JP2001275000A - Character reader, character reading method and computer readable information storage medium with character reading program recorded therein - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a character reader, a character reading method and a computer readable information storage medium where a character reading program is recorded for binarizing the description included in each field of a slip under various conditions in terms of colors when the slip is read and binarized or recognized. SOLUTION: A slip is divided into pixels and fetched in a color by means of photoelectric transfer. At the same time, a point is designated in a color space and the projection value is calculated to binarization axis of every pixel that is fetched with a straight line passing the designated point and the original point of the color space defined as the binarization axis. Then the projection value is binarized by the prescribed threshold and the image obtained by this binarization is processed. Thus, the slip is binarized with a specific color standard by designating a point in the color space.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a character reading apparatus for optically reading a form or the like in which characters to be entered and characters other than the characters to be entered (such as a character frame) are described. The present invention relates to a character reading device, a character reading method, and a computer readable information storage medium on which a character reading program is recorded for efficiently reading a required portion of a form by binarizing the document using a reference to generate a binarized image.

[0002]

2. Description of the Related Art Japanese Patent Application Laid-Open No. 10-51655 discloses an example of an apparatus for recognizing a character required to read a form in color. The means described in the publication extracts lightness and chroma from values read in three primary colors (colors), and changes conditions for binarizing lightness based on these extracted values. Thereby, the character portion necessary for character recognition is binarized except for a chromatic portion such as a character frame which is not necessary for character recognition.

[0003] This means is a configuration for extracting a black character portion excluding a chromatic color portion, binarizing it, and recognizing characters.
For example, recognition of non-black characters, 2 including chromatic parts
It is not intended to acquire and recognize a quantified image.

[0004] By the way, when a form is read and binarized or recognized, it is necessary to binarize the description included in each field (binary or partial area to be recognized) on the form under various conditions. There is. For example, binarization of only handwritten black characters, binarization of only print characters (for example, chromatic colors), binarization of non-character portions such as character frames, and binarization of a combination of these. Further, even in the case of a single form, there are cases where it is desired to change this binarization or recognition condition for each field. By realizing these functions, it is possible to perform fine binarization in the selection of descriptions and the selection of fields, and the functionality as a character reading apparatus is remarkably improved.

[0005]

As described above, conventionally, when a form is read, binarized, or recognized, the description included in each field on the form is as follows.
It could not be binarized under various conditions in terms of color.
Further, in a single form, binarization or recognition conditions cannot be changed for each field.

According to the present invention, there is provided a character reading apparatus capable of binarizing a description contained in each field on a form under various conditions in a case where the form is read and binarized or recognized. It is an object of the present invention to provide a computer readable information storage medium on which a character reading method and a character reading program are recorded.

Further, a character reading device, a character reading method, and a computer readable information storage medium recording a character reading program capable of changing binarization or recognition conditions for each field in a single form are provided. The purpose is to do.

In addition, when a form is read and binarized or recognized, the result of binarizing various descriptions of the description included in each field on the form under various conditions is displayed by test. It is an object of the present invention to provide a character reading device, a character reading method, and a computer readable information storage medium on which a character reading program is recorded, which can confirm and reset the validity of a character.

Further, a character reading device and a character reading method which reliably capture the color of a portion not required for binarization or recognition, such as a character frame, and binarize or recognize a necessary portion with high reliability based on the information. It is another object of the present invention to provide a computer-readable information storage medium storing a character reading program.

It is another object of the present invention to provide a character reading device, a character reading method, and a computer readable information storage medium on which a character reading program is recorded, which can binarize characters and character frames. .

[0011]

In order to solve the above-mentioned problems, a character reading apparatus according to the present invention divides a form into pixels and takes in color by photoelectric conversion, and designates a point in a color space. Means for calculating a projection value of each of the captured pixels onto the binarization axis using a straight line passing through the origin of the color space and the designated point as a binarization axis; and And a means for binarizing the projected value with a predetermined threshold value, and a means for processing an image obtained by the binarization.

By specifying a point in the color space, the form is binarized based on a specific color standard.

Further, the point in the color space is designated for each field constituting the form, and the projection value is determined for a pixel belonging to each field among the captured pixels. It is characterized by the following.

Each field of the form is binarized by an appropriate specific color standard.

[0015] Further, the apparatus further comprises means for displaying a color image of the form taken in the color, and means for designating a specific position or area on the displayed color image, and The point is designated by referring to the color information of the pixel at the designated position or area.

By referring to the color information of the pixel at the specified position or area, a point on the color space for color reference can be specified more easily.

