JP2009152757A - Slip processor and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a slip processor and a program for effectively recognizing a mark, without using a dropout color. <P>SOLUTION: The slip processor is provided with: a read part for reading a slip on which a plurality of items are printed and an area for writing the mark indicating an item to be selected is provided as an image having a plurality of pixels; an extraction part for extracting first, second and third color groups for which the number of the pixels corresponding to the pixels are first, second and third largest respectively from the image inside the area; a color change part for changing the color of the pixels corresponding to the third color group, when a color difference between the second and third color groups is smaller than a prescribed reference color difference; and a recognition part for recognizing the mark separately from the item, on the basis of the image for which the color of the pixels corresponding to the third color group is changed. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a form processing apparatus and a program.

  A technique related to an optical character reader for reading a form is disclosed (for example, see Patent Document 1). Here, a mark such as a circle (“◯”) may be entered on the form. The item selected by the writer can be expressed by writing a mark on one or more of the plurality of items printed on the form. In this case, prints representing the contents of the items and marks entered with a writing instrument or the like exist on the form.

At this time, by setting the print color of the form as a dropout color, it is possible to distinguish the mark entered from the print of the item content. If the print color is the dropout color, the item print is excluded from the reading target and only the mark is read.
JP 11-42309 A

However, when the dropout color is used, the color cannot be used for reading, and the color used for the form is restricted.
In view of the above, an object of the present invention is to provide a form processing apparatus and program that enable effective recognition of a mark without using a dropout color.

  A form processing apparatus according to an aspect of the present invention includes a reading unit that reads a form having an area in which a plurality of items are printed and a mark indicating an item to be selected is written as an image having a plurality of pixels, and the area An extraction unit for extracting the first, second, and third color groups having the first, second, and third largest numbers of pixels corresponding to the pixels from the image, and the second, second, A color changing unit that changes a color of a pixel corresponding to the third color group and a pixel corresponding to the third color group when a color difference between the three color groups is smaller than a predetermined reference color difference; And a recognition unit for recognizing a mark separately from the item based on an image whose color is changed.

  According to the present invention, it is possible to provide a form processing apparatus and program capable of effectively recognizing a mark without using a dropout color.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
FIG. 1 is a block diagram showing a form processing apparatus 100 according to an embodiment of the present invention.
The form processing apparatus 100 is an apparatus for reading a form, in particular, a mark written on the form, and includes a reading apparatus 110, an apparatus main body 120, a storage unit 130, a display unit 140, and an input unit 150.

FIG. 2 shows a part of the form.
Mark areas A1 and A2 are arranged on the form. The mark areas A1 and A2 are areas in which a mark (for example, “◯” (circle mark)) is written by a writing tool (for example, a pencil or a ballpoint pen) (that is, an item on which a plurality of items are printed and selected. This is an area in which a mark indicating is written.

  In the mark areas A1 and A2, items to be selected and marks to be displayed at the time of selection are displayed. In the mark area A1, four items “2 years (24 times)”, “3 years (36 times)”, “4 years (48 times)”, and “5 years (60 times)” are printed. . In the mark area A2, four items “all ages”, “21 years old or older”, “26 years old or older”, and “30 years old or older” are printed. A broken-line circle is printed to indicate a mark (reference mark) to be entered.

In the mark areas A1 and A2, marks M1 and M2 are entered in the items "2 years (24 times)" and "21 years old or older", respectively. The printed dashed circle is hidden under the marked marks M1 and M2.
In the present embodiment, the marks M1 and M2 are read separately from the item contents and the reference marks printed so as to overlap them.

The reading device 110 is a device that optically reads a form, for example, a scanner, and includes a form holding unit 111, a form extraction unit 112, and a form reading unit 113.
The apparatus main body 120 includes a reading control unit 121, a mark area determination unit 122, a color extraction unit 123, a color difference determination unit 124, a color change unit 125, a mark recognition unit 126, and a recognition result correction unit 127.
The storage unit 130 is an auxiliary storage device such as a hard disk or a semiconductor memory, and includes an image holding unit 131, a recognition result holding unit 132, and a correction result holding unit 133.

The form holding unit 111 holds a plurality of forms. The form is taken out from the form holding unit 111, read, and returned.
The form takeout unit 112 takes out the form from the form holding unit 111 and sends it to the form reading unit 113, and returns the read form to the form holding unit 111.
The form reading unit 113 optically reads a form, and functions as a reading unit that reads the form as an image having a plurality of pixels.

