JP2002216119A - Image processing method, image processor and computer readable recording medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To calculate image data from which a specific color is deleted with a few error. SOLUTION: In the case of regarding a color to be distributed within prescribed distance from a straight line including a line segment and the internally dividing position of which to the line segment is distributed in a prescribed value or range as the specific color in a color space, when background color information is used as color information of the line segment, another information is used in place of the background color information when the range regarded as the specific color is judged to be narrow, specific color range distance is adaptively decided from color distribution information of an object image for processing and after that, when the decided range distance is judged to be inappropriate, the range distance is reset.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to an image processing method and apparatus, and more particularly, to an image processing method for generating or extracting an image in which a specific color is deleted.

When an image such as a mark sheet is processed in an objective test, a specific color component originally printed on the sheet is erased and then image processing is performed, so that processing for a specific purpose can be performed more easily and more reliably. Can be done.

For example, yellow is printed in advance as a background color on a mark sheet, and when a mark written by a subject is to be read on the sheet, when the binarization processing is performed as it is, the printing color is used. In some cases, the color component of the mark written by the subject is lost in a certain yellow component, which makes accurate reading difficult. In order to avoid such a situation, a process of removing the yellow component, which is a printing color, from the read image data in advance and leaving only the color component of the mark written by the subject is performed, so that an accurate mark Reading becomes possible.

[0004]

2. Description of the Related Art When a user wants to generate an image in which only a specific color is erased, a method of attaching a lamp of a special color to a scanner and dropping out the scanner when reading an original document has been generally used. . In the mark sheet of the above-mentioned objective test, the mark was read after erasing the printing color by this method. Here, it is assumed that all of the color ranges necessary for performing the above-described erasing process on the read image data are performed by software. In such a way,
Colors to be erased are given as a set of two colors, and a pixel located at a certain distance from a line segment connecting them in the color space is recognized as a specific color, and this is set as an object to be erased. Here, a method of using a background color as color information is conceivable. By using one of the colors to be erased by a human and the background color automatically estimated as the other, the human does not need to provide (input) two colors as a set, and only one color is sufficient. Therefore, the amount of input work is reduced for humans, which is preferable. In the above-mentioned erasing process, a method of adaptively setting a distance range (cylinder radius) from a line segment in the color space, which is a boundary of whether or not to be included in the erasing target, can be considered. A histogram of the distance from the line segment to each pixel is generated, and the distance range is automatically determined based on this information. Here, the distance range may be given manually, but if it can be automatically set, the input operation by a human can be reduced.

However, when the background color is used as the color information as described above, when the background color itself is to be erased, the color given by a human is close to the automatically estimated background color. The line segment to be defined in the space becomes very short, so that erasure may be incomplete. In the above-described method of adaptively setting the distance range from the line segment, the distance range is often too small, and in such a case, the erasure may be incomplete.

In order to solve the above-mentioned problems, the inventions according to the first, twelfth and twenty-third aspects are arranged such that the lines are distributed within a predetermined distance from a straight line including a line segment in a color space, This is an image processing method that regards a color in which an internal division position with respect to a minute is distributed in a predetermined value or range as a specific color. When it is determined, another information is used instead of the background color information.

[0005] With this configuration, it is possible to prevent the range regarded as a specific color from being narrowed, and to appropriately erase the specific color.

According to the second, thirteenth and twenty-fourth aspects of the present invention, when the distance between two end points forming a line segment is smaller than a predetermined value, the range considered as a specific color is narrow. It is characterized by making a judgment.

Further, the invention according to claims 3, 14 and 25 is characterized in that when it is determined that the range regarded as a specific color is narrow, white information is used instead of background color information.

[0008] Thus, the input operation by the user is unnecessary, and the range considered as the specific color can be appropriately determined.

The invention according to claims 4, 15 and 26 is characterized in that when it is determined that the range considered as a specific color is narrow, information specified by a user is used as background color information.

According to a fifth aspect of the present invention, in the color space, distribution is performed within a predetermined distance from a straight line including a line segment, and the position of the internal division with respect to the line segment is a predetermined value or An image processing method in which a color distributed in a range is regarded as a specific color, wherein a specific color range distance is adaptively determined from color distribution information of a processing target image, and then the determined range distance is determined to be inappropriate. In such a case, the range distance is reset.

With such a configuration, it is possible to prevent a distance (a distance from a line segment) in a range regarded as a specific color from being reduced, and to appropriately erase a specific color.

According to a sixth aspect of the present invention, the determination as to whether the determined range distance is appropriate is made when the determined range distance is smaller than a predetermined value. I do.

