JP2006174498A - Recording medium processing method, recording medium processing apparatus, and recording medium processing program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a recording medium processing method, a recording medium processing apparatus, and a recording medium processing program capable of preventing the occurrence of an event that required information cannot be read from image data stored by compressing a data quantity as to a recording medium including visibly recorded image data. <P>SOLUTION: The recording medium processing method for processing the recording medium including visibly recorded data includes: a reception step of receiving the recording medium at a prescribed reception unit; a first acquisition step of obtaining gray scale image data for representing a prescribed part of the received recording medium; a discrimination step of whether or not information represented by the gray scale image of the prescribed part can be reproduced by binary image data on the basis of the acquired image data; a second acquisition step of obtaining binary image data for representing at least part of the received recording medium when the discrimination step discriminates that the information can be reproducible by the binary image data; a recording step of recording the binary image data obtained in the second acquisition step to a storage means; and an output step of providing an output of a prescribed signal when the discrimination step discriminates that the information cannot be reproducible by the binary image data. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

The present invention relates to a recording medium processing method, a recording medium processing apparatus, and a recording medium processing program suitable for processing a recording medium on which a visible record such as a check or bill is recorded.

  Conventionally, when a payment is made by check at a store, the store receives a check from a customer as a buyer and confirms its validity. Then, if it is confirmed that it is valid, it is brought to a clearing institution such as a bank after having written a book or endorsement on the check as necessary. At the clearing house, settlement is performed by performing a transfer process from the account of the transferor to the store account based on the check brought in. Thus, in the conventional check settlement, the bill itself is brought from the store to the settlement organization, and settlement is performed accordingly.

  On the other hand, in recent years, electronic payment processing by checks is also progressing. Data such as account numbers recorded on checks is read by a terminal device installed in a store, and this data is sent to a settlement institution server for settlement. It is also being done. If the check settlement is digitized in this way, it is not necessary to bring the check to a settlement institution, thereby improving business efficiency. In this case, the check that is no longer required to be brought into the clearing house is returned from the store to the buyer on the spot. However, if you return the check used in the transaction without leaving any record, there will be no evidence on the store side that the purchase was made with the check, and there will be a difference in the amount or purchase between the buyer and the buyer. We cannot respond to troubles such as discrepancies between buying and selling. Therefore, when making check settlement electronically, at a store, image data obtained by scanning a check with a scanner is left as evidence of the transaction.

The image data obtained by the check scan usually has a relatively large number of gradations such as 8 bits (256 gradations), and the data size becomes large. In addition, since the store side needs to store images of checks drawn by various buyers for a considerable period of time, the number of image data to be stored also increases. Therefore, if the image data obtained by scanning is stored with the same number of gradations, the capacity of the storage device necessary for data storage becomes enormous, and it takes a long time to transfer the image data. Will occur.
JP-A-8-249399

  In order to avoid such inconvenience, it is effective to store the image data after reducing the data amount by binarizing. However, in order to use the stored image data as evidence that the purchase was made with a check, items such as the check number, the buyer's signature, the amount of money, etc. recorded on the check are represented by the image data. It is necessary to be. There are a wide variety of checks, and in the case of a dark check or when the signature ink is thin, these necessary items may be confused with the check. Therefore, if the image data is binarized in such a case, the above-mentioned necessary items cannot be read, and even if the binarized image data is saved, it can be used as evidence that a purchase was made with a check. It will disappear.

  The present invention has been made in view of the above points, and prevents a situation in which necessary information cannot be read from image data stored by compressing the data amount of a recording medium on which visual recording has been performed. It is an object to provide a recording medium processing method, a recording medium processing apparatus, and a control method for the recording medium processing apparatus.

The invention described in claim 1 is a method of processing a recording medium on which visual recording is performed,
(A) obtaining gray scale image data representing a predetermined portion of the recording medium; (b) creating a gray level frequency distribution for the gray scale image data, and binarizing based on the frequency distribution Determining a threshold; (c) transmitting whether or not the threshold has been determined in step (b);
(D) obtaining binarized image data representing at least a part of the recording medium using the threshold value determined in the step (b);
(E) transmitting the binarized image data obtained in step (d);
It is characterized by providing.

  The invention described in claim 2 is characterized in that, in the step (b), a threshold between the dark side peak and the bright side peak of the frequency distribution is determined.

