JP2012159912A - Image processing device and control method of image processing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To suppress an occurrence of erroneous recognition of a CMC-7 character caused by recognition of a blot, a signature or the like as a part of the CMC-7 character and to accurately recognize the CMC-7 character when reading the CMC-7 character.SOLUTION: By comparing a cumulative quantity NWH that accumulates a quantity of white pixel regions WG as regions between a vertical bar Rn and a vertical bar R(n+1), in other words, the cumulative quantity NWH per a character height H, length in an X-axis direction, of a magnetic ink character, a position of the magnetic ink character having the character height H in the X-axis direction is determined, the position of the magnetic ink character in the X-axis direction is accurately recognized and pattern matching processing of the magnetic ink character can be performed. In such a manner, it is possible to provide an image processing device that suppresses erroneous recognition of the magnetic ink character and realizes recognition, of which a recognition speed is increased, of the magnetic ink character.

Description

  The present invention relates to an image processing apparatus and a control method for the image processing apparatus.

Conventionally, one of the character recognition technologies that read characters (magnetic ink characters) printed on a recording medium such as a check with ink containing a magnetic substance is a magnetic ink character recognition (MICR: Magnetic Ink Character Reader). ) There are several fonts for magnetic ink characters (MICR characters), one of which is CMC-7.
A method and apparatus for optical character recognition (OCR: Optical Character Recognition) of CMC-7 characters printed and recorded on a recording medium is disclosed (for example, see Patent Document 1).

Japanese Patent Laid-Open No. 5-250515

  However, if there is dirt or a sign other than the CMC-7 character in the character frame which is the position or range where the CMC-7 character is printed and recorded, when reading the CMC-7 character, Recognizing dirt or a sign as a part of the CMC-7 character causes erroneous recognition of the CMC-7 character. As a result, in order to accurately recognize the CMC-7 character, it is necessary to repeatedly read and execute it, and there is a problem that it takes time to recognize the CMC-7 character.

  The present invention has been made to solve at least a part of the above problems. It can be realized by the following forms or application examples.

  Application Example 1 In the image processing apparatus according to this application example, a plurality of vertical bars extending in the X-axis direction are arranged at intervals in the Y-axis direction orthogonal to the X-axis direction. An image processing apparatus for processing an image including a character, comprising at least an X-axis direction position detection unit for detecting a position of the magnetic ink character in the X-axis direction, wherein the X-axis direction position detection unit A quantity calculation unit for calculating the quantity of an area between the vertical bar and the other vertical bar in the Y-axis direction of the ink character; and the length of the magnetic ink character in the X-axis direction by accumulating the quantity. A cumulative quantity calculation unit that calculates a cumulative quantity per unit, an X-axis direction position determination unit that determines a position of the magnetic ink character in the X-axis direction based on the cumulative quantity, and pattern matching of the magnetic ink character Execute the process And summarized in that with the turn matching processing unit.

  According to this, in the X-axis direction position determination unit, the cumulative quantity obtained by accumulating the quantity of the area between the vertical bar and the other vertical bars, in other words, the cumulative quantity per length in the X-axis direction of the magnetic ink characters is obtained. By comparing, the position of the magnetic ink character in the X-axis direction can be determined, and the position of the magnetic ink character in the X-axis direction can be accurately recognized to execute the pattern matching process of the magnetic ink character. This causes dirt or a signature to be recognized as a part of the magnetic ink character, causing erroneous recognition of the magnetic ink character, and repeatedly performing pattern matching processing of the magnetic ink character until it is accurately recognized. There is no need. In this way, it is possible to provide an image processing apparatus that can realize recognition of magnetic ink characters that suppresses erroneous recognition of magnetic ink characters and accelerates recognition.

  Application Example 2 In the image processing apparatus according to this application example, the magnetic ink is configured such that a plurality of vertical bars extending in the X-axis direction are arranged at intervals in the Y-axis direction orthogonal to the X-axis direction. An image processing apparatus for processing an image including a character, comprising at least an X-axis direction position detection unit for detecting a position of the magnetic ink character in the X-axis direction, wherein the X-axis direction position detection unit A quantity calculation unit for calculating the quantity of the vertical bar area in the Y-axis direction of the ink character; and a cumulative quantity for calculating the cumulative quantity of the magnetic ink character per length in the X-axis direction by accumulating the quantity. A calculation unit; an X-axis direction position determination unit that determines a position of the magnetic ink character in the X-axis direction based on the accumulated quantity; and a pattern matching process that executes a pattern matching process of the magnetic ink character. And summarized in that a and.

  According to this, in the X-axis direction position determination unit, the cumulative quantity obtained by accumulating the quantity of the vertical bar area, in other words, the cumulative quantity per length in the X-axis direction of the magnetic ink character is compared, thereby comparing the magnetic ink character. The position in the X-axis direction is determined, the position of the magnetic ink character in the X-axis direction can be accurately recognized, and the pattern matching process of the magnetic ink character can be executed. This causes dirt or a signature to be recognized as a part of the magnetic ink character, causing erroneous recognition of the magnetic ink character, and repeatedly performing pattern matching processing of the magnetic ink character until it is accurately recognized. There is no need. In this way, it is possible to provide an image processing apparatus that can realize recognition of magnetic ink characters that suppresses erroneous recognition of magnetic ink characters and accelerates recognition.

  Application Example 3 The control method of the image processing apparatus according to this application example is configured such that a plurality of vertical bars extending in the X-axis direction are arranged at intervals in the Y-axis direction orthogonal to the X-axis direction. A method for controlling an image processing apparatus that processes an image including a magnetic ink character, comprising: an X-axis direction position detection step for detecting at least a position of the magnetic ink character in the X-axis direction, The detecting step includes a quantity calculating step for calculating the quantity of the magnetic ink character in an area between the vertical bar and the other vertical bar in the Y-axis direction, and accumulating the quantity to calculate the magnetic ink character in the magnetic ink character. A cumulative quantity calculating step for calculating a cumulative quantity per length in the X-axis direction; an X-axis direction position determining step for determining a position in the X-axis direction of the magnetic ink character based on the cumulative quantity; And summarized in that includes a pattern matching processing step of performing a pattern matching process of the magnetic ink characters.

  According to this, in the X-axis direction position determination step, the cumulative quantity obtained by accumulating the quantity of the area between the vertical bar and the other vertical bars, in other words, the cumulative quantity per length of the magnetic ink characters in the X-axis direction is obtained. By comparing, the position of the magnetic ink character in the X-axis direction can be determined, and the position of the magnetic ink character in the X-axis direction can be accurately recognized to execute the pattern matching process of the magnetic ink character. This causes dirt or a signature to be recognized as a part of the magnetic ink character, causing erroneous recognition of the magnetic ink character, and repeatedly performing pattern matching processing of the magnetic ink character until it is accurately recognized. There is no need. In this way, it is possible to provide a control method for an image processing apparatus that can realize recognition of magnetic ink characters that suppresses erroneous recognition of magnetic ink characters and accelerates recognition.

  Application Example 4 The control method of the image processing apparatus according to this application example is configured such that a plurality of vertical bars extending in the X-axis direction are arranged at intervals in the Y-axis direction orthogonal to the X-axis direction. A method for controlling an image processing apparatus that processes an image including a magnetic ink character, comprising: an X-axis direction position detection step for detecting at least a position of the magnetic ink character in the X-axis direction, The detecting step includes a quantity calculating step of calculating the quantity of the vertical bar area in the Y-axis direction of the magnetic ink character, and accumulating the quantity to accumulate the magnetic ink character per length in the X-axis direction. A cumulative quantity calculating step for calculating a quantity; an X-axis direction position determining step for determining a position of the magnetic ink character in the X-axis direction based on the cumulative quantity; and a pattern of the magnetic ink character. And summarized in that and a pattern matching processing step of performing the down matching.

  According to this, in the X-axis direction position determination step, by comparing the cumulative quantity obtained by accumulating the quantity of the vertical bar area, in other words, the cumulative quantity per length of the magnetic ink letter in the X-axis direction, The position in the X-axis direction is determined, the position of the magnetic ink character in the X-axis direction can be accurately recognized, and the pattern matching process of the magnetic ink character can be executed. This causes dirt or a signature to be recognized as a part of the magnetic ink character, causing erroneous recognition of the magnetic ink character, and repeatedly performing pattern matching processing of the magnetic ink character until it is accurately recognized. There is no need. In this way, it is possible to provide a control method for an image processing apparatus that can realize recognition of magnetic ink characters that suppresses erroneous recognition of magnetic ink characters and accelerates recognition.

