emitting means 811 when the identification image is captured and the control amount corresponding to the light emission amount. It is explanatory drawing for demonstrating a method.

  FIG. 7A is a diagram showing an example of the density histogram calculated by the histogram calculation unit 84. FIG. 7B shows the light emission by the light emitting means 811 based on the density histogram shown in FIG. It is a figure which shows the reference table for calculating the control amount corresponding to the light emission amount to be made and the light emission amount.

  As shown in FIG. 7A, in the density X (horizontal axis (x)) of the density histogram 700 calculated by the histogram calculation unit 84, the value of the density X is changed from the larger side (density value is 255) to the smaller side (density value is 255). While moving the density value to 0) (in the direction of the arrow 710 in the figure), an accumulated value R is calculated by adding the number of pixels Y corresponding to each density X value.

Then, the specific density value X (D), which is the density X when the calculated accumulated value R reaches the predetermined value S, is specified.
That is, in the density histogram 700 shown in FIG. 7A, the number of pixels Y corresponding to each density X is added while moving the value of density X from Xmax to a smaller value (density value is 0). The accumulated value R is calculated, and the specific density that is the value of the density X when the accumulated value R reaches the predetermined value S (when the area S of the portion surrounded by the broken line of the density histogram 700 reaches the predetermined value S). X (D) is specified.

  The specific density X (D) specified in this way, that is, pixels having a density value equal to or higher than the specific density X (D) are present in the watermark region image (for adjustment) 420 by the number of pixels of approximately the predetermined value S. It shows that you are doing.

  The accumulated value R obtained by adding the number of pixels Y corresponding to each density X while moving the density value of the density histogram 700 from the smaller side (density value 0) to the larger side (density value 255) is a predetermined value S. When the specific density X (D) at the time of reaching is determined, pixels having the density X equal to or lower than the specific density X (D) are approximately the number of pixels of the predetermined value S in the watermark area image (for adjustment) 420. It will show that it exists.

  Therefore, when the value of the specific density X (D) specified from the density histogram 700 is closer to a smaller value (density value is 0), it is determined as a dark image, and a larger value (density value is 255). If it is close to, it is determined that the image is bright.

  Then, based on the determination result of the image brightness and darkness, when the image for identification is captured so that the value of the specific density X (D) of the image captured and read by the image capturing unit 81 is an optimal value without bias in brightness and darkness. The light emission amount of the light emitting means 811 is controlled.

  Specifically, when the specific density X (D) is specified from the density histogram 700, the light emission amount of the light emitting unit 811 is controlled by referring to the reference table 720 and corresponding to the specified specific density X (D). By controlling the light emitting means 811 by the control amount G, it is possible to perform adjustment control to the light emission amount of the light emitting means 811 capable of capturing an optimal image with no bias in light and dark.

  In the reference table 720, as shown in FIG. 7B, the left column shows the range of the value of the specific density X (D) specified from the density histogram, and the right column shows the specific density X (D ), The control amount G corresponding to the range of the control amount G is calculated. For example, if the value of the specific concentration X (D) is in the range of X (0) to X (1), the value of the control amount G Is G1 and the value of the specific concentration X (D) is in the range from X (1) to X (2), the control amount G can be calculated as G2.

  Note that the reference table 720 acquires the correspondence between the specific density X (D) and the control amount G in advance and stores it in the memory 86 as a reference table.

  Specifically, for example, the specific density X (D) of each light and dark image captured and read while changing the light emission amount of the light emitting means 811 corresponding to the control amount is calculated, and the optimum is not biased in light and dark. Corresponding to a predetermined range of the specific density X (D) of the image based on the difference between the control amount of the specific density X (D) of the correct image and each control amount corresponding to the specific density X (D) of each image of light and dark The control amount G of the light emitting means 811 capable of capturing an optimal image with no contrast in brightness is stored in the reference table 720 in association with the control amount G so that it can be calculated.