  According to a third aspect of the present invention, there is provided a recording medium processing apparatus for processing a recording medium on which visible recording has been performed, wherein the first acquisition means obtains grayscale image data representing a predetermined portion of the recording medium. Creating a gray level frequency distribution for the gray scale image data and determining a threshold for binarization based on the frequency distribution; and whether the threshold has been determined by the threshold determining means. First transmission means for transmitting whether or not, second acquisition means for obtaining binarized image data representing at least a part of the recording medium using the threshold determined by the threshold determination means, and the second acquisition And second transmission means for transmitting the binarized image data obtained by the means.

  Further, the invention described in claim 4 is characterized in that the threshold value determining means determines a threshold value between a dark side peak and a bright side peak of the frequency distribution as a threshold value.

According to a fifth aspect of the present invention, there is provided a computer of a recording medium processing apparatus that processes a recording medium on which visual recording is performed.
(A) obtaining gray scale image data representing a predetermined portion of the recording medium;
(B) creating a gray level frequency distribution for the gray scale image data and determining a threshold for binarization based on the frequency distribution;
(C) a procedure for transmitting whether or not the threshold value has been determined in the procedure (b);
(D) obtaining binarized image data representing at least a part of the recording medium using the threshold value determined in the procedure (b);
(E) a procedure for transmitting the binarized image data obtained in the procedure (d);
Is executed.

  According to the present invention, it is possible to prevent a situation in which necessary information cannot be read from image data stored with a data amount compressed for a recording medium on which visible recording has been performed.

  FIG. 1 is an external view of the entire system according to an embodiment of the present invention. FIG. 2 is a block diagram of the system according to this embodiment. The system according to the present embodiment is configured by a check processing device 10 and a register device 20 that are installed at a checkout counter or the like of a store and are connected to be communicable with each other.

  As shown in FIG. 1, the check processing apparatus 10 is covered with, for example, a resin cover 12, provided with an insertion port 14 for inserting a check on the front surface thereof, and discharging the check to the upper surface. A discharge port 16 is provided. In addition, a roll paper discharge port 18 for discharging roll paper stored in a roll paper storage unit (not shown) and printed by a roll paper printing unit 48 described later is provided behind the upper surface of the check processing device 10. Is provided. On the other hand, the register device 20 includes a display unit 22 with a touch panel function, a credit card reader unit 26, a customer display 27, and the like.

  As shown in FIG. 2, the check processing device 10 includes a central processing unit (CPU) 40. The register device 20 includes a host computer 28 and a storage device 30. The CPU 40 of the check processing device 10 and the host computer 28 of the register device 20 can communicate with each other by wire or wireless, and can exchange various data.

  In the register device 20, the display unit 22, the credit card reader unit 26, and the storage device 30 are connected to a host computer 28. Further, the host computer 28 is connected to a settlement institution server (settlement institution server) 34 via a communication line 32.

  Note that the settlement institution may be a settlement agency such as a so-called clearing house (an organization that performs settlement processing in response to a request from a store or the like), or may be a financial institution itself. Such a clearing house has a function to determine the validity of a check based on the account number of the check maker and a function to execute payment by check, but in the case of a clearinghouse, depending on its authority, In some cases, it is only possible to determine the validity of a check, and it does not have a function to perform a payment process. In such a case, an inquiry about the validity of a check may be sent to the clearinghouse server, and a settlement execution instruction may be sent to a financial institution (or a clearinghouse having a settlement function). A description will be given assuming that the settlement institution has both of these functions.

  As shown in FIG. 2, the check processing apparatus 10 also includes a MICR reading unit 42 connected to the CPU 40, a cover printing unit 44, an end printing unit 46, a roll paper printing unit 48, a scanner unit 50, a check transport unit 52, And a check position detector 54 and the like.

  The MICR reading unit 42 has a function of performing reading processing (MICR: Magnetic Ink Character Recognition) of magnetic ink characters recorded on the surface of the check, and supplies the read data to the CPU 40. As will be described later, the CPU 40 makes an inquiry to the settlement institution server 34 about the validity of the check based on the data read by the MICR reading unit 42.

  FIG. 3 is a schematic diagram illustrating an example of a check processed by the check processing apparatus 10 according to the present embodiment. The magnetic ink characters are recorded in the MICR recording area 60 on the check surface shown in FIG. 3, and the recorded data includes the account number of the check maker. Here, the account number means a number that can uniquely identify an account including a bank code, a store code, and the like.

  Each of the front cover print unit 44 and the end print print unit 46 includes a print head of an appropriate method such as a dot impact method. The front cover printing unit 44 prints the front line items such as the payee, date, amount, etc. on the surface of the check in response to an instruction from the CPU 40, and the endorsement printing unit 46 responds to the instruction from the CPU 40. On the back of the check, the endorsement items necessary for the store, such as the shopper's authentication number, date, and usage amount, are printed. The book items are printed in the book region 62 shown in FIG. 3, and the endorsement items are printed in the end document region 64 on the back of the check.