1 is an external perspective view showing a check reading device according to a first embodiment. Explanatory drawing which shows the internal structure of the check reading apparatus of 1st Embodiment. The block diagram which shows the control system of the check reading apparatus of 1st Embodiment. The flowchart which shows operation | movement of the optical character recognition part of 1st Embodiment. The figure which shows typically an example of the state which expand | deployed the character string corresponding | compatible area | region image data of 1st Embodiment to the predetermined coordinate system. The flowchart which shows operation | movement of the optical character recognition part at the time of execution of the labeling process of 1st Embodiment. The flowchart which shows operation | movement of the optical character recognition part at the time of execution of the bandwidth detection process of 1st Embodiment. The figure which shows a mode that the fixed character frame was displayed on the character string corresponding | compatible area | region image data of 1st Embodiment. The figure which shows the character string image data produced | generated of 1st Embodiment. Schematic which shows the image and fixed character frame DM which concern on the magnetic ink character of 1st Embodiment. 4 is a schematic block diagram showing a part of a control system of the optical character recognition unit 80. FIG. The flowchart which shows operation | movement of the optical character recognition part at the time of execution of the optical character recognition process of 1st Embodiment. Schematic which shows the image and fixed character frame DM which concern on the magnetic ink character of 2nd Embodiment. The flowchart which shows operation | movement of the optical character recognition part at the time of execution of the optical character recognition process of 2nd Embodiment.

  In the following embodiments, a check reading apparatus 1 is described as an example of an image processing apparatus.

(First embodiment)
Hereinafter, a first embodiment of the present invention will be described with reference to the drawings.
FIG. 1 is an external perspective view showing a check reading apparatus 1 of the present embodiment.
The check reading device 1 reads a magnetic ink character (MICR character) recorded on the check 4, reads images on both sides of the check 4, and endorses the check 4 with respect to the check 4 that is a sheet-like recording medium. It is an apparatus that performs processing such as recording of a predetermined image.
The check reading device 1 includes a main body case 2 and a lid case 3 placed on the upper side of the main body case 2, and each component is incorporated therein.
The cover case 3 is formed with a conveyance path 5 of a check 4 composed of a narrow vertical groove having a U shape when viewed from above, and one end of the conveyance path 5 is composed of a wide vertical groove. The other end of the conveyance path 5 branches to the left and right, and is connected to a first check discharge unit 7 and a second check discharge unit 8 each having a wide vertical groove.

On the surface 4a of the check 4, a money amount, a payer, a number, a signature, and the like are written on the background on which a predetermined pattern is formed, and the lower end portion thereof has a long side direction as shown in FIG. An extended magnetic ink character string 4A is recorded. The magnetic ink character string 4A is formed by arranging a plurality of magnetic ink characters in the horizontal direction.
In addition, an endorsement column is formed on the back surface 4 b of the check 4. In the endorsement column, a predetermined image related to the endorsement is recorded by the recording device 56 (see FIG. 2).
The check 4 is aligned in the vertical direction so that the upper end 4e is positioned upward and the lower end 4f is positioned downward, and the front and back are aligned so that the surface 4a faces the outside of the U-shaped conveyance path 5 ( In the state shown in FIG. 1, it is inserted into the check supply unit 6. The check 4 inserted into the check supply unit 6 is sent out to the transport path 5 with the rear end 4d as the head.
As the check 4 sent out from the check supply unit 6 is conveyed along the conveyance path 5, the front image as the image of the front surface 4a and the back image as the image of the back surface 4b are read and further recorded on the front surface 4a. The magnetic ink character string 4A being read is magnetically read. The check 4 for which the magnetic ink character string 4 </ b> A has been successfully read is discharged to the first check discharge unit 7 after a predetermined image related to the endorsement is recorded. On the other hand, the check 4 that has failed to be read is discharged to the second check discharge unit 8 without recording a predetermined image related to the endorsement. The check 4 discharged to the second check discharge unit 8 is subjected to processing such as investigation of the cause of reading failure and re-reading.

FIG. 2 is an explanatory diagram showing the internal structure of the check reading device 1.
The check supply unit 6 is provided with a check delivery mechanism 10 for sending the check 4 to the transport path 5. The check feeding mechanism 10 includes a feeding roller 11, a feeding roller 12, a retard roller 13 pressed against the feeding roller 12, a feeding motor 14, and a check pressing hopper 15.
When the feeding motor 14 is driven, the check 4 put in the check supply unit 6 is pressed against the feeding roller 11 by the hopper 15, and in this state, the feeding roller 11 and the feeding roller 12 are rotated synchronously.
The check 4 is fed between the feed roller 12 and the retard roller 13 by the feed roller 11. The retard roller 13 is given a predetermined rotational load, and only one check 4 that is in direct contact with the delivery roller 12 is separated from the other checks 4 and delivered to the transport path 5.
The conveyance path 5 has a U shape as described above, and is connected to the linear upstream conveyance path portion 21 connected to the check supply unit 6 and the first and second check discharge units 7 and 8. A downstream conveyance path portion 23 that extends in a slightly bent state and a curved conveyance path portion 22 that connects them are provided.

A check conveyance mechanism 30 that conveys the check 4 sent from the check supply unit 6 to the conveyance path 5 along the conveyance path 5 is pressed by the first conveyance roller 31 to the sixth conveyance roller 36 and the rotation. The first, second, third, fourth, fifth, and sixth pressing rollers 41, 42, 43, 44, 45, and 46 and the first transport roller 31 to the sixth transport roller 36 are rotationally driven. And a transfer motor 37. The 1st conveyance roller 31-the 6th conveyance roller 36 rotate in synchronization. For example, a stepping motor is used as the transport motor 37. For this reason, the conveyance amount of the check 4 can be known from the number of steps for driving the stepping motor.
The first transport roller 31 to the third transport roller 33 are respectively disposed at the upstream end of the upstream transport path portion 21, the middle position thereof, and the boundary position with the curved transport path portion 22. The fourth transport roller 34 is disposed at a downstream position in the curved transport path portion 22. The fifth transport roller 35 and the sixth transport roller 36 are respectively disposed at the middle position and the downstream end in the downstream transport path portion 23.

  Between the first transport roller 31 and the second transport roller 32 in the upstream transport path portion 21, a magnet 51 for magnetizing magnetic ink characters, a front-side contact image scanner 52, and a back-side contact image scanner from the upstream side. 53 is arranged. The surface-side contact image scanner 52 faces the surface 4a of the check 4 conveyed on the conveyance path 5, and reads a surface image that is an image of the surface 4a. The back surface side contact image scanner 53 faces the back surface 4b of the check 4 conveyed on the conveyance path 5, and reads a back surface image which is an image of the back surface 4b.

Further, a magnetic head 54 for reading magnetic ink characters is disposed between the second transport roller 32 and the third transport roller 33, and the magnetic head 54 is pressed to press the check 4 against the head. The roller 55 is confronting.
Between the fifth conveyance roller 35 and the sixth conveyance roller 36 in the downstream conveyance path portion 23, a recording device 56 for recording a predetermined image related to the endorsement is disposed. The recording device 56 includes a print head, a stamp, and the like that can record a predetermined image in an appropriate direction at an appropriate position on the back surface 4b of the check 4 conveyed on the conveyance path 5.

Various sensors for check conveyance control are arranged in the conveyance path 5. A paper length detector 61 for detecting the length of the check 4 to be sent out is disposed at a position on the front side of the magnet 51. Between the back side contact image scanner 53 and the second conveyance roller 32, a double feed detector 62 for detecting that the check 4 is conveyed in an overlapped state is disposed. A jam detector 63 is arranged at a position on the front side of the fourth transport roller 34, and when the check 4 is continuously detected for a predetermined time or more by this detector, the jam is detected in the transport path 5. It can be seen that the check 4 is jammed. A print detector 64 for detecting the presence or absence of the check 4 endorsed by the recording device 56 is disposed at the position on the labor side of the fifth transport roller 35. Furthermore, it is discharged into these positions on the downstream side of the sixth transport roller 36, that is, the position of the branch path 9 that branches from the transport path 5 to the first check discharge section 7 and the second check discharge section 8. A discharge detector 65 for detecting the check 4 is arranged.
A switching plate 66 that is switched by a drive motor 67 (see FIG. 3) is disposed in the branch path 9. The switching plate 66 is a member for selectively switching the downstream end of the transport path 5 with respect to the first check discharge unit 7 and the second check discharge unit 8 and guiding the check 4 to the selected discharge unit.

