JP2000099701A - Image color processor, image color processing method and recording medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To suppress the personnel expenses required for image color processing and to shorten the workhours. SOLUTION: A color image is divided into plural sections (step S1), a representative color for each sectional image is found out (step S2) and further, a set of non-edge sections not including edge images in the objective images is found out (step S3). Then, a specific section is noticed and a similar color extent coefficient expressing the extent of a section belonging to the set of the non- edge sections out of the sections around the noticed section, being smoothly continued to the surrounding sections and having color similar to the representative color of the noticed section as representative color is found out. The such similar color extent coefficient is found out as of each section out of the set of non-edge sections (step S4). The representative color of a section having the maximum out of the similar color extent coefficient of the obtained respective sections is made to a representative background color. The background color is corrected based on the obtained representative background color or the like.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image color processing apparatus for extracting a background color from a color image, an image color processing method, and a recording medium on which a computer program for realizing the method is recorded.

[0002]

2. Description of the Related Art In general, a product catalog or the like often includes a plurality of product images such as product photos. In such a case, these images are usually included in addition to the product images.
In most cases, the background image is almost a single-color solid background. When a plurality of such images are posted on the same page, it is often desirable that the colors of the background images are unified.

[0003] However, the background color of the image on which such an image is based is usually different for the following reasons. That is, there are cases where the photograph is a product photograph taken in a different photo studio, or when the photographing time, lighting conditions, film brand, and the like are different.

Therefore, in order to unify the background colors of a plurality of images, it is necessary to correct only the background color of the base image.
The operator corrects the background color of each product image one by one using a color correction method disclosed in Japanese Patent No. 2740443 or the like.

[0005]

By the way, in general, for color correction, such as the above method, it is necessary to specify a representative color and a color range of an image to be corrected.
Conventionally, the worker manually extracts these pieces of information about each product image one by one, and therefore spends a lot of labor and work time.

The present invention is intended to overcome the above-mentioned problems in the prior art, and realizes an image color processing apparatus, an image color processing method, and an image color processing apparatus capable of suppressing labor costs and shortening work time. To provide a recording medium on which a computer program for recording is recorded.

[0007]

According to one aspect of the present invention, there is provided an apparatus for dividing a color image into a plurality of sections and a representative color for each of the plurality of sections. A non-edge section set that extracts a non-edge section set that is a set of non-edge sections not including an edge image among a plurality of sections; As a representative color, a similar color which exists continuously around the target partition and which satisfies a predetermined similar color determination condition in the color space with respect to the representative color of the target partition has a representative color. Color width determining means for determining the spread of a non-edge section as a similar color width, and the representative color of the non-edge section having the largest similar color width in the non-edge section set as the representative background color of the color image Height Comprising a color determining means.

According to a second aspect of the present invention, there is provided the image color processing apparatus according to the first aspect, wherein the similar color determination condition is determined based on a point represented by the representative color of the target section in the Lab color space. It is a condition for determining whether or not the distance is equal to or less than a predetermined threshold.

According to a third aspect of the present invention, there is provided the image color processing apparatus according to the first or second aspect, further comprising: when the representative background color is substantially achromatic. Achromatic color selecting means for selecting an achromatic section having substantially an achromatic color among the non-edge sections, and an almost brightest representative color and an almost darkest representative color among the representative colors of the selected achromatic section, respectively, as a brightest background. Light / dark selecting means for selecting the color and the darkest background color.

According to a fourth aspect of the present invention, there is provided the image color processing apparatus according to the first or second aspect, wherein a non-edge image is generated when the representative background color is substantially chromatic. A chromatic color selection unit for selecting a chromatic color section having substantially a chromatic color among the sections, and a lightest representative color and a darkest representative color among the representative colors of the selected chromatic color section; Light / dark selecting means for selecting the darkest background color.

According to a fifth aspect of the present invention, there is provided the image color processing apparatus according to the first or second aspect, further comprising determining whether the representative background color is substantially achromatic or chromatic. Determining means, and achromatic color selecting means for selecting an achromatic section having substantially achromatic color among the non-edge sections when the representative background color is determined to be substantially achromatic by the determining means; First light / dark selecting means for selecting the most lightest representative color and the most darkest representative color among the representative colors of the achromatic color section as the brightest background color and the darkest background color, respectively; A chromatic color selecting means for selecting a chromatic color section having a substantially chromatic color among the non-edge sections when it is determined to be substantially chromatic; and a substantially brightest representative color among the representative colors of the selected chromatic color section And almost the darkest The representative color comprises a second contrast selection means for selecting as the brightest background color and the cycler background color, respectively, the.