In addition, the point in the color space is specified for a plurality of points, and the projection value is obtained for each of a plurality of binarized axes obtained by specifying the plurality of points, and the image is obtained. The processing means includes a process of calculating a logical sum or a logical product of the plurality of images obtained by the binarization.

Dropping out a plurality of colors,
Alternatively, it is obtained as a binarized image.

Further, the apparatus further comprises means for displaying the image obtained by the binarization, and means for updating and specifying a point in the specified color space.

By displaying the image obtained by the binarization, the validity of the binarization standard can be confirmed. If this is not appropriate, update the specification of the point in the color space.

Further, the color space comprises hue, saturation, and brightness.

A point on a color space is designated in accordance with a human color sense.

Further, a character reading method and a computer-readable information storage medium storing a character reading program according to the present invention have steps for realizing the functions of the above-described units.

[0024]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a configuration diagram (FIG. 1A) showing the configuration of a character reading device according to an embodiment of the present invention, and an OCR (optical cc) provided in the character reading device.
FIG. 1B is a configuration diagram (FIG. 2B) showing the configuration of a character reader system.

As shown in FIG. 1A, a character reading apparatus according to this embodiment includes a scanner 1 and a personal computer (PC).
computer2), and inside the PC2,
There is an OCR system 3. The OCR system 3 includes a form definition program 5, an OCR reading program 4, an FC (format control) as shown in FIG.
And a table 6.

The scanner 1 optically reads a form under the control of the PC 2, converts the read information into an electric signal, and supplies the electric signal to the PC 2. The PC 2 is a hardware resource for the OCR system 3 to perform processing on the supplied read information. OC
The R system 3 processes user input information in addition to the supplied read information, and performs generation of a binarized image, character recognition, and the like.

The form definition program 5 sets the reading color conditions for each field of the form.
The set conditions are supplied to the FC table 6. The OCR reading program 4 refers to the contents of the FC table 6 and
A binarized image is generated and character recognition is performed using the supplied read information. FC table 6
Is a storage destination of the reading color condition for each field of the form and a reference destination for generating a binarized image and performing character recognition by combining with the other elements.

Next, the operation of this embodiment will be described.

First, a form to be read is placed on the scanner 1, and a color image is fetched by scanning, and the information of the form definition is set, that is, the color condition of reading is set. The form definition information setting screen is a PC
2 is displayed on the display.

FIG. 2 is a diagram showing an example of a form definition information setting screen. As shown in the figure, a color image 20 of a read form and a field information definition window 30 are displayed on the form definition information setting screen 10.

In the color image 20, there are a plurality of fields 21, 22, 23, and 24 to be read, and a field to be defined and input is selected from these fields. Is displayed. In this figure, field 21 is currently selected.

The field information definition window 30 has a field number 31, a read field selection reception display 32, an image field selection reception display 33, a field start point (start coordinate) display 34, and a field end point (end coordinate) display. 35, font display 36, character type display 37, special attribute display 38, binarization axis selection display 39, saturation display 40 of the selected binarization axis, hue display 41, brightness display 42, binarization Threshold value display 43, test binarization reception display 44
There is.

Here, the reading field selection acceptance display 32 is displayed.
, And a reading field is selected, the field is binarized based on a predetermined standard, and the binarized image is recognized as a character. On the other hand, when an image field is selected by inputting to the image field selection acceptance display 33, the field is binarized based on a predetermined criterion, but the character recognition process is not performed. That is, this choice
The choice is to process the field as a simple image or to recognize and process it as a character.

For the displays other than the read field selection reception display 32 and the image field selection reception display 33,
Hereinafter, description of the information setting of the form definition will be described.

The procedure for setting the form definition information will be described with reference to the flowchart of FIG.

First, as described above, an image of a form is fetched (step 81), and the image is displayed on a form definition information setting screen (step 82).

Next, a field is designated by a rectangle.
This specification can be made, for example, by specifying and dragging from the start point to the end point of the field on the displayed image. As a result, in the start point display 34 and the end point display 35 in FIG. 2, an actual numerical value is displayed in, for example, a unit of mm or dot in accordance with the designated and dragged position. Or, for example, start point display 3
4. A numerical value may be directly input to the end point display 35. In this case, the color image 20 is stored in accordance with the input numerical value.
A rectangle is displayed at the top.