According to a seventh aspect of the present invention, the range distance is reset using a frequency histogram of a distance between a line segment and a pixel included in an image to be processed. Features.

With this configuration, it is possible to adaptively set the range distance in accordance with the characteristics of the image, thereby enabling erasing of a specific color more suitable for the characteristics of the image.

In the invention according to claims 8, 19, and 30, the resetting of the range distance is performed by comparing the frequency of the determined range distance with the frequency of all distances larger than the determined range distance. It is performed when there is no distance with a frequency exceeding the determined distance.

With this configuration, when the image is composed of similar colors, the range distance is reset, so that the range distance suitable for the feature of the image can be set.
It is possible to erase an appropriate specific color.

Further, the invention according to the ninth, twentieth and thirty-first aspects is characterized in that a value which is a predetermined multiple of the initially set distance is set as a new range distance.

Further, in the invention according to claims 10, 21 and 32, the minimum distance which is not less than the distance initially set in the histogram and whose differential value of the frequency is not negative is reset as the range distance.

The invention according to Claims 11, 22, 33, wherein the minimum distance which is a constant multiple of the initially set distance value and which is not less than the initially set distance in the histogram and whose differential value of the frequency is not negative. It is characterized in that the smaller one of the values is reset as the range distance.

[0020]

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

In the image processing, a case where an image in which a color designated by a user is deleted is generated will be described as an example. FIG. 1 shows a configuration for realizing the first embodiment of the present invention.

In the drawing, 1 is an image input means such as a scanner for inputting an image to be processed, 2 is an image storage means for storing input image data, and 3 is a user to designate and input a required color. A designated color storing means for storing information relating to a color to be erased, and a designated color storing means for storing a color to be replaced for the erased pixel after the pixel is erased. , 6 is a background color calculating means for estimating the background color from the processing target image, 7 is a background color storing means for storing information on the background color obtained therefrom,
Reference numeral 8 denotes a background color examination unit that determines whether the background color is suitable for processing, 9 is a background color replacement unit that replaces the background color with another color, and 10 is a position or the like of a cylinder described later in a color space. The cylinder determining means 11 for determining the size is a cylinder storing means for storing the determined cylinder information, and the cylinder storing means 12 is
Pixel color replacement means for replacing the color of an image based on the cylinder information will be described.

FIG. 2 is a flowchart showing the processing executed by the first embodiment. The operation of the first embodiment will be described below with reference to FIG. The image to be processed is input as a color image using the image input means 1 such as a scanner (step 101), and the image data thus input is stored in the image storage means 2. The user inputs a color to be erased using the color specifying means 3 (step 102), and stores information on the color thus input in the specified color storing means 4. In this case, two colors that can be regarded as the same color may be input, but here, only one typical color is input.
Next, the user uses the color designating means 3 to input a color to be replaced after erasing, and information on the color thus input is stored in the pixel color storing means after erasing (step 10).
3). This is a color to be replaced for a pixel to be erased. Here, a color close to the background color of the document is generally given. For example, if the document is printed on a white background, the designated color is erased by setting the color to be replaced to white. Next, the background color calculation means 6 calculates a background color (step 104). The result is stored in the background color storage means 7. There are various methods for obtaining the background color. For example, there is a method of creating a color histogram from the color information of the entire image and using the color having the highest frequency as the background color. Next, the background color examination means 8 determines the color to be erased input in step 102 and the color in step 1.
It is determined whether or not the background color calculated in step 04 is similar (step 105). If the two colors are similar, the cylinder for determining whether the color is a specific color is shown in FIG.
As shown by, the height becomes extremely small, and it is determined that sufficient erasing cannot be performed. If it is determined that the colors are similar, another consideration is required.

Such a situation in which the designated color and the calculated background color are similar often occurs, for example, when it is desired to erase the background color itself. Specifically, for example, when dark yellow is used for the background and information (marks, etc.) written on the yellow background becomes invisible if the read image data is binarized as it is, ,
If the background color is deleted (eg, replaced with white), binarization can be performed while maintaining the written information. in this case,
Although yellow is specified as the color to be erased, the calculated background color is also yellow, so that both colors are similar. There are various methods for determining whether or not both colors are similar, but here, it is similar to the case where the sum of squares of the difference between each of the RGB components of the color is equal to or less than a predetermined value. Is determined.