  The roll paper printing unit 48 includes a print head of an appropriate system such as a thermal transfer system, for example, and in accordance with an instruction from the CPU 40, a consent form (check is electronically printed on the roll paper drawn from the roll paper storage unit. (Consent document for processing), receipt, coupon, receipt, etc. are printed, and then the printed roll paper is cut by an auto cutter (not shown).

  The scanner unit 50 is configured to include an appropriate image sensor such as a contact image sensor, for example, and scans the surface of the check in accordance with an instruction from the CPU 40 and supplies the obtained image data to the CPU 40. In the present embodiment, each pixel of the image data generated by the scanner unit 50 is represented by, for example, a gray scale of 8 bits 256 gradations (0 to 255).

  The check transport unit 52 is for transporting the check along the transport path, and includes a plurality of pairs of transport rollers, a roller driving unit, and the like.

  The check position detection unit 54 is for detecting the position of the check on the conveyance path. The position where the check is inserted by a predetermined amount from the insertion slot 14, the book printing unit 44 and the end document printing unit 46. The CPU 40 supplies a signal indicating whether or not the print start position by the MICR reading unit 42, the scan start position by the scanner unit 50, the scan start position by the scanner unit 50, a pre-scan start position to be described later, and the like have been reached.

  Next, the content of the process performed in this embodiment is demonstrated.

  4 and 5 are flowcharts showing the flow of the entire process executed in the present embodiment. The processing shown in FIGS. 4 and 5 is realized by the CPU 40 and the host computer 28 reading out and executing the programs stored in the memory devices, respectively.

  As shown in FIG. 4, when the cashier process is completed for a shopper, the host computer 28 determines the settlement amount for the shopper (S100). Next, a selection screen for selecting a payment method (whether cash or credit card is a check) is displayed on the display unit 22 (S102), and it is determined which payment method has been selected on this selection screen (S104). . As a result, when cash or a credit card is selected, settlement by cash or a credit card is performed accordingly (S106), and the process ends. On the other hand, when a check is selected as the settlement method, a confirmation screen for confirming whether or not the check settlement can be performed electronically is displayed on the display unit 22 (S108). If the clerk asks the shopper to approve the check settlement electronically, and if the approval is obtained, select the "Confirm" button on the confirmation screen above, but if the approval is not obtained Select the “Conventional processing” button. The host computer 28 determines which button has been selected on the confirmation screen (S110), and if the “confirm” button is selected, a screen for inputting whether or not a book has already been made on the check. Is displayed (S112). The presence / absence of a book input on this screen is used when determining whether or not to perform book printing in the check processing device 10, as will be described later. The processing when the “conventional processing” button is selected will be described later.

  When the presence / absence of a cover is entered, a message prompting the insertion of a check is then displayed (S114). In response to this, the store clerk inserts the check received from the shopper into the insertion section 14 of the check processing device 10. The CPU 40 of the check processing device 10 determines whether or not a check has been inserted based on the detection signal from the check position detection unit 54 (S116), and when inserted, transmits a signal to that effect to the host computer 28 (S116). S118). In response to this signal, the host computer 28 sends an instruction to the CPU 40 to read the magnetic ink character of the check (S120).

  In response to this instruction, the CPU 40 causes the check conveyance unit 52 to convey the check to a reading position by the MICR reading unit 42, and then causes the MICR reading unit 42 to read the magnetic ink character of the check (S122). If the magnetic ink characters are read correctly, the data such as the account number obtained by reading the data, and if the magnetic ink characters are not read, the data to that effect are read as the host computer. 28 (S124).

  The host computer 28 determines whether or not the magnetic ink characters are correctly read based on the reading result transmitted from the CPU 40 (S126). As a result, if it has not been read, a message indicating that the magnetic ink characters cannot be read is displayed (S128), and an instruction to discharge the check is transmitted to the CPU 40, and the process ends. When receiving this instruction, the CPU 40 causes the check transport unit 52 to discharge the check from the discharge port 16 (S131). Thereby, it is possible to prevent the processing from proceeding when an illegal check (or a piece of paper other than the check) without recording of magnetic ink characters is inserted, or when a check is inserted face down.

  On the other hand, when the magnetic ink character is read, the host computer 28 informs the settlement institution server 34 whether or not the account number included in the read data is correct (that is, whether or not an account with the account number exists). An inquiry is made (S132, S134). As a result, if the account number is not correct, a message indicating that the check cannot be handled is displayed (S136), and after the check discharge command is transmitted to the CPU 40 (S130), the process ends. . On the other hand, if the account number is correct, a pre-scan start instruction is transmitted to the CPU 40 (S138).