Next, read information setting is performed (step 8).
4). Here, first, the font, character type, and special attribute are set (step 85). These are performed by inputting to the font display 36, character type display 37, and special attribute display 38 in FIG. Here, the font is, for example, a handwritten character,
Different types of print characters and character types are, for example, numbers, English characters, kana characters, kanji, etc., and special attributes are, for example, binaries and characters for each divided area by further dividing the field. Whether to recognize.

The reading information setting (step 84) further includes a procedure for determining a binarized axis (step 86). Determining a binarized axis refers to uniquely determining an axis that passes through the origin (black) and a coordinate point indicating a certain color in a color space. When this axis is determined, the value of the axis component can be calculated for points indicating all colors in the color space. That is, the value is an on-axis value indicated by a point projected from a point in the color space to the axis. In addition, by determining the binarization axis, a threshold for binarization on the axis is set. These pieces of information are stored in the FC table 6 described in FIG. 1 (Step 87).

For the determination of the binarization axis and the setting of the threshold value, for example, a dropout color (a color that should be ignored in binarization and recognition) is appropriately captured, and each color is ignored while ignoring the dropout color. This is important for highly reliable binarization or recognition. If the binarization axis is determined so as to pass through the coordinates of the dropout color in the color space, the presence or absence of the dropout color can be detected with maximum sensitivity, so that it is easier to distinguish the color other than the dropout color. is there. In the present invention, the determination of the binarization axis and the setting of the threshold can be performed for each field (steps 88 to 83). This makes it possible to set the most appropriate read information for processing the field.

Here, an example of a specific procedure for determining the binarized axis and setting the threshold will be described with reference to FIG. FIG. 9 is a flowchart showing a specific procedure for determining a binarization axis and setting a threshold.

As shown in the figure, in order to determine the binarized axis, first, a representative axis (here, for example, the R, G, and B axes of three primary colors are prepared) is selected. Whether or not to perform is selected (step 91). When a representative axis is selected (step 95), the binarization threshold for that axis is presented as a default, and the user can further change this value (step 96).

The selection of the binarization axis performs binarization and recognition from the viewpoint of the presence or absence of a typical color component. For example, when the dropout color is clearly reddish or greenish, it can be used simply. Although the binarization threshold is presented as a default for the convenience of the user, the user can change the threshold to a more appropriate threshold. This is done in step 96
, The test binarization execution (step 97) flows to the Y side,
Hereinafter, according to the procedure of step 98, step 99, step 91, step 95, and step 96, the binarization can be performed while checking the result of the binarization (step 98).

Note that these selections, presentations, and changes are performed as shown in FIG.
, R axis, G axis in the binarized axis selection display 39
This is reflected on each selection display of the B axis and the binarization threshold value display 43.

If the execution is selected in the test binarization execution selection step 97 from step 96 as described above (this is done by clicking the test binarization display 44 in the field information definition window 30 in FIG. 2). The test binarization is performed on the selected field using the selected binarization axis and the binarization threshold (step 98). This result is displayed on the screen of the PC 1 described with reference to FIG. For example, in the case of the field 21 of the color image 20 in FIG. 2, the result of the test binarization is, for example, as shown in FIG. If this result is acceptable, the determination of the binarized axis and the threshold setting routine are terminated. If the result is not satisfactory, the process is repeated from step 91.

On the other hand, if the representative axis is not used in the determination of the binarized axis, a screen for setting the binarized axis is displayed (step 92). A threshold is set.

That is, on the setting screen of the binarized axis, as described above, for example, the coordinates in the color space of the dropout color are designated and input. Here, the coordinates are coordinates in a color space composed of hue, saturation, and brightness (step 93), because this is easily perceived by humans.

When the coordinates in the color space are designated and input,
The binarization axis is uniquely determined. Thereby, the binarization threshold value in the axis is presented, and the user can change the value (step 94). The reason why the binarization threshold is presented first is to provide the user with convenience by presenting the default value first. As a specific method of presenting the default value, it is possible to refer to this table by preparing a table indicating the default threshold value on the designated binarized axis on the axis.
The change is performed from step 94 through step 97 to the Y side, and by the procedure of step 98, step 99, step 91, step 92, step 93, step 94,
This can be done while checking the result of the binarization (step 98).

These are the "other settings" display and the saturation display 4 in the binarized axis selection display 39 in FIG.
0, hue display 41, lightness display 42, and binarization threshold value display 43 are reflected.

Step 9 for selecting the execution of the test binarization
Steps 7 and below have been described in the case where the above-described representative axis is selected, and thus will not be described.

Next, an example of the binarized axis setting screen (step 92) appearing in the description of FIG. 6 will be described with reference to FIG. FIG. 3 is a diagram illustrating an example of the binarized axis setting screen.