If the two are similar, the background color replacing means 9 replaces the background color estimated (calculated) in step 104 with the erased color given by the user in step 103 (step 106). (Here, in addition to the method of using the color input by the user for replacement, a predetermined color, for example, white, etc., may be used to automatically replace the color.) The cylinder determining means 10 is used in the color space. , The color to be erased given by the user in step 102 and step 104
A cylinder centering on the line connecting the background color calculated in the step is determined and stored in the cylinder storage means 11 (step 107).
The height (length in the axial direction) of the cylinder is set so that the bottom and top surfaces of the cylinder are located near the two points of the color to be erased and the background color, respectively. The radius of the cylinder uses a predetermined value here.

Even if the color to be erased is similar to the background color, the cylinder has a sufficient height because the background color has been replaced with another color in step 106. There should be.

Next, the pixel value replacing means 12 examines all the pixels in the image, deletes the one in the cylinder in the color space, and replaces it with the erased color given by the user in step 103 ( Steps 108 and 109).

When the examination of all pixels is completed, an image resulting from the processing is output and the processing is terminated (steps 110 and 11).
1). In the above-described first embodiment, the cylinder in the color space always has a fixed radius. Next, a method of automatically setting the radius (range distance) will be considered. In this case, depending on the condition, the radius is too small and the erasure becomes insufficient. Therefore, the cylinder radius is reexamined, and as a result, the cylinder radius is reset if necessary. Such a method will be described below.

FIG. 4 shows a configuration for realizing the second embodiment of the present invention. This configuration corresponds to the radius determining section of the cylinder determining means 10 in the first embodiment. FIG. 5 is a flowchart showing a process executed by the second embodiment, and corresponds to a portion for determining a radius in the step of determining a cylinder in step 107 in the first embodiment.

The distance histogram generating means 201 creates a distance histogram between the axis of the cylinder and each pixel in the image and stores it in the histogram storage means 202 (step 301).

Next, the radius calculating means 203 obtains a threshold value t for bisecting the histogram by the discriminant analysis method, and stores this in the radius storing means 204 (step 302). Where t
Is used as the radius of the cylinder. In this case, in the case of an image composed of only the same system colors, such as a black-and-white document and a document consisting of only blue and light blue, the pixel distribution in the color space tends to be concentrated only near the cylindrical axis. In that case, t becomes smaller than it should be. Therefore, in the present embodiment, the following processing is added.

First, the radius examination means 205 compares t with a predetermined value (step 303).
As a result, if t is larger (see the example of the histogram on the left side of the upper row in FIG. 6), it is determined that the above-described problem has not occurred, and t is directly used as the cylindrical radius (step 307). .

Similarly, the radius judging means 205 judges whether or not there exists j whose t <j and their frequency is f (t) <f (j) (step 304). When such j exists, there is a peak of the histogram inside and outside the cylinder as shown in the example of the histogram on the right side of the upper row in FIG. 6, and it can be determined that the above-described problem does not occur. So t
Is used as it is as the cylindrical radius (step 30).
7).

Next, if t is equal to or smaller than the predetermined value, the radius resetting means 206 determines the distance i from the cylinder to each pixel and the frequency f (i)
To find the minimum i that satisfies i > t and f (i) < f (i + 1) (step 305). FIG. 6 shows a histogram when such i exists.
(Example of lower histogram)

Here, if there is no such i, t
The radius is widened by setting a value obtained by multiplying by a certain coefficient k as the radius (step 309). If such an i exists, the i is compared with the above kt, and the smaller one is set as the radius (step 30).
8).

In the above-described processing, even if the radius is set to be narrow, the radius can be reset and the radius can be expanded appropriately, and as a result, it can be expected that appropriate erasing is realized.

FIG. 7 shows a configuration for realizing the third embodiment of the present invention. This embodiment is an embodiment in a case where the present invention is realized by software.
A general-purpose processing device (for example, a personal computer) including the PU 401, the memory 402, the hard disk 403, the input device 404, the CD-ROM drive 405, the display 406, and a mouse is used.

A recording medium 407 such as a CD-ROM includes
A program for realizing the image processing function and the processing procedure according to each embodiment of the present invention described above is recorded in advance. It is assumed that the original image to be processed is stored in, for example, the hard disk 403 or the like.

While appropriately using the memory 402 and the like, the CPU 401 reads out and executes a program for realizing the above-described processing functions and procedures from the recording medium 407, and outputs a result of the image processing realized as a result to the display 406 and the like. . The embodiments of the present invention are not limited to those described above, and it goes without saying that various applications can be considered within the scope described in the claims.

The present invention also includes the following modes.