FIG. 3 is a schematic block diagram showing a control system of the check reading apparatus 1.
The control unit 71 centrally controls each unit of the check reading device 1 under the control of the host-side control unit 73 of the host computer 70, and includes a CPU, ROM, RAM, and other peripheral circuits. Yes.
The control unit 71 drives the check delivery motor 14 and the conveyance motor 37 under the control of the host-side control unit 73 of the host computer 70 to send the checks 4 (see FIG. 1) one by one to the conveyance path 5. The sent check 4 is transported along the transport path 5. The conveyance control of the check 4 by the control unit 71 is based on detection signals from the paper length detector 61, the double feed detector 62, the jam detector 63, the print detector 64, and the discharge detector 65 arranged in the conveyance path 5. Based on.
As the check 4 is transported, the front-side contact image scanner 52 and the back-side contact image scanner 53 respectively read the image on the front side and the back side of the check 4 transported on the transport path 5, and show the read images. The image data is output to the control unit 71. The control unit 71 outputs these image data to the host side control unit 73.
Further, the magnetic head 54 detects an electromotive force generated by a change in the magnetic field formed by the passing magnetic ink character string 4A (see FIG. 1) under the control of the control unit 71, and a signal processing circuit 74 as a detection signal. Output to.
The signal processing circuit 74 includes an amplifier, a noise removal filter circuit, an A / D converter, and the like, amplifies the detection signal input from the magnetic head 54, shapes the waveform, and outputs the amplified signal to the control unit 71 as data. To do. The control unit 71 transmits data indicating the detection signal input from the signal processing circuit 74 to the host side control unit 73.
The operation unit 75 includes various switches such as a power switch and operation switches formed on the main body case 2 (see FIG. 1), detects user operations on these switches, and outputs them to the control unit 71.

A host computer 70 is connected to the check reading device 1 via a communication cable 72.
The host computer 70 includes a host-side control unit 73 that includes a CPU, ROM, RAM, and other peripheral circuits. The host control unit 73 includes an optical character recognition unit 80.
The host-side control unit 73 stores a display 76 capable of displaying various types of information, an operation unit 77 to which input devices such as a keyboard and a mouse are connected, and a rewritable memory such as an EEPROM or a hard disk. Part 78 is connected.
The storage unit 78 stores the front image of the check 4 input from the check reading device 1, data indicating the back image, and data indicating the detection signal.
In the present embodiment, the control unit 71 controls each unit of the check reading device 1 under the control of the host-side control unit 73 of the host computer 70. Specifically, the host-side control unit 73 executes a program stored in the ROM by the CPU to generate control data for controlling the control unit 71, and uses the generated control data for the check reading device 1. By outputting to the control unit 71, each part of the check reading device 1 is controlled. In this embodiment, the host computer 70 and the check reading device 1 cooperate to function as an image processing device.

As described above, the check reading apparatus 1 according to the present embodiment includes the magnetic head 54 that magnetically reads the magnetic ink character string 4 </ b> A and the surface-side contact image scanner 52 that optically reads the surface of the check 4. Yes. Then, the optical character recognition unit 80 included in the host-side control unit 73 of the host computer 70 optically recognizes each of the magnetic ink characters constituting the magnetic ink character string 4A based on the reading result of the front-side contact image scanner 52. To recognize.
A magnetic ink character is a character magnetically printed on a check 4 in accordance with a predetermined font, for example, CMC-7, and one magnetic ink character is a character of a plurality of predetermined character types. It corresponds to any one character type. Recognizing the magnetic ink character is to determine the character type of each magnetic ink character or to detect that the character type of the magnetic ink character cannot be determined for each of the read magnetic ink characters.
The optical character recognition by the optical character recognition unit 80 may be performed together with the magnetic ink character recognition, or may be performed when the magnetic ink character recognition fails.
Hereinafter, the operation of the optical character recognition unit 80 will be described in detail.

FIG. 4 is a flowchart showing the operation of the optical character recognition unit 80.
The function of the optical character recognition unit 80 is realized by the cooperation of hardware and software, such as the CPU of the host-side control unit 73 executing a program stored in the ROM.
In the following description, the font of magnetic ink characters recorded on the check 4 is CMC-7. In CMC-7, shapes corresponding to the character types of 41 characters in total including 10 numbers, 26 uppercase letters, and 5 special symbols are prepared as shapes of magnetic ink characters.
Further, in the following description, it is assumed that the operation of the optical character recognition unit 80 when executing recognition of the magnetic ink character string 4A recorded on the check 4 for a certain check 4 is shown. Then, it is assumed that the image on the surface of the check 4 is read by the surface-side contact image scanner 52 and is output from the controller 71 to the host-side controller 73 as surface image data.

Referring to FIG. 4, the optical character recognition unit 80 acquires surface image data relating to the surface of the check 4 (step SA1).
Next, the optical character recognition unit 80 cuts out a region corresponding to the character string corresponding region from the surface image data, and generates character string corresponding region image data (step SA2). Here, the character string corresponding region is a region defined as a region on the check 4 where the magnetic ink character string 4A is printed according to the standard. Therefore, although an image showing the magnetic ink character string 4A is always included in the character corresponding area data, in order to recognize the magnetic ink character string 4A with high accuracy, the magnetic ink character string 4A in the character corresponding area data is The region of the image to be shown needs to be specified to some extent strictly.
Next, the optical character recognition unit 80 binarizes the character string corresponding area image data (step SA3). Thereby, in the character string corresponding region image data, each pixel arranged in a dot matrix is in a state of holding either a value indicating black or a value indicating white.

FIG. 5 is a diagram schematically illustrating an example of a state in which the binarized character string corresponding region image data is developed in a predetermined coordinate system.
Since the character string corresponding area image data is data in which each pixel is arranged in a dot matrix, the coordinates of each pixel are defined as the origin in the coordinate system by developing the image data in the coordinate system. It is uniquely defined by its relative position from
Referring to FIG. 5, the image indicated by the character string corresponding region image data includes an image P1 indicating the magnetic ink character string 4A. The image P1 related to the magnetic ink character string 4A is configured by arranging images P2 indicating a plurality of magnetic ink characters in the longitudinal direction (Y-axis direction) of the magnetic ink character string 4A. Further, the image indicated by the character string corresponding area image data includes an image P3 indicating a magnetic ink character string 4A, an image P3 related to a sign, an image P4 related to dirt, and the like.
In the present embodiment, it is assumed that the positional relationship between the X-axis and Y-axis in the predetermined coordinate system and the character string corresponding area image data is as shown in FIG. In other words, the character string corresponding area image data indicates that, in the first quadrant of the coordinate system, the longitudinal direction of the magnetic ink character string 4A coincides with the direction in which the Y axis extends, and the width direction of the magnetic ink character string 4A is X It is assumed that the coordinate system is expanded so as to coincide with the direction in which the axis extends.

  After binarizing the character string corresponding area image data, the optical character recognition unit 80 performs a labeling process (step SA4).

FIG. 6 is a flowchart showing the operation of the optical character recognition unit 80 during the execution of the labeling process in step SA4.
In the labeling process (step SA4) shown in FIG. 6, first, the optical character recognition unit 80 detects a block included in the image data from the character string corresponding area image data (step SB1).
More specifically, the optical character recognition unit 80 first scans each pixel constituting the character string corresponding area image data and holds a value indicating black (hereinafter referred to as “black pixel”. Similarly, it indicates white. The pixel holding the value is called “white pixel”). When black pixels are detected, the optical character recognition unit 80 detects all black pixels that are continuous with the black pixels. All black pixels continuous with the black pixel are all black pixels adjacent to each other without being interrupted by white pixels. For example, in FIG. It is a block of black pixels shown. A group of continuous black pixels detected in this way corresponds to one “block”. The optical character recognition unit 80 scans each pixel constituting the character string corresponding area image data and detects all the blocks included in the image indicated by the character string corresponding area image data.
Here, each image of the magnetic ink character corresponding to each character type is composed of a block of one continuous black pixel. The minimum value and the maximum value of the number of pixels constituting the mass are uniquely determined for the mass of black pixels that can be an image of the magnetic ink character and can be a part of the image of the magnetic ink character.
Based on this, the optical character recognizing unit 80 determines whether the number of pixels constituting the cluster out of the continuous black pixel group included in the image indicated by the character string corresponding region image data is less than the minimum value, or If it exceeds the maximum value, the block is not “block”. Accordingly, for example, referring to FIG. 5, the block A1 in which the number of pixels falls below the minimum value and the pool A2 in which the number of pixels exceeds the maximum value are not detected as “blocks”.
By detecting (specifying) a block in this way, it is possible to exclude a block related to a block of black pixels that has no possibility of forming an image showing a magnetic ink character from the detected (specified) block ( Step SB2).