According to a sixth aspect of the present invention, there is provided the image color processing apparatus according to the third or fifth aspect, wherein the achromatic color selecting means comprises an ab in the Lab color space.
A non-edge section having a representative color whose distance from the origin is equal to or less than a predetermined threshold value in a plane is defined as an achromatic section.

According to a seventh aspect of the present invention, there is provided the image color processing apparatus according to the third or fifth aspect, wherein the chromatic color selecting means comprises a representative background color, saturation, and hue. Are characterized in that a non-edge section having a representative color whose difference from the representative background color is within a predetermined threshold range is defined as a chromatic color section.

An apparatus according to an eighth aspect of the present invention is the image color processing apparatus according to any one of the first to seventh aspects, further comprising an image color processing apparatus based on the selected representative background color. Means for correcting the background color of the color image.

According to a ninth aspect of the present invention, there is provided a method for dividing a color image into a plurality of sections, a representative color extracting step for extracting a representative color of each of the plurality of sections, and a plurality of sections. A non-edge selection step of extracting a non-edge section set that is a set of non-edge sections not including an edge image, and sequentially executing each non-edge section of the non-edge section set,
A similar color that exists continuously around the target section and has a similar color as a color that satisfies a predetermined similar color determination condition in the color space with respect to the representative color of the target section while being set as the target section. A similar color width determining step of determining the spread of the non-edge section as a similar color width, and a background in which the representative color of the non-edge section having the largest similar color width in the non-edge section set is set as a representative background color of the color image A color determining step.

A method according to a tenth aspect of the present invention is the image color processing method according to the ninth aspect, further comprising the step of determining whether the representative background color is substantially achromatic or chromatic. And an achromatic color selecting step of selecting an achromatic section having substantially an achromatic color among the non-edge sections when the representative background color is determined to be substantially an achromatic color by the determining step; and the selected achromatic color section. And a first light / dark selection step of selecting a lightest representative color and a darkest representative color among the representative colors as a brightest background color and a darkest background color, respectively. A chromatic color selection step of selecting a chromatic color section having a substantially chromatic color among the non-edge sections when it is determined that there is a color; and a substantially brightest representative color and a substantially darkest representative color among the representative colors of the selected chromatic color section. The representative color, Respectively brightest background color and comprises a second contrast selection step of selecting as the cycler background color, the.

A method according to an eleventh aspect of the present invention is the image color processing method according to the ninth or tenth aspect, further comprising a color image processing method based on the selected representative background color. Correcting the color of the background.

A recording medium according to a twelfth aspect of the present invention is a recording medium in which a program for extracting a background color from a color image by a computer is recorded, wherein the color image is divided into a plurality of sections. A function, a representative color extraction function for extracting a representative color of each of the plurality of sections, a non-edge selection function for extracting a non-edge section which is a set of non-edge sections not including an edge image among the plurality of sections, Each non-edge section of the edge section set is sequentially set as a target section, and a non-edge that exists continuously around the target section and has a representative color that is visually similar to the representative color of the target section as a representative color. A similar color breadth that determines a similar color breadth for each non-edge partition of a non-edge partition set by calculating a similar color breadth that is the extent of a partition belonging to the partition set. A constant function, a non-edge section of similar Irohiro of the set is recording a program for realizing a background color determination function for a representative background color of the color image representative colors of the largest non-edge section.

[0019]

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

<1. FIG. 1 is a system configuration diagram of an image color processing apparatus according to an embodiment of the present invention. Hereinafter, this image color processing apparatus will be described with reference to FIG.

The apparatus of the present invention is an apparatus for automatically obtaining a background color and a range of the gradation of the background color for an image.

As shown in FIG. 1, an image color processing apparatus 100
The CPU 1 (division means, representative color extraction means, non-edge selection means, similar color width determination means, background color determination means, achromatic color selection means, light / dark selection means, chromatic color selection means, first and second light / dark selection And a means for correcting the background color together with the input unit 5 described later), a ROM 2 storing a basic program, a RAM 3 storing an image conversion program and the like described in detail below, and its image conversion. A reading unit 4 for reading a program from a recording medium 4a such as a magnetic disk or a CD-ROM; an input unit 5 for receiving an input from a keyboard or a mouse; a color display 6 for displaying a target image and an image color processing result; A main body in which a hard disk 7 for storing data and the like are connected by a bus line BL via an interface (I / F) as appropriate. Minute of the addition, the scanner 8, a printer 9
Is a device realized by executing an image conversion program by an internal CPU 1 or the like in one general computer system including peripheral devices such as. In addition,
This computer system can communicate with other devices (not shown) through a bus line BL to exchange various data and the like. The image conversion program may be stored in the ROM 2 or the hard disk 7 in advance, and read out and stored in the RAM 3 for use. In such a case, the ROM 2 or the hard disk 7 functions as a recording medium.