An image processing method which regards a color which is distributed within a predetermined distance from a straight line including a line segment in a color space and whose internal division position with respect to the line segment is distributed in a predetermined value or range, as a specific color. When the background color information is used as the line segment color information, when it is determined that the range considered as the specific color is narrow, another color information is used instead of the background color information, and the color distribution of the image to be processed is An image processing method, wherein a specific color range distance is adaptively determined from information, and if the determined range distance is determined to be inappropriate, the range distance is reset.

The image processing method according to claim 1, wherein the determination as to whether or not the determined range distance is appropriate is performed when the determined range distance is smaller than a predetermined value. The image processing method according to claim 1, wherein the resetting of the range distance is performed using a frequency histogram of a distance related to a color distribution of an image to be processed. Each computer readable recording medium on which a program for realizing each of the above image processing methods is described.

An image processing apparatus which regards a color which is distributed within a predetermined distance from a straight line including a line segment in a color space and whose internal division position with respect to the line segment is distributed in a predetermined value or range, as a specific color. When the background color information is used as the line segment color information, when it is determined that the range considered as the specific color is narrow, another color information is used instead of the background color information, and the color distribution of the image to be processed is An image processing apparatus, wherein a specific color range distance is adaptively determined from information, and when it is determined that the determined range distance is inappropriate, the range distance is reset.

The image processing apparatus according to the present invention, wherein the determination as to whether the determined range distance is appropriate is made when the determined range distance is smaller than a predetermined value. The image processing apparatus according to claim 1, wherein the resetting of the range distance is performed using a frequency histogram of a distance relating to a color distribution of an image to be processed.

As described above, when background color information is used as color information of a line segment, when it is determined that the range considered as a specific color is narrow, another color information is used instead of the background color information. As a result, it is possible to prevent the range considered as a specific color from being narrowed, and to appropriately erase the specific color.

Further, when it is determined that the range regarded as a specific color is narrow, white information is used instead of the background color information, thereby making it unnecessary for the user to perform an input operation and being regarded as an appropriate specific color. Range can be determined.

After the specific color range distance is adaptively determined from the color distribution information of the image to be processed, if it is determined that the determined range distance is inappropriate, the range distance is reset. Distance within the range considered to be a specific color
(Distance from the line segment) can be prevented from being reduced, so that the specific color can be appropriately erased.

Since the range distance is reset using a frequency histogram of the distance between the line segment and the pixel included in the processing target image, the range distance is adaptively adjusted according to the characteristics of the image. Can be set, and a specific color that is more suitable for the characteristics of the image can be erased.

As described above, since the color desired by the user can be specified with high accuracy, color erasure with few errors can be performed.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a first exemplary embodiment of the present invention.

FIG. 2 is a flowchart showing an operation of the image processing apparatus shown in FIG.

FIG. 3 is a diagram showing replacement of a background color and a cylinder in a color space.

FIG. 4 is a block diagram showing a configuration of a second exemplary embodiment of the present invention.

FIG. 5 is a flowchart showing an operation of a radius determining unit of the cylinder determining means shown in FIG.

FIG. 6 is a diagram illustrating an example of a histogram of distances of pixels included in a processing target image from a cylinder axis of a cylinder in a color space as illustrated in FIG. 3;

FIG. 7 is a diagram illustrating a configuration of a third exemplary embodiment of the present invention.

[Explanation of symbols]

 6: Background color calculation means 8; Background color examination means 9: Background color replacement means 10: Cylinder determination means 12: Pixel replacement means 201; Distance histogram generation means 205: Radius examination means 206: Radius resetting means

Continued on front page F-term (reference)

Claims (33)