  The form reading unit 113 includes a light source, a linear sensor, and a scanning drive unit. The light source irradiates the form with light. In a linear sensor, a photoelectric conversion element such as a CCD is arranged one-dimensionally, and photoelectrically converts three primary colors (for example, R (red), G (green), and B (blue)) in reflected light from a form. , Output as an electrical signal of luminance data of multiple gradations (for example, 256 gradations (8 bits)). The scan driver scans the linear sensor on the form. By continuously scanning the linear sensor on the form, two-dimensional image data of the form can be obtained. This image data is composed of a large number of pixel data. Each pixel data has RGB luminances Lr, Lg, and Lb. That is, the form reading unit 113 reads the form as an image (image) composed of a large number of pixels having colors.

  The color here refers to a combination of RGB luminances Lr, Lg, and Lb, and includes both chromatic and achromatic colors. Further, even if the ratios of the luminances Lr, Lg, and Lb are the same, if the absolute values of the luminances Lr, Lg, and Lb are different, the colors are different. That is, the color here is broader than the color in chromatics.

  The apparatus main body 120 can be configured by a computer and a program (for example, an OS (operating system), an application program). A CPU (Central Processing Unit) in the computer operates as a program, so that the computer functions as the reading control unit 121 and the like.

  The reading control unit 121 controls the reading device 110 to read a form as an image and cause the image holding unit 131 to hold the read image data. The image includes a plurality of pixels each having a color (RGB luminance Lr, Lg, Lb).

  The mark area determination unit 122 determines the mark areas A1 and A2 from the form image data. The mark area determination unit 122 has a table T1 for this determination. In this table T1, for example, distances (relative distances) between the reference position of the form and the four corners of the mark areas A1 and A2 are represented. The reference position is appropriately defined on the form (for example, the upper left corner and the lower right corner of the form). The reference position may be one, but it is preferable that the reference position is two or more in consideration of the possibility that the form is read in a tilted state.

  The color extraction unit 123 extracts a plurality of color groups from the images in the mark areas A1 and A2, and functions as an extraction unit that extracts a plurality of color groups corresponding to pixels from the image. Specifically, the color groups C1 to C3 are extracted in descending order of the corresponding number of pixels (area ratio on the form). These color groups C1 to C3 are considered to be the ground color of the form (paper color), the color of the printed item, and the color of the mark entered. When the color group C3 does not substantially exist (error range), it is considered that no mark is entered.

  The color difference determination unit 124 determines the magnitude of the color difference ΔC between the color groups C2 and C3. Specifically, the color difference determination unit 124 determines whether or not the color difference ΔC is larger than the reference color difference ΔC0.

  The color changing unit 125 changes the color of the pixel corresponding to the color group C3 based on the determination result of the color difference determining unit 124. If the color difference ΔC between the color groups C2 and C3 is smaller than the reference color difference ΔC0, the color of the pixel corresponding to the color group C3 is changed. If the color difference ΔC is greater than or equal to the reference color difference ΔC0, the pixel color is not changed.

  The mark recognition unit 126 recognizes which item in the mark areas A1 and A2 is marked, and holds the recognition result in the recognition result holding unit 132. The recognition unit recognizes the mark separately from the item. Function as. The mark recognition unit 126 has a table T2 for this recognition. In this table T2, for example, the relative positions of items in the mark areas A1 and A2 are represented.

  The recognition result correction unit 127 corrects the recognition result in the mark recognition unit 126 based on the operation of the operator (operator), and holds the correction result in the correction result holding unit 133. For this correction, the display unit 140 and the input unit 150 are used.

The image holding unit 131 holds image data of the form read by the form reading unit 113.
The recognition result holding unit 132 holds the recognition result from the mark recognition unit 126.
The correction result holding unit 133 holds the correction result from the recognition result correction unit 127.

The display unit 140 is a display device that displays an image, for example, a CRT or a liquid crystal display device.
The input unit 150 is an input device for inputting information, for example, a keyboard and a mouse.

(Operation procedure of the form processing apparatus 100)
FIG. 3 is a flowchart showing an example of the operation procedure of the form processing apparatus 100.
(1) Reading of form (step S11)
When the reading control unit 121 controls the reading device 110, the form is optically read and held in the image holding unit 131 as image (image) data. This image is represented by a plurality of pixels each having a color. This color is represented by a combination of RGB luminances Lr, Lg, and Lb.

(2) Determination of mark area (step S12)
The mark area determination unit 122 determines the mark areas A1 and A2 of the form by using the form image data and the table T1. Specifically, the reference position of the form is read from the form image, and the mark areas A1 and A2 in the image are determined based on the relative distance from the reference position.