The following processing is performed by the above-described apparatus configuration.

<2. Image color processing of the embodiment>
Image color processing in this embodiment will be described.
FIG. 2 and FIG. 3 are flowcharts showing the whole image color processing of this embodiment. It is assumed that image data from which a background color is to be extracted is prepared in advance prior to the following processing, and that an operator has confirmed that the image includes a background portion. Unless otherwise specified, the following processing is automatically performed by software by the CPU 1 executing the image color processing program read into the RAM 3.

First, a color image to be prayed (hereinafter referred to as “target image”) is divided into a plurality of sections (step S).
1). FIG. 4 is a diagram showing how the target image is divided in this embodiment. As shown, in this embodiment, the target image is divided into rectangular sections that are equal to each other. Then, two-dimensional orthogonal coordinates XY are set in the image, and each section and each pixel of the section are designated by coordinate values. For the target image, M × N sections are set which are equally divided into M in the X direction and N in the Y direction. Hereinafter, for convenience, the X direction and Y
The number of pixels in the direction is represented by a and b, respectively. In FIG. 4, M = N = 4.

By the way, the present invention is premised on image color processing for a color image. In this embodiment, a color image is an image having each color component of R (red), G (green) and B (blue). And

Next, a representative color of each section image is obtained (step S2). Specifically, one of the sections is sequentially processed (hereinafter, in each step,
A section to be processed is referred to as a “target section”), and a histogram of the gradation distribution of each of the three color components of each pixel in each target section (each of R, G, and B components has a range of class values) Therefore, a rectangular parallelepiped class (having a class block) in the RGB color space is created, and the most frequent class block, which is the class block having the highest frequency, is obtained from the obtained histogram, and the most frequent class block and the class block adjacent thereto are obtained. The average value of the gradation values for each color component of the pixel belonging to the target section is set as each color component value of the representative color of the target section. The method of extracting the representative colors is disclosed in Japanese Patent Application No. 10-94569 filed by the present applicant.

Next, a non-edge section belonging to a smooth portion which does not include an edge image in the target image is obtained for all the sections (step S3). Specifically, it is obtained as follows.

First, to determine whether the target section belongs to a smooth part or a part including an edge image,
A parameter d (i, j) indicating smoothness is set for each pixel. FIG. 5 is a diagram showing a method for specifying a pixel. As shown, each pixel is represented by P (i, j), and the indices of the X and Y components are represented by i and j, respectively. And
The parameter d (i, j) is made to correspond to the pixel P (i, j). Here, the parameter d (i, j) is a quantity representing the smoothness of the image at the pixel P (i, j), and is determined by the following conditional expression.

[0030]

(Equation 1)

[0031]

(Equation 2)

[0032]

(Equation 3)

[0033]

(Equation 4)

Here, R, G, and R of the pixel P (i, j) are
The B components are Pr (i, j), Pg (i, j), Pb (i,
j). As can be seen from these equations, the parameter d (i, j) is the pixel P (i-1, j-1), P (i + 1, j-1) obliquely adjacent to the pixel of interest P (i, j). ),
The absolute value of the difference between the average value of the gradations of P (i-1, j + 1) and P (i + 1, j + 1) and the gradation of the pixel of interest P (i, j). Therefore, the closer the value of the parameter d (i, j) is to “0”, the smaller the gradation difference between the target pixel and the surrounding pixels, which means that the image is smooth and does not include edge components. . And such a parameter d
(i, j) is obtained for all the pixels included in the target section except for the pixels at the periphery of the target section. The reason for excluding the peripheral pixels is that the peripheral pixels have a small number of adjacent pixels, and it is inappropriate to include them in the image smoothness determination target. This is because even if it is excluded from the target of the determination, the determination result is not significantly affected. Note that the parameter d (i, j) does not take a negative value because it is the sum of absolute values.

FIG. 6 shows a parameter d (i,
6 is a flowchart showing a procedure for obtaining j). Less than,
A specific method for obtaining the parameter d (i, j) will be described.

First, a parameter d (i, j) of each pixel P (i, j) other than the periphery (outermost) included in the target section is determined (FIG. 6: step S21). That is, 0 <i <(a
−1), 0 <j <(b−1), the parameter d (i, j) for each pixel of interest P (i, j) whose index i and j are changed over the range of Equations 1 to 4 above. j) is obtained and stored in the RAM 3.

Next, the sum sd of the parameters d (i, j) of the pixels P (i, j) is obtained (FIG. 6: step S2).
2). That is, the sum of the parameters d (i, j) is obtained,
The value is defined as a sum sd.