As shown in FIG. 3, the binarized axis setting screen 51
Includes a color plane 52 having a hue 53 and a saturation 54 as a y-axis and an x-axis, respectively, and a display for indicating a lightness 58. That is, for example, the coordinates in the color space of the dropout color for determining the binarization axis are determined by the position of the cross pointer 55 on the color plane 52 for the hue 53 and the saturation 54, and the position pointer 57 for the lightness 58. Can be instructed by the height of each.

In this case, if the user sets these instructions without any clues, the burden is large and cannot be said to be practical. For the purpose of improvement, a case will be described as an example where the field 21 of the form color image 20 in the form definition information setting screen 10 shown in FIG. 2 is currently set.

If the handwritten description is to be binarized and recognized and the chromatic character frame is not to be binarized and recognized, any of the character frames in the field 21 is Specify the cursor with the part. As a result, the hue, saturation, and lightness of the part designated by the cursor are automatically reflected on the positions of the cross pointer 55 and the lightness indicator pointer 57 on the color plane 52 of the binarization axis setting screen 51. This allows
Hue, saturation, and lightness can be presented as defaults.

As another method, the hue, saturation,
The average of the lightness is obtained, and these values are displayed on the binarization axis setting screen 51.
Can be reflected on the positions of the cross pointer 55 and the lightness indication pointer 57 on the color plane 52. In this case, there is an advantage that the default value is not affected by some pixels.

Further, on the binarized axis setting screen 51, a plurality of colors may be specified on the color plane 52 of the hue 53 and the saturation 54. This effect
For example, when two colors are selected and designated, a binary image based on the first designation and a binarized image based on the second designation can be obtained for the field. By calculating the logical sum or logical product of two binarized images, it is possible to drop out two colors or obtain a binarized image. Similarly, for three or more selections / designations, the same effect can be obtained by increasing the colors.

Further, the hue 53 and the saturation 54 may be specified by specifying an area on the color plane 52. The designation in this case is performed, for example, as in a designation frame 56. The effect in this case is also essentially the same as the case where a plurality of colors can be designated. That is, when an area is specified, it can be considered that each color included in the area is specified, and a binary image can be obtained for each color specification. Therefore, by calculating a logical sum or a logical product of these binarized images, it is possible to drop out the color of the designated area or obtain a binarized image.

In this case, for all colors (colors corresponding to respective digital values) included in the specified area, 2
Obtaining the quantified image considerably increases the processing load, so that the processing may be performed assuming that the color is specified for the predetermined representative point included in the area.

When the setting of the form definition is completed as described above, the actual form reading process is performed based on this information.

[0061]

As described above in detail, according to the character reading device, the character reading method, and the computer readable storage medium on which the character reading program is recorded according to the present invention, a point in a color space is designated, A projection value of each pixel taken on a straight line passing through the origin of the space and the designated point as a binarization axis is calculated on the binarization axis, and the calculated projection value is binarized by a predetermined threshold value. , So that the form can be
A valuation is made. As a result, a process such as character recognition can be performed by obtaining an optimal binarized image for each form and each field while securing the degree of freedom of the user.

[Brief description of the drawings]

FIG. 1 is a configuration diagram (FIG. 1A) showing a configuration of a character reading device according to an embodiment of the present invention, and an OCR (optical character) provided in the character reading device.
FIG. 1 is a configuration diagram illustrating a configuration of an (er reader) system.

FIG. 2 is a diagram showing an example of a form definition information setting screen.

FIG. 3 is a diagram showing an example of a binarized axis setting screen.

FIG. 4 is a diagram showing an example of display of a result of test binarization.

FIG. 5 is a flowchart showing a procedure for setting information of a form definition.

FIG. 6 is a flowchart showing a specific procedure for determining a binarization axis and setting a threshold.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Scanner 2 PC 3 OCR system 4 OCR reading program 5 Form definition program 10 Form definition information setting screen 20 Color image 20 21, 22, 23, 24 Field 30 Field information definition window 31 Field number 32 Reading field selection acceptance display 33 Image field selection acceptance display 34 Field start point (start coordinate) display 35 Field end point (end coordinate) display 36 Font display 37 Character type display 38 Special attribute display 39 Binary axis selection display 40 Selected binarization Axis saturation display 41 Hue display of selected binarized axis 42 Lightness display of selected binarized axis 43 Binarization threshold display 44 Test binarization reception display 51 Binarization axis setting screen 52 Color plane 53 Hue 54 Saturation 55 Crosshair pointer 56 Designation frame 57 Position pointer 8 brightness