(3) Extraction of color groups C1 to C3 from the mark area (step S13)
The color extraction unit 123 extracts color groups C1 to C3 from the mark areas A1 and A2.
FIG. 4 is a graph showing the luminance distribution of R, G, and B in the mark area. The horizontal and vertical axes of the graph are the brightness and appearance frequency, respectively. Here, the ground color of the paper (form), the print color (color of the printed item), and the color of the entered mark are distributed in correspondence with the appearance frequency (number of pixels) and the brightness of large, medium and small.

  FIG. 5 is a schematic diagram showing the colors of the pixels in the mark area on the RGB color space. The RGB color space is expressed by using three orthogonal axes as RGB luminances Lr, Lg, and Lb. The color of each pixel is represented as a point P on the RGB color space. This point P represents one color (one pixel). Color regions S1 to S3 each including color groups C1 to C3 as an accumulation of points P are arranged. Color groups C1 to C3 are extracted in descending order of the number of corresponding pixels. Note that the color groups C1 to C3 correspond to the order in which the area ratio on the form is large and the order in which the pixel density D is large on the color space.

Here, the color regions S1 to S3 are represented as a space on a cube having centers P1 (Lr1, Lg1, Lb1), P2 (Lr2, Lg2, Lb2), and P3 (Lr3, Lg3, Lb3).
Color region S1: (Lr1 ± ΔL1, Lg1 ± ΔL1, Lb1 ± ΔL1)
Color region S2: (Lr2 ± ΔL2, Lg2 ± ΔL2, Lb2 ± ΔL2)
Color region S3: (Lr3 ± ΔL3, Lg1 ± ΔL3, Lb3 ± ΔL3)
Here, the sizes (distances) of the color areas ΔL1 to ΔL3: the distances from the centers P1 to P3 of the color areas S1 to S3 to the respective outer circumferences

Specifically, the color groups C1 to C3 (color regions S1 to S3) can be selected as follows.
1) A color space is divided into a plurality of blocks (for example, luminance Lr, Lg, and Lb are each divided into a plurality of blocks (for example, a block divided into eight), and a block having the maximum number of points (pixels) inside is selected. To do.

  2) The center of the selected block is adjusted, and the center P1 (Lr1, Lg1, Lb1) that maximizes the number of pixels in the block is selected.

  3) The block size ΔL is adjusted, and the block size ΔL1 is selected based on the pixel density D in the block.

FIG. 6 is a graph showing an example of the relationship between the block size ΔL in the RGB color space and the pixel density D in the block. The density D is expressed by the number of pixels in the block / the volume of the block (2 · ΔL) ³ .

  The density D near the center P1 of the block is large (the density D is the maximum density Dmax near ΔL to 0), and the density D decreases as the distance from the center P1 increases (approaches the average density Dav in the entire color space). As the size ΔL1, the size ΔL that maximizes the amount of change in the density D is selected. Instead, a size ΔL at which the density D becomes the reference density D1 may be adopted. As the density D1, a value (Dmax / a) obtained by dividing the maximum density Dmax by a predetermined value a can be used.

  4) The block determined as described above is the color region S1. The color region S2 is obtained by performing the same processing except for the color region S1 and its vicinity from the color space.

  5) The color region S3 is obtained by performing the same processing except for the color regions S2 and S3 and the vicinity thereof from the color space.

(4) Determination of presence / absence of substantial existence of color group C3 (step S14)
It is determined whether or not the color group C3 is substantially present. Specifically, if the density of the pixels in the extracted color area S3 is equal to or higher than a predetermined reference density D0, it is determined that the extracted color area S3 substantially exists. On the other hand, if the density of the pixels in the extracted color region S3 is smaller than the predetermined reference density D0, the extracted color region S3 is a kind of noise (error range), and the color group C3 does not substantially exist ( The mark is not entered.

(5) Determination of the color difference ΔC between the color groups C2 and C3 (step S15)
A color difference ΔC between the color groups C2 and C3 is calculated. The color difference ΔC can be calculated as follows, for example.
ΔC = ((Lr2−Lr3) ² + (Lg2−Lg3) ² + (Lb2−Lb3) ² ) ^1/2
This color difference ΔC is compared with the reference color difference ΔC0 to determine whether or not it is necessary to change the color of the pixel corresponding to the color group C3. If the color difference ΔC is smaller than the reference color difference ΔC0, it is determined that the color of the pixel corresponding to the color group C3 needs to be changed.

(6) Changing the color of the pixel corresponding to the color group C3 (step S16)
If it is determined in step S15 that the colors of the pixels corresponding to the color group C3 need to be changed, the colors of these pixels are changed. Specifically, the color of the pixel in the color region S3 is changed to the color C31. The color C31 is appropriately selected so that a color difference from the color group C2 is ensured.