Next, all pixels P (i,
The average value of the parameter d (i, j) of j) is obtained (FIG. 6: step S23). That is, the sum sd is divided by the number (a−2) × (b−2) of all pixels excluding the peripheral pixels, and the parameter d
An average value ad of (i, j) is obtained.

Next, it is determined whether or not all sections have ended (FIG. 6: step S24), and the processing in steps S21 to S24 is performed while changing the target section until the average value ad is obtained for all the sections (M × N). repeat. The sum sd and the average value ad are values representing the smoothness of each section.

Next, a procedure for obtaining a non-edge section as a section having a smooth image containing no edge component among all the sections using the obtained average value ad will be described with reference to the flowchart of FIG. Will be explained.

First, a non-edge section candidate whose section average value ad is equal to or smaller than the first threshold th1 is determined (FIG. 7: step S31). Here, the first threshold value th1 is a value in which a value that is determined to clearly have an edge component in a section image having an average value ad larger than this threshold value is obtained in advance by experiments or the like.

Next, the average value ad of all the non-edge section candidates
(FIG. 7: Step S3)
2). Then, discrimination analysis is performed on the obtained histogram to obtain a second threshold value th2 (FIG. 7: step S3).
3). The discriminant analysis performed here will be described more specifically. In the non-edge section candidates obtained in step S31, the sections whose average value ad is equal to or smaller than the first threshold th1 are sections in which it is difficult to determine whether or not there is an edge component in the image. The set of such sections includes a section including an edge component and a section of a smooth image not including the edge component. Accordingly, the obtained histogram is roughly divided into two groups: a section including an edge component, that is, a section having a large average value ad, and a section not including an edge component, that is, a section having a small average value ad. It is considered something.
Therefore, here, as a discriminant analysis, the histogram is divided into two groups so that the distance between the centroids (frequency is the weight) of both groups is maximized, and an average value ad serving as a boundary between both groups is obtained. The value ad is set to a second threshold value th2.

Next, a section whose average value ad is equal to or smaller than the second threshold value th2 is set as a non-edge section which is a section having a smooth image (FIG. 7: step S34). That is, the section belonging to the group having the smaller average value ad among the two groups obtained as described above is set as the non-edge section.

As can be seen from the above description, in this embodiment, attention is paid only to the section whose average value ad is equal to or less than the first threshold value th1, that is, the section in which it is difficult to determine whether or not there is an edge component in the image. The non-edge section is determined in more detail about the section. Therefore, the first threshold value t
A non-edge section can be extracted more accurately and efficiently than when a histogram of the average value ad is created for all the sections without using h1, and the above-described discriminant analysis is performed on the histogram to obtain a non-edge section. .

Returning to the description of FIG. Next, for each of the section images included in the set of non-edge sections, the number of sections that have smooth colors that are smoothly continuous around and have a similar color that is visually recognized as a color that is very similar to the relevant section image. A similar color area coefficient c (m, n) is obtained (step S4).

FIG. 8 is a flowchart showing a procedure for obtaining the maximum similar color area coefficient. Hereinafter, a method of obtaining the maximum similar color area coefficient will be described in more detail with reference to FIG.

First, all sections (M × N) in the target image
) Is assigned a similar color width coefficient c (m, n) indicating the size of a range having a color similar to that of the section. Where m,
n represents the coordinates of the section in the target image. The similar color width coefficient c (m, n) exists around a section belonging to the set of non-edge sections as a noticed section, and also belongs to the non-edge section, and its representative color is similar to the noticed section. Indicates the number of sections (hereinafter referred to as “similar color sections”). In other words, the larger the similar color area coefficient c (m, n), the greater the number of non-edge sections around which the noticeable section and the representative color are similar.

First, the similar color width coefficient c for each section
As an initial value of (m, n), “1” is set for a non-edge section,
"0" is set for the other sections (FIG. 8: step S41).

Next, "1" is set as the parameter T for obtaining the maximum similar color area coefficient (FIG. 8: step S42).

Next, the section where c (m, n) = T is set as the target section (FIG. 8: step S43).

FIG. 9 shows a similar color area coefficient c (m, n) of each section.
Is a diagram showing the setting of each rectangle, each rectangle represents a section B (reference numerals are partially omitted), and a numerical value in each rectangle represents a similar color width coefficient c (m, n). As shown in FIG. 9A, initially, the similar color width coefficient c (m, n) of each section is “1” or “0”.
It has become.

Next, it is determined whether or not the majority of the sections in the vicinity of (T + 2) × (T + 2) of the section of interest have a similar color breadth coefficient c (m, n) and whether or not the representative colors are similar. 8: Step S4
4) If this condition is satisfied, the similar color width coefficient c (m, n) of the target section is set to T + 2 (FIG. 8: step S45), and the process proceeds to step S46. Conversely, if the condition is not satisfied, the process proceeds directly to step S46.