  When the CPU 40 receives the pre-scan start instruction, the CPU 40 performs a pre-scan check and executes a series of threshold value determination processes until the threshold value for binarizing the image data is determined (S140).

  FIG. 6 is a diagram for explaining the principle of threshold determination in the threshold determination processing of this embodiment, and shows an example of the gray level frequency distribution of each pixel constituting the check image. In the threshold value determination process according to the present embodiment, the frequency distribution of gray levels is determined based on the peak P1 of the portion having a small gray level value (that is, the dark portion) corresponding to the printed or handwritten character portion, and the background of the check. It is separated into a peak P2 having a large gray level value corresponding to the portion (that is, a bright portion), and an upper limit value Pmin of the dark side peak P1 and a lower limit value PMax of the bright side peak P2. Is determined as the threshold value TH.

  FIG. 7 is a flowchart showing an outline of threshold determination processing executed by the CPU 40 based on the above principle. As shown in FIG. 7, first, image data is acquired by pre-scanning for a specific portion of a check (S200). FIG. 8 shows a portion of a check that is an object of image acquisition by pre-scanning. As shown in FIG. 8, in the pre-scan, a portion (hereinafter referred to as MICR portion) SMICR that overlaps the MICR recording area 60 where magnetic ink characters are recorded, and a portion other than the magnetic ink character portion (hereinafter referred to as non-MICR portion). ) Obtain image data for Sb1 to Sb3. In the present embodiment, the MICR part Sm and the non-MICR parts Sb1 to Sb3 are selected so as to be arranged side by side in the width direction of the check. When scanning a check, the scanner unit 50 reads an image while feeding the check in the longitudinal direction. Therefore, by selecting the MICR part SMICR and the non-MICR parts Sb1 to Sb3 as described above, if the check is sent by the length of these parts (L shown in FIG. 8), the prescan ends. The time required can be shortened.

  However, the target part of the pre-scan is not limited to the above, and may be appropriately selected so that a more appropriate threshold can be determined. For example, the number of MICR parts is not limited to one, but may be two or more, and the number of non-MICR parts is not limited to three, but may be one or two, or four or more. You can choose.

  When the pre-scanning is completed, a frequency distribution representing the appearance frequency of the gray level is created for the image data obtained by the pre-scanning (S202).

  FIG. 9 is a flowchart showing details of the frequency distribution creation process. As shown in the figure, first, the frequency distribution Hm (g) is determined from the image data for the MICR unit SMICR (S202A). However, g is a gray level of the pixel (0 to 255 in the present embodiment), and a larger value indicates brighter (closer to white). In creating the frequency distribution Hm (g), the MICR part SMICR is a part representing particularly important information in the check, and thus the frequency value is weighted by, for example, doubling the value. That is, Hm (g) = 2 · Hm (g).

  Next, the frequency distributions Hb1 (g), Hb2 (g), and Hb3 (g) of each part are obtained from the respective image data of the non-MICR parts Sb1 to Sb3 (S202B).

  Next, the frequency distributions Hm (g), Hb1 (g), Hb2 (g), and Hb3 (g) obtained as described above are respectively expressed as H'm (g) and H'b1 (g) as follows. ), H′b2 (g), and H′b3 (g) (S202C).

When g ≦ BGth, H′m (g) = Hm (g)
H′bi (g) = Hbi (g) (i = 1, 2, 3)
When g> BGth, H′m (g) = 0 H′bi (g) = 0 (i = 1, 2, 3)
Here, BGth is a lower limit value of the gray level detected when a scanning target does not exist in front of the light receiving unit of the scanner unit 50 and a background image is captured, and is set to a value of about 254, for example. That is, even when the image of the back of the check is included in the image data due to the above correction, the frequency distribution for only the check portion is correctly obtained by clearing the frequency for the image portion to zero. Will be.

  Next, for each frequency distribution H′m (g), H′b1 (g), H′b2 (g), and H′b3 (g) corrected as described above, Average value of the total of (2 · K + 1) frequencies of the frequency of gray level g and the frequency of adjacent 2 · K gray levels (K on the side larger than gray level g and K on the smaller side) By taking the above, each frequency distribution is averaged. (S202D).

That is, for the MICR part SMICR:

H ″ m (g) is calculated by the above, and the non-MICR parts Sb1 to Sb3 are calculated.

  H ″ b1 (g), H ″ b2 (g), and H ″ b3 (g) are calculated. For example, 4 is used as the value of K.