[Claims] 1. An image processing method in which a color distributed within a predetermined distance from a straight line including a line segment in a color space, and an inner segment position with respect to the line segment being distributed in a predetermined value or range is regarded as a specific color. Image processing characterized by using background color information as line segment color information, and using another color information instead of the background color information when it is determined that the range considered as a specific color is narrow. Method. 2. The method according to claim 1, wherein when the distance between two end points forming the line segment is smaller than a predetermined value, it is determined that the range considered as the specific color is narrow.
The image processing method according to 1. 3. The image processing method according to claim 1, wherein white information is used instead of background color information when it is determined that the range considered as a specific color is narrow. 4. The image processing method according to claim 1, wherein when it is determined that a range considered as a specific color is narrow, color information designated by a user is used as background color information. 5. An image processing method in which a color which is distributed within a predetermined distance from a straight line including a line segment in a color space and whose internal division position with respect to the line segment is distributed in a predetermined value or range is regarded as a specific color. Wherein, after adaptively determining a specific color range distance from the color distribution information of the processing target image, if it is determined that the determined range distance is inappropriate, the range distance is reset. Image processing method. 6. The image processing method according to claim 5, wherein the determination whether the determined range distance is appropriate is performed when the determined range distance is smaller than a predetermined value. 7. The image processing method according to claim 5, wherein the resetting of the range distance is performed using a frequency histogram of a distance related to a color distribution of an image to be processed. 8. The resetting of the range distance is performed by comparing the frequency of the determined range distance with the frequency of all distances larger than the determined range distance, and there is no distance having a frequency exceeding the frequency of the determined distance. 8. The image processing method according to claim 7, wherein the method is performed in a case. 9. The image processing method according to claim 8, wherein at the time of the resetting, a value that is a predetermined multiple of the initially set distance is set as a new range distance. 10. The method according to claim 8, wherein at the time of the resetting, a minimum distance which is not less than a distance initially set in the histogram and whose frequency differential value is not negative is reset as a new range distance. The image processing method according to 1. 11. In the above-mentioned resetting, a smaller one of a constant multiple of the initially set distance value and a minimum distance value which is equal to or larger than the initially set distance in the histogram and whose differential value of the frequency is not negative. 9. The image processing method according to claim 8, wherein the distance is reset as a new range distance. 12. An image processing apparatus which regards a color which is distributed within a predetermined distance from a straight line including a line segment in a color space and whose internal segment position with respect to the line segment is distributed in a predetermined value or range as a specific color. Image processing characterized by using background color information as line segment color information, and using another color information instead of the background color information when it is determined that the range considered as a specific color is narrow. apparatus. 13. The method according to claim 12, wherein when the distance between two end points forming the line segment is smaller than a predetermined value, it is determined that the range considered as the specific color is narrow. Image processing device. 14. The image processing apparatus according to claim 12, wherein when it is determined that the range considered as a specific color is narrow, white information is used instead of background color information. 15. The image processing apparatus according to claim 12, wherein when it is determined that a range regarded as a specific color is narrow, color information designated by a user is used as background color information. 16. An image processing apparatus which regards a color which is distributed within a predetermined distance from a straight line including a line segment in a color space and whose internal division position with respect to the line segment is distributed in a predetermined value or range, as a specific color. Wherein, after adaptively determining a specific color range distance from the color distribution information of the processing target image, if it is determined that the determined range distance is inappropriate, the range distance is reset. Image processing apparatus. 17. The image processing apparatus according to claim 16, wherein the determination as to whether or not the determined range distance is appropriate is performed when the determined range distance is smaller than a predetermined value. 18. The image processing apparatus according to claim 16, wherein the resetting of the range distance is performed using a frequency histogram of a distance related to a color distribution of an image to be processed. 19. The resetting of the range distance is performed by comparing the frequency of the determined range distance with the frequency of all distances larger than the determined range distance, and there is no distance having a frequency exceeding the frequency of the determined distance. The image processing apparatus according to claim 18, wherein the processing is performed in a case. 20. The image processing apparatus according to claim 19, wherein when resetting the range distance, a value that is a predetermined multiple of the initially set distance is set as a new range distance. 21. When resetting the range distance, resetting a minimum distance that is equal to or more than the distance initially set in the histogram and whose differential value of the frequency is not negative as a new range distance. The image processing device according to claim 19, wherein: 22. When resetting the range distance, a constant multiple of the initially set distance value and a minimum distance value that is equal to or greater than the initially set distance in the histogram and whose frequency derivative is not negative. 20. The image processing apparatus according to claim 19, wherein a smaller one of the distances is reset as a new range distance. 23. A computer-readable recording medium on which a program for implementing the image processing method according to claim 1 is described. 24. A computer-readable recording medium on which a program for realizing the image processing method according to claim 2 is described. 25. A computer-readable recording medium on which a program for realizing the image processing method according to claim 3 is described. 26. A computer-readable recording medium on which a program for realizing the image processing method according to claim 4 is described. 27. A computer-readable recording medium on which a program for realizing the image processing method according to claim 5 is described. 28. A computer-readable recording medium in which a program for implementing the image processing method according to claim 6 is described. 29. A computer-readable recording medium on which a program for realizing the image processing method according to claim 7 is described. 30. A computer-readable recording medium on which a program for realizing the image processing method according to claim 8 is described. 31. A computer-readable recording medium in which a program for realizing the image processing method according to claim 9 is described. 32. A computer-readable recording medium on which a program for realizing the image processing method according to claim 10 is described. 33. A computer-readable recording medium in which a program for realizing the image processing method according to claim 11 is described.