This determination will be described with reference to a specific example of FIG.
First, for all the sections in the image where the similar color width coefficient c (m, n) is “1” (parameter T), the 3 × 3
((T + 2) × (T + 2)) A similar color area coefficient c (m, n) of a section near ((hatched section in FIG. 9A)) and whether the representative color is similar to the section Check whether or not. More specifically, if the similar color area coefficient c (m, n) is positive and the section whose representative color is similar to the section of interest AB is a majority of the neighboring sections, the similar color area of the section of interest AB is used. The coefficient c (m, n) is set to "3", otherwise (as it is "1")
And

Here, whether or not the representative color of the section adjacent to the section of interest is similar to the section of interest is determined according to a predetermined similar color determination condition in the color space. In this embodiment, the following conditions are employed as such similar color determination conditions. That is, the representative color of the target section and the representative color of the section adjacent to the target section are converted from the values of the RGB color system into the values of the Lab color system, and are represented as points in the Lab color space. FIG. 10 is a diagram schematically illustrating determination of a similar color in the Lab color system. In FIG. 10, the two representative colors are schematically represented as points p1 and p2. Incidentally, the distance d between the two points p1 and p2 in the Lab color space is known as representing the degree of color similarity. Therefore, the distance d in the Lab space between the representative color of the section of interest and the representative color of the section adjacent to the section of interest is determined, and a threshold (L
(the threshold of the distance in the ab color space), and if it is equal to or less than the threshold, it is determined that the two representative colors are similar. In order to convert the values of the RGB color system to the values of the Lab color system, first, the ROM 2
Is converted using a look-up table for converting values of the RGB color system prepared in advance into values of the Lab color system.

Next, it is determined whether or not the above processing has been completed for all the blocks of c (m, n) = T (FIG. 8: step S46). If not, the flow returns to step S43.
The processing of steps S43 to S46 is repeated for all the blocks of c (m, n) = T. When it is determined that the processing has been completed for all the partitions, it is determined whether or not there is a partition of c (m, n) = T + 2 (FIG. 8: step S4).
7). And if there is such a section, the parameter T
(Step S48 in FIG. 8), and the process returns to step S43, and repeats the processing of steps S43 to S48 until it is determined that there is no section of c (m, n) = T + 2.

The description of the specific example of FIG. 9 will be continued. Step S43 with the parameter T = 1 as described above.
Steps S46 to c (m, n) = 1 (that is, c (m, n) =
FIG. 9B shows a state performed for all the sections in T).
It is. As shown in FIG. 9B, since there is a section of c (m, n) = 3 (that is, c (m, n) = T + 2), “Yes” is determined in step S47, and T = The process returns to step S43 and returns to step S43. In step S44, this time, for all the blocks in the image where the similar color breadth coefficient c (m, n) is “3” (T), the 5 × 5
The similar color width coefficient c (m, n) and the representative color of the neighboring section are checked, and the similar color width coefficient c (m, n) takes a positive value, and
If the number of sections in which the representative color is similar to the noticed section AB is a majority, the similar color width coefficient c (m, n) of the section is set to “5”.
(T + 2). FIG. 9C shows this state. In this manner, the range of detecting the spread of the non-edge section having a similar color around each section is sequentially increased (the parameter T is increased by “2”), and the similar color is increased as described above. The value of the width coefficient c (m, n) is updated. As a result, the surrounding non-edge sections having similar colors have a large spread, and the similar color width coefficient c
As (m, n) gradually increases,
The number of sections in which the value of the similar color area coefficient c (m, n) is updated decreases.

Then, in the determination in step S47, "N
If it is determined to be "o", the parameter T is set as the maximum similar color area coefficient (FIG. 8: step S49). That is, the detection of the spread of the similar color has been completed for all the non-edge sections, and the similar color width coefficient c (m,
Since the maximum value of n) should be the parameter T at that time, the parameter T is set to the maximum similar color area coefficient.

Next, returning to FIG. 2, the section having the largest similar color area coefficient is determined as the largest similar color section (step S).
5) The representative color of the section having the largest similar color area coefficient (hereinafter, referred to as “maximum similar color area”) is determined as the representative background color that is the representative background color of the image. Is stored (step S6). That is, it is a non-edge section, and
The representative color of a section that is smoothly continuous around and has the widest spread of sections having similar colors is determined as a representative background color. However, when there is one section in which the similar color area coefficient c (m, n) matches the maximum similar color area, the representative color of that section is used as the representative background color of the image. There may be more than one. In such a case,
The hues of the representative colors of the sections are obtained, and the average value of the hues is obtained. Note that the method of calculating the hue of the representative color can be obtained by performing conversion to the HSL color system described below. Then, the representative color of the section having the hue closest to the average value of the obtained hues among the representative colors is determined as the representative background color of the image, and stored in the RAM 3.