  By this averaging process, the noise portion of the frequency distribution is removed, and the characteristics of the image of each portion are more correctly represented by the frequency distribution.

  Then, the frequency distributions H ″ b1 (g), H ″ b2 (g), and H ″ b3 (g) of the non-MICR part are summed as shown in the following formula to integrate the non-MICR part. The frequency distribution H ″ b (g) is calculated (S202E).

H ″ b (g) = H ″ b1 (g) + H ″ b2 (g) + H ″ b3 (g)
(G = 0-255)
Thus, the frequency distribution creation process ends.

  Next, a process for determining PMin and PMax, which are parameters for calculating the threshold, is performed from the frequency distributions H ″ m (g) and H ″ b (g) created as described above (FIG. 7 S204).

FIG. 10 is a flowchart showing details of the process for determining the parameters PMin and PMax. As shown in FIG. 10, first, the parameter PMin is determined from the frequency distribution H ″ m (g) of the MICR unit SMICR (S204A). This parameter PMin is such that the cumulative value of g = 0 to PMin is a predetermined value R (for example, 0.1) with respect to the relative frequency r (g) (ratio to the total frequency) of the frequency distribution H ″ m (g) of the MICR unit SMIR. ) (See the left part of FIG. 6 above). That is, PMin is

As

  Is obtained as the maximum value of g that satisfies the above.

  Next, the parameter PmMax is obtained from the frequency distribution H ″ m (g) of the MICR part Sm (S204B). Specifically, the value of g that maximizes the frequency of the frequency distribution H ″ m (g) is set as the mode value nmod, and the value of g is decreased by 1 from nmod in the frequency distribution H ″ m (g). In this case, the value of g at the point where the value of H ″ m (g) first falls below a predetermined value M (for example, 30) is defined as PmMax (see the right side of FIG. 6).

  Next, the parameter PbMax is obtained from the frequency distribution Hb (g) of the non-MICR part (S204C). Similarly to the case where the parameter PmMax is obtained, the parameter PbMax is also obtained by determining the value nmod of g that maximizes the frequency distribution H ″ b (g), and decreasing the value of g by 1 from nmod. In this case, the value of g at the point where the value of H ″ b (g) first falls below M is defined as PbMax.

  Then, PmMax for the MICR part SMICR is compared with PbMax for the non-MICR part, and the smaller value is adopted as PMax (S204D).

  When the parameters PMin and PMax are obtained as described above, it is next determined whether or not a binarization threshold can be determined (S206 in FIG. 7). Specifically, when the value of PMax is 0, it is determined that the threshold value cannot be determined, and in other cases, it is determined that the threshold value can be determined. That is, the value of PMax being 0 means that the frequency is the maximum for at least one of the frequency distribution H ″ m (g) of the MICR part SMICR or the frequency distribution H ″ b (g) of the non-MICR part Sb. This means that it does not fall below the predetermined value M on the lower side of the peak (the side with the smaller gray level). In such a case, the peak P2 on the bright side of the color corresponding to the background portion where no check character is recorded does not appear clearly in the frequency distribution, and therefore the threshold cannot be determined ( In other words, it is determined that the binarization makes it impossible to reproduce the information represented by the original image data).

  As a result of the determination process, if it is determined that the threshold value can be determined, the threshold value TH is calculated using Pmin and Pmax (S208 in FIG. 7). Specifically, the threshold value R is set to a value of T percent (for example, 58%) from the PMin side between PMin and PMax. That is, the threshold value TH is calculated by TH = PMin + (PMin−PMax) · T / 100.

  However, when PMin ≧ PMax (that is, when the dark-colored peak P1 and the bright-colored peak P2 overlap), the value of PMin is set as the threshold value TH.

  When the threshold value determination process described above is completed, the CPU 40 transmits a processing result indicating whether or not the threshold value has been determined to the host computer 28 (S142 in FIG. 5).

  The host computer 28 determines the processing result transmitted from the CPU 40 (S144). As a result, if the threshold value can be determined, it is determined that the necessary items for specifying the check can be reproduced even if the image data is binarized. In this case, since the binarized image data can be used as proof of the transaction, the processing based on the premise that the check settlement is performed electronically in S146 and after. On the other hand, if the threshold value could not be determined, if the image data is binarized, the necessary items for specifying the check cannot be reproduced. Therefore, the check settlement as before is performed non-electronically in S174 and thereafter. Processing that assumes this is executed.