After detecting the block (chunk A) included in the character string corresponding region image data, the optical character recognition unit 80 selects a block to be subjected to the processing of the following steps SB3 to SB7 from among a plurality of blocks. Identify. In the following description, the block identified in step SB2 is referred to as a “processing target block”. The following processing from step SB3 to step SB7 is executed once for each detected block.
After identifying the processing target block from the detected block, the optical character recognition unit 80 detects the block frame of the processing target block (step SB3).
Specifically, the block frame is a rectangular frame that surrounds the processing target block, and includes two sides extending in the X-axis direction and two sides extending in the Y-axis direction. The length of the two sides extending in the X-axis direction of the block frame is defined corresponding to the length of the block in the X-axis direction, and the length of the two sides extending in the Y-axis direction of the block frame is Y of the block It is defined corresponding to the axial length. With reference to FIG. 5, for example, a frame indicated by a symbol BW corresponds to a block frame of a corresponding block.
Note that detecting the block frame means detecting information indicating the position of the block frame in the coordinate system (for example, the respective coordinates of the four corners of the block frame).

Next, the optical character recognition unit 80 determines whether there is a possibility that the detected block frame is a character frame of one magnetic ink character (step SB4).
More specifically, the character frame of one magnetic ink character is a frame that encloses an image showing the magnetic ink character of the character string corresponding area image data in the same manner as the block frame.
Here, as described above, 41 types of characters are prepared in CMC-7, but for the character frames of the images of the 41 types of character types, the side of the side extending in the X-axis direction of the character frame is displayed. The ratio of the vertical side length, the horizontal side length of the side extending in the Y-axis direction, and the vertical side length of the side extending in the X-axis direction to the horizontal side length of the side extending in the Y-axis direction (hereinafter referred to as “aspect ratio”) , Respectively. Accordingly, when the block frame is a character frame of magnetic ink characters, the range of the vertical side length of the side extending in the X-axis direction, the range of the horizontal side length of the side extending in the Y-axis direction, and the vertical and horizontal The range of the ratio is determined. Based on this, in step SB4, the vertical side length of the side extending in the X-axis direction, the horizontal side length of the side extending in the Y-axis direction, and the aspect ratio of the block frame of the processing target block are detected. It is discriminated whether or not it falls within each range.

When it is determined in step SB4 that the block frame may correspond to the character frame (step SB4: YES), the optical character recognition unit 80 uses the block frame as a candidate character frame KM (KM1, KM2, see FIG. 5). ), And the process proceeds to step SA8.
In step SA8, it is determined whether or not all the blocks detected in step SA1 are processing target blocks. If all the blocks are not processing target blocks (step SB8: NO), the optical character recognition unit 80 Then, the process procedure is shifted to step SB2, and when all the blocks are processed blocks (step SB8: YES), the labeling process (step SA4) is ended.

On the other hand, when it is determined in step SB4 that the block frame is not likely to correspond to the character frame (step SB4: NO), the processing target block is integrated into the blocks existing adjacent to the processing target block. It is determined whether there is a possible block (step SB6).
More specifically, if there is no possibility that the block frame of the processing target block corresponds to a character frame, the processing target block is not a block corresponding to one magnetic ink character alone, but a plurality of components constituting a control character. There is a possibility that the block corresponds to any one of the parts. When the processing target block is a block corresponding to any of a plurality of parts constituting the control character, the control character is configured around the processing target block and adjacent to the processing target block. This means that there is a block. The block that can be integrated with the processing target block is another block that is close to the processing target block to the extent that one control character can be configured. In step SB6, it can be integrated with the processing target block. It is determined whether or not a correct block exists.
In the following description, associating a block corresponding to one part constituting one control character with a block corresponding to another part is expressed as “integration”.

In step SB6, when there is no block that can be integrated with the processing target block (step SB6: NO), the optical character recognition unit 80 performs the processing procedure without specifying the block frame of the processing target block as the candidate character frame KM. The process proceeds to SB8. This is because the block to be processed is not a block related to the magnetic ink character, and is not a block related to a part constituting the magnetic ink character related to the control character.
On the other hand, in step SB6, when there is a block that can be integrated with the processing target block (step SB6: YES), the optical character recognition unit 80 executes integration processing (step SB7).

  After detecting the block group (integrated block) corresponding to one control character including the processing target block by the integration process in step SB7, the optical character recognition unit 80 uses the block frame of the entire integrated block as a candidate character frame. After specifying (step SB5), the process proceeds to step SB8 described above.

In FIG. 5, the candidate character frame detected by executing the labeling process (step SA4) is indicated by reference numeral KM. As shown in FIG. 5, the candidate character frame KM is an appropriate one surrounding the block corresponding to each part constituting the control character with respect to the control character.
On the other hand, in the labeling process, as long as there is a possibility that the block frame of one block may become a character frame of magnetic ink characters due to the length of the side and the aspect ratio, for example, as shown in the candidate character frame KM2. Even a block frame that does not normally correspond to one magnetic ink character is identified as a candidate character frame.

  In the labeling process (step SA4) shown in FIG. 4, after detecting the candidate character frame, the optical character recognition unit 80 performs a bandwidth detection process (step SA5).

FIG. 7 is a flowchart showing the operation of the optical character recognition unit 80 when the bandwidth detection process is executed.
Referring to FIG. 7, in the bandwidth detection process (step SA5), first, optical character recognition unit 80 executes a predetermined process after step SC2 from the candidate character frames detected in the labeling process (step SA4). A candidate character frame to be processed is specified (step SC1). That is, the predetermined processing after step SC2 is performed once for each candidate character frame. In the following description, the candidate character frame specified in step SC1 is referred to as a “processing target candidate character frame”.
Next, the optical character recognition unit 80 performs optical character recognition on a block corresponding to the processing target candidate character frame (a block surrounded by the processing target candidate character frame) (step SC2).
Optical character recognition is a process of determining the character type of the image indicated by the block or determining that the character type of the image indicated by the block cannot be determined. For example, a bitmap pattern corresponding to each of the 14 character types And the image data cut out by the processing target candidate character frame.
Next, the optical character recognition unit 80 determines whether or not the optical character recognition of the block corresponding to the processing target candidate character frame has succeeded (the character type of the image indicated by the block is determined) (step SC3).
If successful (step SC3: YES), the optical character recognizing unit 80 specifies the processing target candidate character frame as a confirmed character frame (step SC4), and moves the processing procedure to step SC5. That is, the fixed character frame is a character frame of an image related to a magnetic ink character that has been successfully recognized by optical characters.
On the other hand, when the optical character recognition fails (step SC3: NO), the optical character recognition unit 80 shifts the processing procedure to step SC5 without specifying the processing target candidate character frame as the confirmed character frame.
In step SC5, it is determined whether or not all of the candidate character frames detected in the labeling process are processing target candidate character frames. If all of the candidate character frames are not processing target candidate character frames (step SC5: NO), optical The character recognizing unit 80 moves the processing procedure to step SC1. On the other hand, when all of the candidate character frames become processing target candidate character frames (step SC5: YES), the processing procedure moves to step SC6.

FIG. 8 is a diagram illustrating a state in which a confirmed character frame is displayed on the character string corresponding region image data.
In FIG. 8, the frame indicated by the symbol DM corresponds to the confirmed character frame.
As shown in FIG. 8, a fixed character frame DM is detected for each of the images P2 related to the magnetic ink characters of the magnetic ink character string 4A. On the other hand, as shown in the images P5 and P6, the block formed by continuously and integrally combining the magnetic ink characters and the stains fails as a result of optical character recognition in step SC2 ( Step SC3: NO), the corresponding fixed character frame DM is not detected.

In step SC6, the optical character recognition unit 80 determines whether there are more than a predetermined number of characters that have been determined to have been successfully recognized in step SC3.
When the predetermined number or more exists (step SC6: YES), the optical character recognition unit 80 detects the lower end value of the band (step SC7).