Thus, the representative background color of the target image has been obtained.

Next, it is determined whether or not the representative background color is close to an achromatic color (step S7). Specifically, a representative background color RP (RG
B) is a representative background color RP represented by the HSL color system
(HSL), the S (saturation) component of the color components is compared with a predetermined saturation threshold, and if smaller than the saturation threshold, it is determined that the representative background color is achromatic or close to achromatic. And
Proceed to step S8. Conversely, if the saturation threshold is exceeded, it is determined that the representative background color is a chromatic color or close to a chromatic color, and the process proceeds to step S10.

In order to convert the values of the RGB color system into the values of the HSL color system, the values of R, G, and B are converted into three perceived color values using a look-up table prepared in the ROM 2. It is converted into a certain hue H, saturation S, and lightness L. Where R,
Since the values of G and B do not directly correspond to the values of hue H, saturation S and lightness L, in this conversion, the values of R, G and B are temporarily changed to the Lab color system (1 corresponding to lightness L). The three-dimensional coordinate space, the saturation S, and the hue H are converted into values in a three-dimensional coordinate space (two-dimensional coordinate space composed of two-dimensional coordinates drawn on a plane indicated by reference symbols a and b). Then the L
The value of the ab color system is converted to the value of the HSL color system. FIG.
FIG. 4 is a diagram illustrating a relationship between a Lab color space and an HSL color space. At this time, the Lab color space and the HSL color space have a relationship as shown in FIG. In FIG. 11, Lab
The L axis of the color space corresponds directly to the lightness L of the HSL color space, the radius on the two-dimensional coordinates of the a axis and the b axis corresponds to the saturation S, and the angle θ between the a axis and the radius is the hue. H is supported.
The details of this conversion are also described in Japanese Patent Application No. 10-9 by the present applicant.
4569.

Next, when it is determined that the representative background color is achromatic or close to achromatic, the section having the representative color of achromatic or close to achromatic in the set of non-edge sections. A set of achromatic similar sections is obtained (step S8).
Specifically, the representative color rp (RGB) of the non-edge section is converted into the value rp (Lab) of the Lab color system by the method described above, and the ab component of the representative color rp (Lab) in the set of the non-edge sections is converted. Finds a set of sections (considered to be sections with similar colors) within a predetermined distance (threshold) from the origin O (L axis representing achromatic color) in the ab plane. This is because, in the case of a color that is close to an achromatic color, the similarity of the color does not depend much on the hue, and thus the similarity of the color is uniformly determined only from the distance from the achromatic color. In this case, L (brightness)
Ignore components.

Next, among the representative colors of each achromatic similar section,
The brightest representative color and the darkest representative color, respectively,
The brightest background color and the darkest background color of the target image are determined and stored (step S9). Specifically, the brightest (L is the largest) section belonging to the set of sections whose colors are close to the achromatic color is the brightest section. Similarly, the darkest (the smallest L) section belongs to the set of sections whose colors are close to the achromatic color. Then, the representative colors of the brightest section and the darkest section are determined as the brightest background color and the darkest background color, respectively, and are stored in the RAM 3.

Conversely, if it is determined in step S7 that the representative background color is a chromatic color or close to a chromatic color,
Among the non-edge sections, a chromatic similar section whose representative color is close to the representative background color in hue and saturation is obtained (step S1).
0). Specifically, the representative color rp (RG
B) is converted into an HSL-based value rp (HSL), and the saturation (S) and hue (H) of the representative color rp (HSL) and the saturation of the representative background color RP (HSL) in the set of non-edge sections. When the difference between the degree (S) and the hue (H) is equal to or less than a predetermined threshold, the representative color is determined to be close to the representative background color, and the section is regarded as a chromatic similar section. Find a set of parcels.

Next, among representative colors of each chromatic similar section,
The brightest representative color and the darkest representative color, respectively,
The lightest background color, which is the lightest background color of the target image, and the darkest background color, which is the darkest background color, are determined and stored (step S11). Specifically, the brightest section (L is the maximum value) that belongs to the set of chromatic similar sections and is the brightest section. Similarly, a section belonging to the set of chromatic similar sections and being the darkest (L is the minimum value) is defined as the darkest section. Then, the representative colors of the brightest section and the darkest section are determined as the brightest background color and the darkest background color, respectively, and are stored in the RAM 3.