  That is, when making a payment based on a check electronically, the check itself is returned from the store to the buyer, so the store stores the image data of the check as evidence that the check was made. At that time, in order to compress the data amount, the image data is binarized and stored. However, in order to use binarized image data as proof of transaction, it is essential to be able to identify the check by reading from the image data the items necessary to identify the check, such as check number, bank account number, bank signature, etc. It is. Therefore, if the threshold value cannot be determined, it is determined that the check cannot be specified from the binarized image data and cannot be used as evidence of the transaction, and settlement based on the check is performed electronically. To avoid it.

  When the threshold value TH for binarizing the image data can be determined, first, an instruction for endorsement printing is transmitted from the host computer 28 to the CPU 40 (S146). In response to this, the CPU 40 causes the check transport unit 52 to transport the check to the endorsement printing start position, and then causes the endorsement printing unit 46 to perform endorsement printing on the back side of the check (S148).

  When the host computer 28 receives the notification of completion of the endorsement printing from the CPU 40, it next transmits an instruction to check invalid printing to the CPU 40 (S150). In response to this, the CPU 40 causes the check transport unit 52 to transport the check to the invalid print start position, and then causes the book print unit 44 to perform check invalid printing such as “VOID” on the check surface (S152). If it is not desired to read the check invalid print in the image data, the check invalid print may be performed as a separate process.

  Further, when the host computer 28 receives a notification of the check invalid print completion from the CPU 40, the host computer 28 determines whether or not a book is placed on the check based on the result of the book presence / absence input (S112) (S154). ). As a result, if the cover is not performed, it is determined that the cover print is necessary, and an instruction to print the cover matter including the payee, date, amount, etc. is transmitted to the CPU 40. (S156). In response to this, the CPU 40 causes the check transport unit 52 to transport the check to the book print start position, and causes the book print unit 44 to print the book on the check (S158). When the host computer 28 receives a notification of the completion of the book printing from the CPU 40, the process proceeds to S160. On the other hand, if the book has already been written on the check in S154, it is determined that the book printing is unnecessary, and the process proceeds to S160 without sending a book printing command.

  Of the payee, amount, and date included in the book item, the payee is the name of the store where the system is installed and is preset for each system, and the amount is the settlement determined in S100 above. A monetary value is used and the date is obtained from the system clock. That is, it is not necessary to input a new book item when printing a book. The endorsement item is set in advance for each system. For example, when there are a plurality of cash registers in a store, a register number can be included as an endorsement item. The host computer 28 or the check processing device 10 can be set not to perform endorsement printing. In this case, the endorsement printing instruction in S146 and later-described S176 is omitted (set by the host computer 28). Or if the instruction for endorsement printing is received in S148, this instruction is ignored (when set by the check processing device 10).

  In S160, the host computer 28 sends an instruction to the CPU 40 indicating that a main scan of the check should be performed. Upon receiving this instruction, the CPU 40 causes the scanner unit 50 to execute an image scan while conveying the check by the check conveying unit 52, and discharges the checked check from the discharge port 16 (S162). Then, the 256 gradation image data obtained by scanning by the scanner unit 50 is binarized using the threshold value TH determined in the threshold value determination process (S140) (S164), and the binarized image data is converted into a host. The data is transmitted to the computer 28 (S166). The host computer 28 records the transmitted binarized image data in the storage device 30 together with settlement data including information such as the check account number and amount (S168). The acquisition of image data by the main scan in S162 may be performed on the entire check, or a part including matters necessary for specifying a check such as a check number, an account number, a maker signature, and an amount of money. You may go about.

  Further, the host computer 28 transmits an instruction to the CPU 40 to print the consent form (consent form for electronically processing the check) (S170). In response to this, the CPU 40 causes the roll paper printing unit 48 to print two primary and secondary consent documents (S172).

  In the store, the check scanned and discharged in S162 is purchased together with the two primary and secondary consent documents printed on the roll paper in S172 and the detailed receipt printed by the receipt printing unit 24 during the cash register operation. Give it to the customer. Of these, the positive consent document is collected with the signature of the shopper and stored at the store. Then, as will be described later, payment processing is electronically performed by communication between the host computer 28 and the payment institution server 34 at night or the like.

  On the other hand, if it is determined in S144 that the threshold value could not be determined, the host computer 28 causes the display unit 22 to display a message indicating that it is not appropriate to perform check settlement electronically (S174). Then, in the same manner as S146, S154, and S156 when the threshold value can be determined, the endorsement print instruction (S176), the presence / absence determination of the cover (S178), and the cover print instruction (S180) are performed, and then the check is sent to the CPU 40. A discharge instruction is transmitted (S182), and in response to this, the CPU 40 discharges the check from the discharge port 16 (S184).