  In this way, light and darkness of the watermark region image (for adjustment) 420 is determined based on the specific density X (D) specified from the density histogram, and light applied to the paper sheet 20 corresponding to the specific density X (D). By controlling the light emission means 811 with a control amount corresponding to the optimal light emission amount, the watermark density of the paper sheet 20 can be identified without being affected by variations in the image density due to dirt, scratches, etc. It is possible to acquire an image for use.

  FIG. 8 is a diagram illustrating an example of a method of reading an identification image by irradiating a paper sheet with an optimal amount of light emission based on an adjustment image.

FIG. 8A is a diagram illustrating an image of the surface portion of the paper sheet, and FIG. 8B is a paper sheet when an image for identification is captured from the paper sheet 20 illustrated in FIG. The figure which shows the imaging position of the leaves 20, FIG.8 (c) is a figure which shows the image for an identification imaged and read at the imaging position shown in FIG.8 (b).
As shown in FIGS. 8A and 8B, a control amount corresponding to the amount of light emitted to the paper sheet 20 is calculated based on the adjustment image, and the light emission means 811 is calculated based on the calculated control amount. 8C, when the predetermined image area including the entire watermark area 21 of the paper sheet 20 enters the imaging range 87 of the imaging unit 81, the paper sheet 20 is imaged. Such an identification image 800 is read.

  As shown in FIG. 8 (c), before the image of the paper sheet 20 is captured and the identification image is read, the optimal light emission amount is obtained corresponding to the density variation of the adjustment image captured and read of the paper sheet 20. After the adjustment, an image for identification is captured, so that it is possible to acquire a feature image with a constant image density.

  There is a small hole in the feature image of the paper sheet due to, for example, mischief by adjusting and controlling the amount of light emitted to the paper sheet when the feature image of the paper sheet is captured by the method described above Even if there is a noise component such as, a feature image with a constant image density can be read.

  In this embodiment, an example is shown in which the light emission amount of the light emitting unit 811 of the imaging unit 81 is adjusted and controlled. However, the output of a signal corresponding to the amount of light received by the light receiving unit 812 It may be configured to perform adjustment control corresponding to the image density due to scratches or the like, and both the light emitting means 811 and the light receiving means 812 are adjusted and read according to the image density due to dirt or scratches on the paper sheet. Alternatively, the density of the identification image may be constant.

Schematic configuration diagram of a paper sheet identification apparatus according to the present invention Schematic configuration diagram of an image reading apparatus Flowchart showing the operation of the image reading device for acquiring a feature image of a paper sheet The figure which shows an example of the method of extracting a watermark area | region from the image for adjustment acquired by imaging paper sheets Explanatory diagram of how the histogram calculation unit calculates the density histogram The figure which shows an example of the correspondence of the brightness and darkness of a watermark area | region image (for adjustment), and a density histogram Explanatory drawing of the method of calculating the control amount corresponding to the light emission amount made to emit light with the light emission means 811 based on a density histogram. The figure which shows an example of the method which irradiates the light of the light-emission amount calculated based on the image for adjustment to a paper sheet, and acquires the image for identification

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Paper sheet identification device 2 Control part 3 Paper sheet insertion part 4 Paper sheet conveyance part 5 Paper sheet receiving part 6 Drive part 7 Identification part 8 Image reader 20 Paper sheet 21 Watermark area 22 Normal area 23 Watermark pattern 80 Image Reading Control Unit 81 Imaging Unit 82 Binarization Processing Unit 83 Watermark Area Extraction Unit 84 Histogram Calculation Unit 85 Control Amount Calculation Unit 86 Memory 400 Adjustment Image 410 Adjustment Image (Binarization)
420 Watermark area image (for adjustment)
800 Identification Image 811 Light Emitting Element 812 Light Receiving Element

Claims (14)