(7) Recognition of mark corresponding to color group C3 (step S17)
It is recognized to which item a mark composed of pixels corresponding to the color group C3 (or the changed color C31) is attached. Specifically, continuous pixels corresponding to the color group C3 (or color C31) are extracted, and the positions of the centers (centers of gravity) of these pixels, that is, the positions where marks are added, are recognized. Furthermore, by referring to the table T2, the item with the mark is determined from this position. This recognition result is held in the recognition result holding unit 132.

(8) Determination of presence / absence of processing in all mark areas (step S18)
It is determined whether or not processing has been executed in all mark areas. If there is an unprocessed mark area, steps S13 to S17 are repeatedly executed.

(9) Correction of mark recognition result (step S19)
The recognition result of the mark is corrected. Specifically, the form image and the mark recognition result are displayed on the display unit 140, and the operator determines whether there is a difference between the image and the recognition result. When there is a difference between the image and the recognition result, information for correcting the recognition result is input by the input unit 150. When the color of the pixel corresponding to the color group C3 is changed, it is preferable to display an image based on the changed color on the display unit 140.

  As described above, based on the difference in pixel color, the entered mark can be recognized separately from the printed items. In particular, (1) the mark can be recognized effectively even if the overlap between the item and the mark is large, and (2) even if the difference in color between the print and the mark is small. In cases such as (1) and (2), there is generally a possibility that the mark recognition rate, and thus the form reading process (entry) efficiency, will be reduced. In this embodiment, a decrease in the mark recognition rate is prevented by using the difference in pixel colors. As a result, it is possible to secure the degree of freedom of arrangement and color arrangement of items and marks on the form. In the case of (2), the mark can be recognized more reliably by changing the color of the mark on the image (enlarging the color difference).

(Other embodiments)
Embodiments of the present invention are not limited to the above-described embodiments, and can be expanded and modified. The expanded and modified embodiments are also included in the technical scope of the present invention.
In the above embodiment, a color group is extracted using the color of each pixel. Instead of this, it is also possible to use an average value of colors at a plurality of adjacent pixels.
In the above embodiment, the RGB color space is used as the color space. Instead, another color space, for example, an xyY color space may be used.

It is a block diagram showing the form processing apparatus which concerns on one Embodiment of this invention. It is a figure showing a part of form. It is a flowchart showing an example of the operation | movement procedure of a form processing apparatus. It is a graph showing the luminance distribution of each of R, G, and B in the mark area. It is the model which represented the color of the pixel in a mark area on RGB color space. It is a graph showing an example of the relationship between the size of the block on the RGB color space and the density of the pixels in the block.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 100 ... Form processing apparatus, 110 ... Reading apparatus, 111 ... Form holding part, 112 ... Form taking-out part, 113 ... Form reading part, 120 ... Main part of apparatus, 121 ... Reading control part, 122 ... Mark area determination part, 123 ... Color Extraction unit 124 ... color difference determination unit 125 ... color change unit 126 ... mark recognition unit 127 ... recognition result correction unit 130 ... storage unit 131 ... image holding unit 132 ... recognition result holding unit 133 ... correction result Holding unit, 140 ... display unit, 150 ... input unit

Claims (4)

A reading unit that reads a form having an area in which a plurality of items are printed and a mark representing an item to be selected is written as an image having a plurality of pixels;
An extraction unit for extracting the first, second, and third color groups having the first, second, and third largest numbers of pixels corresponding to the pixels from the image in the region;
A color changing unit that changes a color of a pixel corresponding to the third color group when a color difference between the second and third color groups is smaller than a predetermined reference color difference;
A recognition unit for recognizing a mark separately from the item based on an image in which the color of a pixel corresponding to the third color group is changed;
A form processing apparatus comprising: The form processing apparatus according to claim 1, wherein the first to third color groups respectively correspond to a background color of the form, a color of the printed item, and a color of a written mark. . A display unit for displaying a recognition result in the recognition unit;
An input unit for inputting information for correcting the recognition result;
A recognition result correction unit for correcting the recognition result based on the input;
The form processing apparatus according to claim 1, further comprising: A control unit for controlling a reading device that reads a form having an area in which a plurality of items are printed and a mark representing an item to be selected is written, as an image having a plurality of pixels;
An extraction unit for extracting the first, second, and third color groups having the first, second, and third largest numbers of pixels corresponding to the pixels from the image in the region;
A color changing unit that changes a color of a pixel corresponding to the third color group when a color difference between the second and third color groups is smaller than a predetermined reference color difference;
Based on an image in which the color of the pixel corresponding to the third color group is changed, as a recognition unit that recognizes a mark separately from the item,
A program that allows a computer to function.

Images