  In addition, in the above-described step S110 (FIG. 4), if the consent for the digitization of the payment process is not obtained, the host computer 28 indicates that a check should be inserted to print the front / end document. Is displayed on the display unit 22 (S184), and thereafter, the processing after S176 is executed.

  When the check is discharged in S184, only the detailed receipt is given to the shopper, the check is stored at the store side, and the previous non-electronic check settlement is performed.

  The host computer 28 transmits a settlement request to the settlement institution server 34 based on the settlement data recorded in the storage device 30 at an appropriate timing such as at night every day. In response to the settlement request, the settlement institution server 34 performs a process for transferring the designated amount from the account of the check account number to the account of the requesting store, thereby completing the settlement of the check.

  As described above, in the present embodiment, a threshold value for binarizing image data can be determined based on 256-gradation image data obtained by pre-scanning the MICR and non-MICR portions of the check. It is determined whether or not. If the threshold value can be determined, the information necessary for specifying the check is not lost even if the image data is binarized and the data amount is compressed (that is, the information necessary for the binarized image data is not lost). And binarized image data is recorded to prepare for electronic settlement processing. Therefore, according to the present embodiment, when performing check settlement electronically, it is possible to store check image data as proof of transaction with a small data amount.

  On the other hand, if the above threshold value cannot be determined, it is determined that the information necessary for specifying the check will be lost if the data amount is compressed by binarization, and electronic settlement processing is prohibited. Perform the same processing as before. Therefore, according to the present embodiment, after returning the check to the buyer (drawer) to perform settlement electronically, it was found that the stored binary image data cannot be used as evidence that the check was traded. It is possible to prevent such a situation. In addition, in order to avoid such a situation, it is unnecessary to display a binarized image on the screen and a store clerk or the like to visually judge whether or not binarization is possible. Mistakes can also be prevented.

  As described above, since the binarized data is stored only when it can be determined that the necessary information can be read even if the image data is binarized and compressed, there is a situation in which necessary information cannot be read from the stored image data. Can be prevented.

  Also, in this embodiment, before pre-scanning, it is input whether the shopper has accepted the check settlement electronically (S110), and the threshold is determined by pre-scanning only when there is an approval. Process. For this reason, when the shopper does not wish to make an electronic payment, unnecessary processing such as pre-scanning and threshold determination processing is not required.

  Also, before performing the pre-scan, the magnetic ink characters are read, and if the magnetic ink characters cannot be read from the check, the check is immediately discharged, so the check is inserted face down as described above. In such a case, it is possible to prevent the processing from proceeding. On the other hand, when the magnetic ink characters are read, the account number is identified and stored from the read data, so that electronic payment can be automatically executed by communication with the payment server 34.

  In addition, when the threshold value can be determined in the threshold value determination process, invalid printing is performed on the check, so that the used check can be prevented from being reused, and invalid to prevent such reuse. The trouble of handwriting the display can be eliminated. On the other hand, when electronic payment is not performed because the threshold value cannot be determined, invalid printing is not performed, and therefore it is possible to prevent invalid printing from being performed on a check before the settlement processing.

  In addition, it is possible to write a book or endorsement on a check regardless of whether or not the threshold value can be determined by the threshold determination process (that is, whether or not check settlement is performed electronically). And the hassle of handwriting endorsements. In such a case, an input is made on whether or not a book has already been made on the check (S112), and if the book is made, the book is not printed on the check processing device 10. , It is possible to prevent the book from being overlaid on the checked check.

  Further, when electronic processing for check settlement is performed, the check is scanned after the book is printed, and the image data is binarized and stored. Therefore, the contents of the book are included in the stored image data, and the certainty as the transaction evidence of the image data can be further increased.

  In the above embodiment, the original image data of the check is collected on a gray scale of 256 gradations, and this is binarized to compress the data amount. However, the original image data is not limited to 256 gradations, and may be higher gradations such as 512 gradations, or may be low gradations such as 126 gradations. Further, the compression of the data amount is not limited to binarization, and may be performed by converting to a low gradation number such as 4 gradations (2 bits).

  The compression of the image data amount is not limited to reducing the number of gray scale gradations of the image data, and other appropriate image data compression methods may be used. In short, only when necessary items can be read from image data compressed by some compression method, the image data compressed by the compression method may be stored.

  In the above embodiment, when it is not possible to determine the threshold value, the host computer 28 displays a message indicating that it is not appropriate to make a payment electronically. This display is provided on the check processing device 10 side. You may make it carry out with display apparatuses, such as a light emitting element.