Claims (18)

[Claims] 1. A means for dividing a form into pixels and taking in color by photoelectric conversion, a means for specifying a point in a color space, and a straight line passing through the origin of the color space and the specified point.
Means for calculating a projection value of each of the pixels taken into the binarization axis as a value axis, means for binarizing the calculated projection value with a predetermined threshold value, Means for processing the obtained image. 2. Designation of a point in the color space is performed for each field constituting the form, and calculation of the projection value is performed for a pixel belonging to each field among the captured pixels. The character reading device according to claim 1, wherein: 3. The character reading device according to claim 1, further comprising: a display unit configured to display a color image of the form captured in the color, and a specific position or area on the displayed color image. 3. The method according to claim 1, further comprising a step of: specifying the point in the color space by referring to color information of a pixel at the specified position or area. Character reader. 4. Designation of points in the color space is performed for a plurality of points, and the projection value is obtained for each of a plurality of binarized axes obtained by designating the plurality of points. 4. The character reading device according to claim 1, wherein the processing unit includes a process of obtaining a logical sum or a logical product of the plurality of images obtained by the binarization. 5. The character reading device according to claim 1, further comprising: means for displaying an image obtained by the binarization; and updating a point on the designated color space. 5. The character reading device according to claim 1, further comprising a designation unit. 6. The character reading device according to claim 1, wherein the color space includes hue, saturation, and brightness. 7. A step of dividing a form into pixels and taking in color by photoelectric conversion, a step of designating a point in a color space, and forming a line passing through the origin of the color space and the designated point by 2
Calculating a projection value of each of the pixels taken into the binarization axis as a binarization axis, binarizing the calculated projection value with a predetermined threshold value, Processing the obtained image. 8. A point in the color space is specified for each field constituting the form, and the projection value is calculated for a pixel belonging to each field among the captured pixels. 8. The character reading method according to claim 7, wherein: 9. The character reading method according to claim 7, further comprising: displaying a color image of the form read in the color, and designating a specific position or area on the displayed color image. 9. The method according to claim 7, further comprising the step of: specifying the point in the color space by referring to color information of a pixel in the specified position or area. Character reading method. 10. Designation of a point in the color space is performed for a plurality of points, and the projection value is obtained for each of a plurality of binarized axes obtained by specifying the plurality of points. 10. The character reading apparatus according to claim 7, wherein the processing step includes a process of obtaining a logical sum or a logical product of the plurality of images obtained by the binarization. 11. The character reading method according to claim 7, further comprising: displaying an image obtained by the binarization; and updating a point in the designated color space. 11. The character reading method according to claim 7, further comprising: a designation step. 12. The color space according to claim 7, wherein the color space comprises hue, saturation, and lightness.
Character reading method described in section. 13. A step of dividing a form into pixels and taking in color by photoelectric conversion, a step of designating a point in a color space, and a step of forming a line passing through the origin of the color space and the designated point by 2
Calculating a projection value of each of the pixels taken into the binarization axis as a binarization axis, binarizing the calculated projection value with a predetermined threshold value, And a computer-readable information storage medium storing a character reading program for causing a computer to execute the step of processing the obtained image. 14. Designation of a point in the color space is performed for each field constituting the form, and calculation of the projection value is performed for a pixel belonging to each field among the captured pixels. A computer-readable information storage medium storing the character reading program according to claim 13. 15. A computer-readable information storage medium storing the character reading program according to claim 13 or 14, further comprising a step of displaying a color image of the form captured in the color. Specifying a specific position or area on the displayed color image, and specifying a point on the color space refers to color information of a pixel at the specified specific position or area. 15. A computer-readable information storage medium on which a character reading program according to claim 13 or 14 is recorded. 16. Designation of a point in the color space is performed for a plurality of points, and the projection value is determined for each of a plurality of binarized axes obtained by specifying the plurality of points. 14. The processing step according to claim 13, wherein the processing step includes a process of obtaining a logical sum or a logical product of the plurality of images obtained by the binarization.
A computer-readable information storage medium storing the character reading program according to any one of claims 15 to 15. 17. A computer-readable information storage medium on which the character reading program according to claim 13 is recorded, wherein the character reading program further comprises: an image obtained by the binarization. 17. The computer readable information recorded with the character reading program according to claim 13, further comprising a step of displaying, and a step of updating and specifying the point in the specified color space. Storage medium. 18. The computer-readable information storage medium according to claim 13, wherein the color space includes hue, saturation, and lightness.