As described above, the representative background color, the brightest background color, and the darkest background color have been obtained. Then, if necessary, the background color is corrected using these color data (step S1).
2).

More specifically, the representative background color is set as a color to be corrected, the brightest background color and the darkest background color are set as a range of the color correction processing based on the color to be corrected, and each of them is defined as an HSL color system. Convert to Further, the operator designates the changed target color represented by the HSL color system through the input unit 5. Next, color correction parameters are set based on the representative background color and the target color. This parameter is a value such as a ratio between the representative background color and the target color. Further, the image data in pixel units of the target image is converted into image data of the HSL color system, and it is determined whether or not the converted image data falls within the range of the color change processing. , The original RGB image data of the pixel is multiplied by the above parameter to correct the target color or a color close to the target color. The details of this color correction method are described in Japanese Patent No. 2740436 and Japanese Patent No.
No. 3 discloses this.

The representative background color, the brightest background color, and the darkest background color may be recorded on the hard disk 7, the recording medium 4a, or the like, and may be taken out and corrected when necessary.

In some cases, a plurality of target images are prepared, and in such a case, the above processing is repeated. Thus, the image color processing according to the embodiment ends.

As described above, according to the image color processing apparatus of this embodiment, the representative background color, the brightest background color and the darkest background color are automatically obtained for the color image, and these are obtained. To correct the background color of the color image based on the
The work load can be reduced as compared with the case where the background color is extracted from the color image by hand and the color is corrected, whereby the cost due to labor costs can be reduced and the work time can be shortened.

In addition, since the representative color of the non-edge section having the largest similar color width among the set of non-edge sections is set as the representative background color of the color image, even if the background color has shading, it is included in the non-edge section. , A representative background color can be accurately extracted. Further, by performing the color correction of the background color based on such a representative background color, the color correction can be performed based on a color suitable for the color correction, and the quality of the color correction can be improved.

When calculating the similar color width coefficient, La
Since it is determined whether or not the distance between the representative colors in the b color space is equal to or less than a predetermined threshold value, the similar color can be objectively extracted, and the spread of the color visually similar to the background can be accurately determined. In addition to being able to capture, the representative background color can be accurately extracted.

Further, it is determined whether the representative background color is substantially achromatic or chromatic. When the representative background color is determined to be substantially achromatic and when the representative background color is determined to be substantially chromatic, the brightest background color and the chromatic color are determined as described above. Since the darkest background color is determined, the representative background color can be used for both chromatic images and achromatic images, so that the versatility is high. In addition, the range of the background color can be accurately specified using the brightest background color and the darkest background color obtained thereby, so that even when the background color is corrected, only the color of the background image can be easily corrected. . In addition, thereby, costs due to labor costs can be further reduced, and work time can be further reduced.

Further, when the representative background color is an achromatic color, a non-edge section having a representative color whose distance from the origin is equal to or smaller than a predetermined threshold in the ab plane in the Lab color space is defined as an achromatic section, and In the case where the representative background color is a chromatic color, a non-edge section having a representative color whose difference between the representative background color and the representative background color for each of the saturation and the hue is within a predetermined threshold range is provided. Since it is a colored section, the achromatic section and the chromatic section can be accurately extracted, respectively, and therefore, the exact brightest background color and the darkest background color can be extracted, so that the range of the background image can be more accurately extracted. Color correction can be performed more accurately.

<3. Modifications> In the above embodiment, an example of the image color processing device and the image color processing by the image color processing device has been described, but the present invention is not limited to this.

For example, in the above-described embodiment, the image color processing is performed using the image data from which the background color is to be extracted as it is. However, the image color processing is started (step S1 in FIG. 2 is executed). The size of the target image data may be compared with a predetermined threshold value before, and if the size is larger than the threshold value, the image color processing may be performed after reducing the size by an appropriate method (such as pixel thinning). In this case, since the data amount of the image data is reduced, the storage capacity of the image data such as the RAM is reduced, and the processing speed is improved.

Further, in the above embodiment, it is determined in step S7 of the flowchart in FIG. 3 that the representative background color is substantially achromatic or almost chromatic. However, the background color is substantially achromatic or chromatic. If only the image is targeted, after performing step S6, only steps S8 and S9 or steps S10 and S11 may be performed, and step S12 may be further performed.