  In the above-described embodiment, the image data of 256 gradations generated by scanning the check by the scanner unit 50 is binarized by the CPU 40. However, the scanner unit 50 has a threshold value instructed by the CPU 40 to 2 with the threshold value. It may be configured to have a function of generating digitized image data and supply the binarized data to the CPU 40. In this case, instead of the check scan and image binarization processing in S162 and S164 (FIG. 5), an instruction to generate image data binarized with the threshold value TH from the CPU 40 may be sent. .

  In the above embodiment, the check processing apparatus 10 performs each process in response to an instruction from the host computer 28, and the display and input are performed on the host computer 28 side. However, the check processing apparatus 10 and the host computer 28 You may comprise as an apparatus which integrated.

  In the above embodiment, endorsement printing, invalid printing, and front cover printing are performed in this order. However, the order of printing is not limited to this, and depending on the positional relationship of each printing unit, for example, a check What is necessary is just to determine suitably so that a conveyance distance may become the shortest.

  In the above embodiment, the threshold determination process for binarizing the image data is performed by the process shown in FIG. 7, but the threshold determination is not limited to this, and other appropriate threshold determination methods are used. It may be used.

  Furthermore, the present invention is not limited to checks, and can be widely applied to a case where image data is compressed and stored for a recording medium on which various documents, documents, etc. are printed or visually recorded by handwriting. is there.

  According to the first to fifth aspects of the present invention, it is possible to prevent a situation in which necessary information cannot be read from image data stored by compressing the data amount of a recording medium on which visual recording has been performed. Therefore, the industrial applicability is very large.

It is an external view of the whole system which is one embodiment of the present invention. It is a block block diagram of the system of this embodiment. It is the schematic showing an example of the check processed with the check processing apparatus. It is a flowchart (the 1) showing the flow of the whole process performed by this embodiment. It is a flowchart (the 2) showing the flow of the whole process performed by this embodiment. It is a figure for demonstrating the principle of the threshold determination in the threshold determination process of this embodiment, and shows an example of the frequency distribution of the gray level of each pixel which comprises a check image. It is a flowchart showing the outline | summary of the threshold value determination process which CPU of a check processing apparatus performs. It is a figure which shows the location of the check used as the object of the image acquisition by prescan. It is a flowchart which shows the detail of a frequency distribution creation process. It is a flowchart showing the detailed content of the determination process of parameters PMin and PMax.

Explanation of symbols

10 Check Processing Device 14 Insertion Port 16 Discharge Port 22 Display Unit 30 Storage Device 34 Settlement Institution Server 40 Central Processing Unit (CPU)
42 MICR reading section 44 cover printing section 46 end printing section 50 scanner section

Claims (5)

A method of processing a recording medium on which a visual recording has been made,
(A) obtaining grayscale image data representing a predetermined portion of the recording medium;
(B) creating a gray level frequency distribution for the gray scale image data, and determining a threshold for binarization based on the frequency distribution;
(C) transmitting whether or not the threshold value has been determined in step (b);
(D) obtaining binarized image data representing at least a part of the recording medium using the threshold value determined in the step (b);
(E) transmitting the binarized image data obtained in step (d);
A recording medium processing method comprising:   2. The recording medium processing method according to claim 1, wherein, in the step (b), a threshold value is determined between a dark side peak and a bright side peak of the frequency distribution. A recording medium processing apparatus for processing a recording medium on which visual recording is performed,
First acquisition means for obtaining grayscale image data representing a predetermined portion of the recording medium;
Threshold value determining means for creating a gray level frequency distribution for the gray scale image data and determining a threshold value for binarization based on the frequency distribution;
First transmission means for transmitting whether or not the threshold value could be determined by the threshold value determination means;
Second acquisition means for obtaining binarized image data representing at least a part of the recording medium using the threshold determined by the threshold determination means;
Second transmission means for transmitting the binarized image data obtained by the second acquisition means;
A recording medium processing apparatus comprising:   4. The recording medium processing apparatus according to claim 3, wherein the threshold value determining unit determines a threshold value between a dark side peak and a bright side peak of the frequency distribution. In a computer of a recording medium processing apparatus that processes a recording medium on which visual recording is performed,
(A) obtaining gray scale image data representing a predetermined portion of the recording medium;
(B) creating a gray level frequency distribution for the gray scale image data and determining a threshold for binarization based on the frequency distribution;
(C) a procedure for transmitting whether or not the threshold value has been determined in the procedure (b);
(D) obtaining binarized image data representing at least a part of the recording medium using the threshold value determined in the procedure (b);
(E) a procedure for transmitting the binarized image data obtained in the procedure (d);
A recording medium processing program for executing

Images