In the following description, for convenience of explanation, of the four sides of the confirmed character frame DM, the two sides extending in the Y-axis direction are defined as the “lower end of the confirmed character frame DM” as the side close to the Y axis, The side far from the position is expressed as “the upper end of the confirmed character frame DM”.
In step SC7, the optical character recognizing unit 80 specifies the confirmed character frame DM in which the lower end 4f of the confirmed character frame DM is located closest to the Y axis.
Then, the optical character recognition unit 80 detects the value in the X-axis direction of the lower end 4f of the specified confirmed character frame DM, and sets the detected value as the band lower end value.

Next, the optical character recognition unit 80 detects the band upper end value (step SC8).
More specifically, the optical character recognition unit 80 identifies a confirmed character frame DM having an upper end 4e having the largest separation distance from the Y axis among the confirmed character frames DM.
Then, the optical character recognition unit 80 detects the value in the X-axis direction of the upper end 4e of the identified fixed character frame DM, and sets the detected value as the band upper end value.

Next, the optical character recognition unit 80 determines whether or not there is a logical contradiction between the band lower limit value detected in step SC7 and the band upper limit value detected in step SC8 (step SC9).
The logical contradiction is, for example, that the band lower end value is larger than the band upper end value, and that the difference between the band upper end value and the band lower end value is significantly smaller or larger than an assumed value. .
If there is no contradiction (step SC9: NO), the optical character recognizing unit 80, on the X axis of the predetermined coordinate system, sets a range obtained by adding a slight margin to a range sandwiched between the band lower end value and the band upper end value. It detects as a bandwidth (step SC10), cuts out character corresponding area image data with the detected bandwidth, and generates character string image data (step SC11).

FIG. 9 is a diagram showing the character string image data generated in step SC11.
As shown in FIG. 9, by detecting the bandwidth by the method described above and generating character string image data by cutting out the character string corresponding region image data based on the detected bandwidth as shown in FIG. The character string image data is data cut out in a region corresponding to the magnetic ink character string 4A corresponding to the width of the magnetic ink character string 4A. As a result, it is possible to eliminate as much as possible the image related to the dirt existing around the magnetic ink character string 4A from the character string image data, and it is particularly clear in the comparison between the images P5 and P6 in FIGS. As described above, the image related to the stain formed continuously and integrally with the image related to the magnetic ink character is also excluded from the character string image data.

In order to ensure the reliability of the detected bandwidth value when the band upper limit value and the band lower limit value are calculated using the detected fixed character frame and the band width is detected based on these values. Therefore, it is required that a certain number of fixed character frames as a premise of hand width detection exist. Based on this, in step SC6 (see FIG. 7), is the number of blocks successfully recognized (= number of confirmed character frames detected) equal to or greater than a predetermined number that can ensure the reliability of the bandwidth? If the number is less than the predetermined number, the optical character recognition unit 80 performs a preliminary bandwidth detection process (step SC12) for detecting the bandwidth by another method.
If there is a logical contradiction between the detected band upper limit value and the band lower limit value (step SC9), the bandwidth cannot be detected, so the optical character recognition unit 80 performs the preliminary bandwidth detection process ( Step SC12) is executed.

After detecting the bandwidth by the preliminary bandwidth detection processing (step SC12), the optical character recognition unit 80 cuts out the character string corresponding region image data with the detected bandwidth and generates character string image data (step SC11). ).
Thus, in this embodiment, in step SC6, it is determined that the number of blocks that have been successfully recognized has fallen below a predetermined number, or there is a logical contradiction between the calculated band upper limit value and band lower limit value. Even in this case, the bandwidth is detected by the spare bandwidth detection process.

After the bandwidth is detected by the bandwidth detection process (see FIG. 4) in step SA5 and character string image data (see FIG. 9) is generated, the optical character recognition unit 80 executes the optical character recognition process (step SA6). ).
In this optical character recognition process, the optical character recognition unit 80 performs optical character recognition, such as detection of a block that may be an image of a magnetic ink character, character recognition by pattern matching for each block, based on character string image data. A series of such processing is executed, and optical recognition of the magnetic ink character included in the magnetic ink character string 4A is executed again.

FIG. 10 is a schematic diagram showing an image and a confirmed character frame DM relating to magnetic ink characters.
CMC-7, which is one of the fonts of magnetic ink characters (MICR characters), has 10 numbers (“0” to “9”), 26 capital letters (“A” to “Z”), and special symbol 5 A total of 41 characters. The CMC-7 characters are printed and recorded on a recording medium, and can be discriminated magnetically. The CMC-7 character has seven vertical bars Rn (n is a natural number from 1 to 7) arranged at a plurality of types of intervals.
In the character frame (determined character frame DM) of the image of each of the 41 CMC-7 characters, the length of the side (vertical side length) extending in the X-axis direction (vertical axis direction) of the character frame (determined character frame DM), The length of the side (horizontal side length) extending in the Y-axis direction (horizontal axis direction) and the ratio of the vertical side length to the horizontal side length (hereinafter referred to as “aspect ratio”) are respectively determined. Then, by comparing the length of the vertical side with the size of a plurality of types of characters of CMC-7, the character height H, which is the length of the magnetic ink character in the X-axis direction, is defined.

As shown in FIG. 10, the CMC-7 character is displayed by the black pixel region BG, and the black pixel region BG is arranged in the X-axis direction and is configured by arranging seven vertical bars Rn formed in a bar shape. ing. The seven vertical bars Rn are arranged in the Y-axis direction and are configured by combinations of the vertical bars Rn. In FIG. 10, for convenience, seven columns of vertical bars Rn are illustrated as vertical bars R1, R2, R3, R4, R5, R6, and R7 from the left side of the figure.
The combinations of the vertical bars Rn define, for example, an interval between the vertical bar Rn and the vertical bar R (n + 1) (white pixel region WG, hatched portion in the figure and / or a white portion surrounded by a broken line), and 7 columns. One of the vertical bars Rn is defined by being divided into a plurality of parts in the X-axis direction, and is displayed with a black pixel region BG and a white pixel region WG (a shaded portion and / or a white portion surrounded by a broken line in the drawing). In this way, CMC-7 characters are formed.

A character height H of the CMC-7 character is a pixel row formed by applying a plurality of pixels G in the Y-axis direction by applying a pixel G having a height (H / 16) that divides the character height H into 16 parts. L is configured by arranging 16 rows continuous in the X-axis direction.
A character frame (determined character frame DM) surrounding the CMC-7 character is configured by arranging a plurality of pixel rows L in succession.
Here, as described above, the size (dimension) of the pixel G in the X-axis direction applies the pixel G that divides the character height H into 16, but the present invention is not limited to this. It is determined appropriately depending on the resolution of the character recognition device. In the present embodiment, the size of the pixel G in the Y axis direction is described as a configuration in which the size of the vertical bar Rn in the Y axis direction is the same as the size of one pixel G in the Y axis direction. However, the size of the vertical bar Rn in the Y-axis direction may be configured by a plurality of pixels G.

Since each pixel row L arranged in the X-axis direction intersects (orthogonalizes) seven columns of vertical bars Rn arranged in the Y-axis direction, a black pixel region BG displaying CMC-7 characters, A white pixel region WG (a white portion surrounded by a hatched portion and / or a broken line in the drawing) that displays an interval between the bar Rn and the vertical bar R (n + 1). For this reason, when the pixel row L constitutes a magnetic ink character having a character height H, a maximum of seven black pixel regions BG are arranged in the pixel row L, and a white pixel is between the black pixel region BG and the black pixel region BG. A pixel region WG (a shaded portion and / or a white portion surrounded by a broken line) is arranged.
On the other hand, when the pixel row L does not constitute a magnetic ink character having a character height H, the pixel row L is constituted only by the white pixel region WG (white portion surrounded by a broken line) or a continuous black pixel region BG or It is composed of a small number of black pixel regions BG and white pixel regions WG (the shaded portion in the figure and / or the white portion surrounded by a broken line).

  FIG. 11 is a schematic block diagram showing a part of the control system of the optical character recognition unit 80. FIG. 12 is a flowchart showing the operation of the optical character recognition unit 80 when the optical character recognition process in step SA6 is executed.

  As shown in FIG. 11, the optical character recognition unit 80 includes a Y-axis direction position detection unit 81 and an X-axis direction position detection unit 82. The X-axis direction position detection unit 82 includes a quantity calculation unit 83, a cumulative quantity calculation unit 84, an X-axis direction position determination unit 85, and a pattern matching processing unit 86.