In a paper sheet identification device that reads an image of a paper sheet and identifies the paper sheet based on the read image,
Conveying means for conveying the paper sheet;
A light emitting means for irradiating light on the paper sheet conveyed by the conveying means;
Image reading means for reading an image from the paper sheet irradiated with light by the light emitting means;
An adjustment image acquisition means for acquiring an adjustment image of the paper sheet based on the output of the image reading means;
Adjusting means for adjusting at least one of the light emission amount of the light emitting means and the output of the image reading means based on the adjustment image acquired by the adjustment image acquiring means;
After adjusting at least one of the light emission amount of the light emitting means and the output of the image reading means by the adjusting means, the output of the image reading means is output in the same conveying process as the adjustment image acquiring means acquiring the adjustment image. An identification image acquisition means for acquiring an identification image including a characteristic image of the paper based on the basis,
A paper sheet identification device comprising: a paper sheet identification unit that identifies the paper sheet based on the identification image acquired by the identification image acquisition unit. The adjusting means is
Partial image extraction means for extracting a part of the feature image included in the adjustment image acquired by the adjustment image acquisition means;
Density detecting means for detecting the density of the partial image extracted by the partial image extracting means;
Density histogram calculation means for calculating a density histogram based on each density value of the partial image detected by the density detection means,
The paper sheet identification apparatus according to claim 1, wherein at least one of a light emission amount of the light emitting unit and an output of the image reading unit is adjusted based on the density histogram calculated by the density histogram calculating unit. The adjusting means is
The paper sheet identification apparatus according to claim 2, wherein at least one of a light emission amount of the light emitting unit and an output of the image reading unit is adjusted based on an area of the density histogram calculated by the density histogram calculating unit. The density histogram calculation means includes:
The paper sheet identification apparatus according to claim 3, wherein the density is calculated by thinning out the density values of the partial images detected by the density detection unit. The feature image is
The paper sheet identification apparatus according to claim 1, wherein the paper sheet identification image includes a watermark area including a watermark pattern. The conveyance speed of the paper sheets by the conveyance unit is fixed at the time of acquisition of the adjustment image by the adjustment image acquisition unit and at the time of acquisition of the identification image by the identification image acquisition unit. The paper sheet identification device according to any one of claims 1 to 5. The paper sheet identification apparatus according to claim 1, wherein the adjustment image acquired by the adjustment image acquisition unit is used for identifying the paper sheet. In a paper sheet identification method for reading an image of a paper sheet by an image reading unit and identifying the paper sheet based on the read image,
Transport the paper sheets,
Irradiate the conveyed paper sheet with light by a light emitting means,
An image for adjustment is obtained from the paper sheet irradiated with the light based on the output of the image reading means,
Based on the acquired adjustment image, adjust at least one of the light emission amount of the light emitting means and the output of the image reading means by an adjusting means,
After adjusting at least one of the light emission amount of the light emitting unit and the output of the image reading unit by the adjusting unit, an identification image is acquired based on the output of the image reading unit in the same conveyance process as the acquisition of the adjustment image. ,
The paper sheet identifying method, wherein the paper sheet is identified based on the acquired identification image. Extracting a part of the feature image included in the adjustment image;
Detecting the density of the extracted partial image by a density detecting means;
A density histogram is calculated by the density histogram calculation means based on each density value of the partial image detected by the density detection means,
9. The paper sheet identification method according to claim 8, wherein at least one of a light emission amount of the light emitting unit and an output of the image reading unit is adjusted based on the density histogram calculated by the density histogram calculating unit. The paper sheet identification method according to claim 9, wherein at least one of a light emission amount of the light emitting unit and an output of the image reading unit is adjusted based on an area of the density histogram calculated by the density histogram calculating unit. The paper sheet identification method according to claim 10, wherein the density histogram calculation unit performs calculation by thinning out the density values of the partial images detected by the density detection unit. The paper sheet identification method according to claim 8, wherein the feature image is an image of a watermark region including a watermark pattern. The paper sheet identification method according to any one of claims 8 to 12, wherein a conveyance speed of the paper sheet is constant at the time of acquisition of the adjustment image and at the time of acquisition of the identification image. The paper sheet identification method according to claim 8, wherein the adjustment image is used for identifying the paper sheet.

Images