Referring to FIG. 12, in the optical character recognition process (step SA6), first, in the Y-axis direction position detection unit 81 of the optical character recognition unit 80, the character string image data (FIG. 5) generated by the bandwidth detection process (step SA5). 9), a Y-axis direction position detecting step for detecting the position of the magnetic ink character in the Y-axis direction is performed (S101).
Specifically, the position of the magnetic ink character in the Y-axis direction is detected by detecting the black pixel region BG from the plus (+) side or the minus (−) side in the Y-axis direction of the character string corresponding region.

Next, the X-axis direction position detection unit 82 determines the position in the X-axis direction of the magnetic ink character having the character height H indicated by the solid-line double-ended arrow and the dashed-dotted double-ended arrow in FIG. An X-axis direction position detecting step for detecting is performed (S102).
First, in the X-axis direction position detection step (S102), white as a region between the vertical bar Rn and the vertical bar R (n + 1) in the pixel row L in the Y-axis direction per pixel G in the fixed character frame DM. A quantity calculating step for calculating the quantity NW of the pixel region WG is performed by the quantity calculating unit 83 (S103). For example, the quantity NW (NWa, NWb,..., NWv, NWw) of the white pixel area WG for each pixel row L (La, Lb,..., Lv, Lw) shown in FIG. 10 is calculated.
Here, the quantity NW of the white pixel area WG is not limited to calculating the quantity of the white pixel area WG (the white part surrounded by the hatched portion and / or the broken line in the drawing) one by one. In the case where a plurality of white portions surrounded by a broken line are connected adjacently in the Y-axis direction of the pixel row L, the quantity NW of the white pixel regions WG is calculated as one calculation unit, and the white pixel region WG (in broken lines) In the case where a plurality of enclosed white portions) are not connected adjacently in the Y-axis direction of the pixel row L and are independent, the number NW of the white pixel regions WG is calculated as 1.

Then, in 16 pixel rows L arranged continuously in the X-axis direction, the quantity NW of the white pixel region WG is accumulated to accumulate the quantity per character height H (length in the X-axis direction) of the magnetic ink characters. The cumulative quantity calculation step for calculating NWH is performed by the cumulative quantity calculation unit 84 (S104). Thereby, the cumulative quantity NWH of the white pixel area WG per character height H is calculated. In this way, the cumulative quantity NWH obtained by accumulating the quantity of the white pixel area WG as the area between the bar Rn and the vertical bar R (n + 1) is the character height H (length in the X-axis direction) of the magnetic ink character. The cumulative quantity NWH per unit is shown.
More specifically, as shown in FIG. 10, by shifting the character height H by one pixel G (one pixel row L) in the X-axis direction along the length of the vertical side of the confirmed character frame DM, the character of the magnetic ink character is displayed. The accumulated quantity NWH (NWHa, NWHb,..., NWHv, NWHw) of the white pixel region WG per high H is calculated.
For example, the cumulative quantity NWH (NWHb) is calculated by accumulating the quantity NW of the white pixel region WG from the 16 pixel lines Lb to the pixel line Lq arranged continuously in the X-axis direction. Thereby, the cumulative quantity NWHb of the white pixel region WG per character height H of the magnetic ink character is calculated. Similarly, in the pixel row Lc to the pixel row Lr, the pixel row Ld to the pixel row Ls, the pixel row Le to the pixel row Lt, the pixel row Lf to the pixel row Lu, and the pixel row Lg to the pixel row Lv, the cumulative quantity NWHc, The cumulative quantity NWHd, the cumulative quantity NWHHe, the cumulative quantity NWHf, and the cumulative quantity NWHg are calculated. As shown in FIG. 10, at the position of the character height H indicated by the solid-line double-ended arrow and the dashed-dotted double-ended arrow, the accumulated quantity NWH is accumulated quantity NWHb = 30, accumulated quantity NWHc = 36, accumulated quantity NWHd = 41, accumulated quantity, respectively. NWHe = 47, cumulative quantity NWHf = 46, cumulative quantity NWHg = 44.

Further, by comparing the cumulative quantity NWH of the white pixel area WG per character height H, an X-axis direction position determining step for determining the position in the X-axis direction of the magnetic ink character having the character height H is determined. It implements in the part 85 (S105).
As described above, when the pixel row L constitutes a magnetic ink character having a character height H, seven black pixel regions BG are arranged in the pixel row L at a maximum, and between the black pixel region BG and the black pixel region BG. Since the white pixel region WG (the hatched portion in the figure) is arranged, the quantity NW of the white pixel region WG when the pixel row L constitutes a magnetic ink character with a character height H is the magnetic quantity with the pixel row L being a character height H. Larger than when ink characters are not configured.
Therefore, the position in the X-axis direction of the character height H at which the cumulative quantity NWH of the white pixel area WG per character height H is maximum is determined as the position of the magnetic ink character with the character height H. In FIG. 10, the cumulative quantity NWHe = 47 from the pixel row Le to the pixel row Lt is the maximum, and the position indicated by the solid-line double-ended arrow is determined as the position of the magnetic ink character having the character height H.

More specifically, when the character height H matches the position of the magnetic ink character in the X-axis direction, in other words, when the character height H is the position indicated by the solid-line double-ended arrow, the accumulated white pixel area WG per character height H is accumulated. The quantity NWH (cumulative quantity NWHe = 47) is the maximum.
On the other hand, when the character height H does not coincide with the position of the magnetic ink character in the X-axis direction, in other words, when the character height H is the position indicated by the double-dotted arrow, the accumulation of the white pixel area WG per character height H is accumulated. The quantity NWH is smaller than the case where the quantity NWH matches the position of the magnetic ink character in the X-axis direction.
For example, if the cumulative quantity NWHb = 30, the cumulative quantity NWHc = 36, and the cumulative quantity NWHd = 41, the quantity NWb = 0 and NWc = 1 of the white pixel areas WG of the pixel rows La, Lb, and Lc that do not constitute the magnetic ink character. , NWd = 0 are added (calculated), respectively, and the numbers NWr = 6, NWs = 6, and NWt = 6 of the white pixel regions WG of the pixel rows Lr, Ls, and Lt constituting the magnetic ink character are respectively subtracted ( Calculated). For this reason, when the magnetic ink character matches the position in the X-axis direction, the cumulative quantity NWH of the white pixel region WG per character height H is the maximum.

  A pattern matching processing step for executing the pattern matching processing of magnetic ink characters is performed by the pattern matching processing unit 86 (S106). Thereby, recognition of magnetic ink characters is realized.

  According to this embodiment, the cumulative quantity NWH obtained by accumulating the quantity of the white pixel area WG as the area between the vertical bar Rn and the vertical bar R (n + 1), in other words, the character height H (X-axis direction) of the magnetic ink character. The X-axis direction position of the magnetic ink character with the character height H is determined by accurately comparing the accumulated quantity NWH per length of the magnetic ink character, and the magnetic ink character accurately recognizes the X-axis direction position of the magnetic ink character. Character pattern matching processing can be executed. This causes dirt or a signature to be recognized as a part of the magnetic ink character, causing erroneous recognition of the magnetic ink character, and repeatedly performing pattern matching processing of the magnetic ink character until it is accurately recognized. There is no need. In this way, it is possible to provide an image processing apparatus that can realize recognition of magnetic ink characters that suppresses erroneous recognition of magnetic ink characters and accelerates recognition.

  In the position determination step in the X-axis direction, the cumulative quantity NWH obtained by accumulating the quantity of the white pixel area WG as an area between the vertical bar Rn and the vertical bar R (n + 1), in other words, the character height H ( By comparing the cumulative quantity NWH per length in the X-axis direction, the position in the X-axis direction of the magnetic ink character with the character height H is determined, and the position in the X-axis direction of the magnetic ink character is accurately recognized. Thus, pattern matching processing of magnetic ink characters can be executed. This causes dirt or a signature to be recognized as a part of the magnetic ink character, causing erroneous recognition of the magnetic ink character, and repeatedly performing pattern matching processing of the magnetic ink character until it is accurately recognized. There is no need. In this way, it is possible to provide a control method for an image processing apparatus that can realize recognition of magnetic ink characters that suppresses erroneous recognition of magnetic ink characters and accelerates recognition.

(Second Embodiment)
The second embodiment of the present invention will be described below with reference to the drawings.
The check reading device 1 of the second embodiment and the control method of the check reading device 1 have the same configuration as that of the first embodiment shown in FIGS. 1 to 9 and 11, but are shown in FIGS. 10 and 12. The optical character recognition process of the control method of the check reading device 1 is different. For this reason, about the same structure, the same code | symbol is provided and description is abbreviate | omitted. Hereinafter, with reference to FIG. 13 and FIG. 14, the optical character recognition process of the control method of the check reading apparatus 1 different from the first embodiment will be described.

FIG. 13 is a schematic diagram showing an image and a confirmed character frame DM relating to magnetic ink characters.
CMC-7, which is one of the fonts of magnetic ink characters (MICR characters), has 10 numbers (“0” to “9”), 26 capital letters (“A” to “Z”), and special symbol 5 A total of 41 characters. The CMC-7 characters are printed and recorded on a recording medium, and can be discriminated magnetically. The CMC-7 character has seven vertical bars Rn (n is a natural number from 1 to 7) arranged at a plurality of types of intervals.
In the character frame (determined character frame DM) of the image of each of the 41 CMC-7 characters, the length of the side (vertical side length) extending in the X-axis direction (vertical axis direction) of the character frame (determined character frame DM), The length of the side (horizontal side length) extending in the Y-axis direction (horizontal axis direction) and the ratio of the vertical side length to the horizontal side length (hereinafter referred to as “aspect ratio”) are respectively determined. And the character height H is prescribed | regulated by comparing the length of a vertical side and the magnitude | size of the multiple types of character of CMC-7.

As shown in FIG. 13, the CMC-7 character is displayed by the black pixel region BG, and the black pixel region BG is arranged in the X-axis direction and is configured by arranging seven columns of vertical bars Rn formed in a bar shape. ing. The seven vertical bars Rn are arranged in the Y-axis direction and are configured by combinations of the vertical bars Rn. In FIG. 13, for convenience, seven rows of vertical bars Rn are illustrated as vertical bars R1, R2, R3, R4, R5, R6, and R7 from the left side of the figure.
As the combinations of the vertical bars Rn, for example, the interval between the vertical bar Rn and the vertical bar R (n + 1) (white pixel region WG, white portion surrounded by a broken line in the figure) is defined. This is defined by being divided into a plurality of parts in the X-axis direction, and is displayed by a black pixel region BG and a white pixel region WG (a white portion surrounded by a broken line in the drawing). In this way, CMC-7 characters are formed.

A character height H of the CMC-7 character is a pixel row formed by applying a plurality of pixels G in the Y-axis direction by applying a pixel G having a height (H / 16) that divides the character height H into 16 parts. L is configured by arranging 16 rows continuous in the X-axis direction.
A character frame (determined character frame DM) surrounding the CMC-7 character is configured by arranging a plurality of pixel rows L in succession.
Here, as described above, the size (dimension) of the pixel G in the X-axis direction applies the pixel G that divides the character height H into 16, but the present invention is not limited to this. It is determined appropriately depending on the resolution of the character recognition device. In the present embodiment, the size of the pixel G in the Y axis direction is described as a configuration in which the size of the vertical bar Rn in the Y axis direction is the same as the size of one pixel G in the Y axis direction. However, the size of the vertical bar Rn in the Y-axis direction may be configured by a plurality of pixels G.

Since each pixel row L arranged in the X-axis direction intersects (orthogonalizes) seven columns of vertical bars Rn arranged in the Y-axis direction, a black pixel region BG displaying CMC-7 characters, A white pixel region WG (a white portion surrounded by a broken line in the drawing) that displays an interval between the bar Rn and the vertical bar R (n + 1). For this reason, when the pixel row L constitutes a magnetic ink character having a character height H, a maximum of seven black pixel regions BG are arranged in the pixel row L, and a white pixel is between the black pixel region BG and the black pixel region BG. A pixel region WG (white portion surrounded by a broken line in the drawing) is arranged.
On the other hand, when the pixel row L does not constitute a magnetic ink character having a character height H, the pixel row L is constituted only by the white pixel region WG (white portion surrounded by a broken line) or a continuous black pixel region BG or It is composed of a small number of black pixel regions BG and white pixel regions WG (white portions surrounded by broken lines in the drawing).

  FIG. 14 is a flowchart showing the operation of the optical character recognition unit 80 when the optical character recognition process in step SA6 is executed.

Referring to FIG. 14, in the optical character recognition process (step SA6), first, as in the first embodiment, the Y-axis direction position detection unit 81 of the optical character recognition unit 80 performs the bandwidth detection process (step SA5). A Y-axis direction position detecting step for detecting the position of the magnetic ink character in the Y-axis direction from the generated character string image data (see FIG. 9) is performed (S101).
Specifically, the position of the magnetic ink character in the Y-axis direction is detected by detecting the black pixel region BG from the plus (+) side or the minus (−) side in the Y-axis direction of the character string corresponding region.

Next, an X-axis direction position detecting step for detecting the position in the X-axis direction of the magnetic ink character having the character height H indicated by the solid-line double-ended arrow and the dashed-dotted double-ended arrow in FIG. 13 from the detected fixed character frame DM is performed. Implement (S112).
First, in the fixed character frame DM, the quantity calculation unit 83 performs a quantity calculation step for calculating the quantity NB of the black pixel area BG as the area of the vertical bar R in the pixel row L in the Y-axis direction per pixel G ( S113). For example, the quantity NB (NBa, NBb,..., NBv, NBw) of the black pixel region BG for each pixel row L (La, Lb,..., Lv, Lw) shown in FIG. 13 is calculated.

Then, the cumulative quantity NBH per character height H (length in the X-axis direction) of the magnetic ink character is accumulated by accumulating the quantity NB of the black pixel regions BG in the 16 pixel rows L arranged continuously in the X-axis direction. Is calculated. As a result, the cumulative quantity calculation step for calculating the cumulative quantity NBH of the black pixel region BG per character height H is performed by the cumulative quantity calculation unit 84 (S114).
More specifically, as shown in FIG. 13, by shifting the character height H by one pixel G (one pixel row L) in the X-axis direction along the length of the vertical side of the confirmed character frame DM, the character of the magnetic ink character is displayed. The cumulative quantity NBH (NBHa, NBHb,..., NBHv, NBHw) of the black pixel region BG per high H is calculated.
For example, the cumulative quantity NBH (NBHb) is calculated by accumulating the quantity NW of the black pixel region BG in 16 pixel lines Lb to Lq arranged continuously in the X-axis direction. Thereby, the cumulative quantity NBHb of the black pixel region BG per character height H of the magnetic ink character is calculated. Similarly, in the pixel row Lc to the pixel row Lr, the pixel row Ld to the pixel row Ls, the pixel row Le to the pixel row Lt, the pixel row Lf to the pixel row Lu, and the pixel row Lg to the pixel row Lv, the cumulative quantity NBHc, Cumulative quantity NBHd, cumulative quantity NBHe, cumulative quantity NBHf, and cumulative quantity NBHg are calculated. As shown in FIG. 13, at the position of the character height H indicated by the solid-line double-ended arrow and the dashed-dotted double-ended arrow, the accumulated quantity NBH is accumulated quantity NBHb = 45, accumulated quantity NBHc = 52, accumulated quantity NBHd = 57, accumulated quantity, respectively. NBHe = 63, cumulative quantity NBHf = 61, cumulative quantity NBHg = 58.

Further, by comparing the accumulated quantity NBH of the black pixel region BG per character height H, an X-axis direction position determining step for determining the position in the X-axis direction of the magnetic ink character having the character height H is determined. It implements in the part 85 (S115).
As in the first embodiment, when the pixel row L constitutes a magnetic ink character having a character height H, a maximum of seven black pixel regions BG are arranged in the pixel row L, and the black pixel region BG and the black pixel region BG Since the white pixel region WG (the hatched portion in the drawing) is arranged between the black pixel region BG when the pixel row L constitutes a magnetic ink character having a character height H, the pixel row L has a character height. It is larger than the case where H magnetic ink characters are not formed.
For this reason, the position in the X-axis direction of the character height H at which the cumulative quantity NBH of the black pixel region BG per character height H is maximum is determined as the position of the magnetic ink character with the character height H. In FIG. 13, the cumulative quantity NBHe = 63 from the pixel row Le to the pixel row Lt is the maximum, and the position indicated by the solid-line double-ended arrow is determined as the position of the magnetic ink character with the character height H.

More specifically, when the character height H matches the position in the X-axis direction of the magnetic ink character, in other words, when the character height H is the position indicated by the solid-line double-ended arrow, the black pixel region BG per character height H is accumulated. The quantity NBH (cumulative quantity NBHe = 63) is the maximum.
On the other hand, when the character height H does not match the position of the magnetic ink character in the X-axis direction, in other words, when the character height H is the position indicated by the dashed-dotted double-ended arrow, the black pixel region BG per character height H is accumulated. The quantity NBH is smaller than the case where the quantity NBH matches the position of the magnetic ink character in the X-axis direction.
For example, if the cumulative quantity NBHb = 45, the cumulative quantity NBHc = 52, and the cumulative quantity NBHd = 57, the quantity NBb = 0, NBc = 2 of the black pixel areas BG of the pixel rows La, Lb, Lc that do not constitute the magnetic ink character. , NBd = 1 are added (calculated), respectively, and the quantities NBr = 7, NBs = 7, and NBt = 7 of the black pixel regions BG of the pixel rows Lr, Ls, and Lt constituting the magnetic ink character are respectively subtracted ( Calculated). For this reason, when the magnetic ink character matches the position in the X-axis direction, the cumulative quantity NBH of the black pixel region BG per character height H is the maximum.

  A pattern matching processing step for executing the pattern matching processing of magnetic ink characters is performed by the pattern matching processing unit 86 (S106). Thereby, recognition of magnetic ink characters is realized.

  In the present embodiment, as illustrated in FIG. 13, the size (dimension) of the vertical bar Rn in the Y-axis direction and the size of one pixel G in the Y-axis direction have been described as being the same. The size of the vertical bar Rn in the Y-axis direction may be configured by a plurality of pixels G. In that case, the quantity of the black pixel region BG is calculated. The number of black pixel regions BG is not calculated as the number of black pixel regions BG, but when the plurality of black pixel regions BG are adjacently connected, the number of black pixel regions BG is calculated as one.

  According to the present embodiment, by comparing the cumulative quantity NWH obtained by accumulating the quantity of the vertical bar area, in other words, the cumulative quantity NWH per character height H (length in the X-axis direction) of the magnetic ink character, the character height H The position of the magnetic ink character in the X-axis direction is determined, the position of the magnetic ink character in the X-axis direction can be accurately recognized, and the pattern matching process of the magnetic ink character can be executed. This causes dirt or a signature to be recognized as a part of the magnetic ink character, causing erroneous recognition of the magnetic ink character, and repeatedly performing pattern matching processing of the magnetic ink character until it is accurately recognized. There is no need. In this way, it is possible to provide an image processing apparatus that can realize recognition of magnetic ink characters that suppresses erroneous recognition of magnetic ink characters and accelerates recognition.

  Then, in the X-axis direction position determining step, by comparing the accumulated quantity NWH obtained by accumulating the quantity of the vertical bar area, in other words, the accumulated quantity NWH per character height H (length in the X-axis direction) of the magnetic ink character. The position of the magnetic ink character having the character height H in the X-axis direction is determined, and the position of the magnetic ink character in the X-axis direction can be accurately recognized to execute the magnetic ink character pattern matching process. This causes dirt or a signature to be recognized as a part of the magnetic ink character, causing erroneous recognition of the magnetic ink character, and repeatedly performing pattern matching processing of the magnetic ink character until it is accurately recognized. There is no need. In this way, it is possible to provide a control method for an image processing apparatus that can realize recognition of magnetic ink characters that suppresses erroneous recognition of magnetic ink characters and accelerates recognition.

  DESCRIPTION OF SYMBOLS 1 ... Check reading device, 2 ... Main body case, 3 ... Cover case, 4 ... Check, 4A ... Magnetic ink character string, 4a ... Front surface, 4b ... Back surface, 4d ... Rear end, 4e ... Upper end, 4f ... Lower end, 5 ... Conveyance path, 6 ... check supply unit, 7 ... first check discharge unit, 8 ... second check discharge unit, 8 ... second check discharge unit, 9 ... branch path, 10 ... check feed mechanism, 11 ... feed roller, 12 DESCRIPTION OF SYMBOLS ... Sending roller, 13 ... Retard roller, 14 ... Sending motor, 15 ... Hopper, 21 ... Upstream conveyance path part, 22 ... Curve conveyance path part, 23 ... Downstream conveyance path part, 30 ... Check conveyance mechanism, 31 ... 1st conveyance roller, 32 ... 2nd conveyance roller, 33 ... 3rd conveyance roller, 34 ... 4th conveyance roller, 35 ... 5th conveyance roller, 36 ... 6th conveyance roller, 37 ... Motor for conveyance, 41, 42, 4 , 44, 45, 46 ... pressure roller, 51 ... magnet, 52 ... front side contact image scanner, 53 ... back side contact image scanner, 54 ... magnetic head, 55 ... pressure roller, 56 ... recording device, 61 ... paper length detection 62 ... double feed detector, 63 ... jam detector, 64 ... print detector, 65 ... discharge detector, 66 ... switching plate, 67 ... drive motor, 70 ... host computer, 71 ... control unit, 72 ... communication Cable 73 73 Host side control unit 74 Signal processing circuit 75 Operation unit 76 Display unit 77 Operation unit 78 Storage unit 80 Optical character recognition unit 81 Y-axis direction position detection unit , 82 ... X-axis direction position detection unit, 83 ... Quantity calculation unit, 84 ... Cumulative quantity calculation unit, 85 ... X-axis direction position determination unit, 86 ... Pattern matching processing unit, A, A1, A2 ... Accumulation, KM, KM1, KM2 ... candidate character frames, P1, P2, P3, P4, P5, P6 ... image, R, R1, R2, R3, R4, R5, R6, R7, Rn, R (n + 1) ... vertical bar.

Claims (4)

An image processing apparatus for processing an image including magnetic ink characters in which a plurality of vertical bars extending in the X-axis direction are arranged at intervals in the Y-axis direction orthogonal to the X-axis direction,
An X-axis direction position detection unit that detects at least the position of the magnetic ink character in the X-axis direction,
The X-axis direction position detection unit is a quantity calculation unit that calculates the number of regions between the vertical bar and the other vertical bar in the Y-axis direction of the magnetic ink character;
A cumulative quantity calculation unit for accumulating the quantity and calculating a cumulative quantity per length of the magnetic ink characters in the X-axis direction;
An X-axis direction position determining unit that determines a position of the magnetic ink character in the X-axis direction based on the accumulated quantity;
An image processing apparatus comprising: a pattern matching processing unit that executes pattern matching processing of the magnetic ink characters. An image processing apparatus for processing an image including magnetic ink characters in which a plurality of vertical bars extending in the X-axis direction are arranged at intervals in the Y-axis direction orthogonal to the X-axis direction,
An X-axis direction position detection unit that detects at least the position of the magnetic ink character in the X-axis direction,
The X-axis direction position detection unit is a quantity calculation unit that calculates the number of the vertical bar regions in the Y-axis direction of the magnetic ink characters;
A cumulative quantity calculation unit for accumulating the quantity and calculating a cumulative quantity per length of the magnetic ink characters in the X-axis direction;
An X-axis direction position determining unit that determines a position of the magnetic ink character in the X-axis direction based on the accumulated quantity;
An image processing apparatus comprising: a pattern matching processing unit that executes pattern matching processing of the magnetic ink characters. This is a control method for an image processing apparatus for processing an image including magnetic ink characters in which a plurality of vertical bars extending in the X-axis direction are arranged at intervals in the Y-axis direction orthogonal to the X-axis direction. And
An X-axis direction position detecting step for detecting at least a position of the magnetic ink character in the X-axis direction,
The X-axis direction position detection step includes a quantity calculation step for calculating the quantity of the magnetic ink character in an area between the vertical bar and the other vertical bar in the Y-axis direction;
A cumulative quantity calculating step of calculating the cumulative quantity per length of the magnetic ink characters in the X-axis direction by accumulating the quantity;
An X-axis direction position determining step for determining a position of the magnetic ink character in the X-axis direction based on the accumulated quantity;
And a pattern matching processing step for performing pattern matching processing of the magnetic ink characters. This is a control method for an image processing apparatus for processing an image including magnetic ink characters in which a plurality of vertical bars extending in the X-axis direction are arranged at intervals in the Y-axis direction orthogonal to the X-axis direction. And
An X-axis direction position detecting step for detecting at least a position of the magnetic ink character in the X-axis direction,
The X-axis direction position detecting step includes a quantity calculating step for calculating the quantity of the vertical bar area in the Y-axis direction for the magnetic ink character;
A cumulative quantity calculating step of calculating the cumulative quantity per length of the magnetic ink characters in the X-axis direction by accumulating the quantity;
An X-axis direction position determining step for determining a position of the magnetic ink character in the X-axis direction based on the accumulated quantity;
And a pattern matching processing step for performing pattern matching processing of the magnetic ink characters.

Images