In the above embodiment, the parameter d (i, j) representing the smoothness of the image of each pixel is obtained by using the target pixel P for each color component as shown in Expressions 1 to 3.
The pixel P (i−1, j−) that is obliquely adjacent to (i, j)
1), P (i + 1, j-1), P (i-1, j + 1), P (i +
The absolute value of the difference between the average value of the gradation of (1, j + 1) and the gradation of the pixel of interest P (i, j) is obtained, and the sum of them is used as a parameter d (i, j) according to the equation (3). The target pixel P
Pixels P (i−1, j), P (i + 1, j), P (i, j−) adjacent to the positive side and the negative side of the X axis and the Y axis with respect to (i, j)
1), the average value of the gradations of P (i, j + 1) and the target pixel P (i,
The absolute value of the difference from the gradation of j) is taken as the parameter d (i, j), or the pixel P (i + r, j + s) adjacent to the pixel P (i, j)
(R = 0, ± 1 and s = 0, ± 1) The absolute value of the difference from the average value of all the gradations is used as the parameter d (i, j). The absolute value of the difference from the average value of the gradations of the pixels adjacent on the outside may be used.

Further, in the above embodiment, the representative colors of the brightest section and the darkest section are set to the brightest background color and the darkest background color in steps S9 and S11 of the flowchart of FIG. A color slightly lighter than the representative color and a color slightly darker than the representative color of the darkest section may be set as the brightest background color and the darkest background color, respectively. This is because the representative color of each section image extracted in step S2 in FIG. 2 is a representative color of those sections, so that not all pixels in those sections have the representative color. However, it is considered that the image usually has a predetermined range around the representative color. Therefore, it is considered that a slightly lighter (or darker) pixel is also present in the section having the brightest (or darkest) representative color. If such a representative color is set as the brightest (darkest) background color, At the time of color correction, pixels that are originally included in the background image but are not color-corrected (uncorrected) in those sections, or conversely, pixels that are not originally included in the background color are corrected. This is because things (overcorrection) may occur.

[0080]

As described above, according to the first to seventh aspects of the present invention, a dividing means for dividing a color image into a plurality of sections, and a representative color for extracting a representative color of each of the plurality of sections. Extracting means, non-edge selecting means for extracting a set of non-edge sections not including an edge image among a plurality of sections, and sequentially setting each non-edge section of the set of non-edge sections as a target section, By calculating the similar color width, which is the number of sections belonging to a set of non-edge sections that have a similar color that is visually similar to the representative color of the noticeable section as a representative color while being continuously present around, A similar color width determining means for obtaining a similar color width for each non-edge section of the set of non-edge sections; and a representative color of the non-edge section having the largest similar color width in the set of non-edge sections as a representative of the color image. Target Since the background color determining means for the scenery is provided, the work load can be reduced as compared with a case where the background color is manually extracted from the color image. You can save time.

According to the invention of claims 1 to 12, the representative color of the non-edge section having the largest similar color width among the set of non-edge sections is set as the representative background color of the color image. Even when the background color has shading, a representative background color can be accurately extracted therefrom. Further, by performing the color correction of the background color based on such a representative background color, the color correction can be performed based on a color suitable for the color correction, and the quality of the color correction can be improved.

According to the second aspect of the present invention, it is determined whether or not a similar color is present based on whether or not the distance between representative colors in the Lab color space is equal to or smaller than a predetermined threshold value. In addition, the spread of colors visually similar to the background can be accurately grasped, and the representative background color can be accurately extracted.

Further, claim 3, claim 5, and claim 1
According to the invention of No. 0, when the representative background color is substantially achromatic, the brightest representative color and the darkest representative color among the representative colors of the achromatic section having the substantially achromatic color among the non-edge sections are determined. In order to obtain the brightest background color and the darkest background color, respectively,
According to the invention, when the representative background color is substantially a chromatic color, the brightest representative color and the darkest representative color among the representative colors of the chromatic sections having the substantially chromatic colors among the non-edge sections are represented by: Since the background color is the brightest background color and the darkest background color, respectively, the range of the background color can be accurately specified using the brightest background color and the darkest background color, respectively. Only the color of the image can be easily corrected. In addition, thereby, costs due to labor costs can be further reduced, and work time can be further reduced.

According to the inventions of claims 5 and 10, it is determined whether the representative background color is substantially achromatic or chromatic. When the representative background color is determined to be substantially achromatic, it is determined that the representative background color is substantially chromatic. Since the brightest background color and the darkest background color are obtained for each case as described above, the versatility is higher because the representative background color can be used for both chromatic and achromatic images.

According to the invention of claim 6, a non-edge section having a representative color whose distance from the origin is equal to or smaller than a predetermined threshold in the ab plane in the Lab color space is defined as an achromatic section. Since a non-edge section having a representative color whose difference between the representative background color and the representative background color for each of the saturation and the hue is within a predetermined threshold value is defined as a chromatic section, the achromatic section and the non-achromatic section are respectively provided. Since the colored section can be accurately extracted, and thus the accurate brightest background color and the darkest background color can be extracted, the range of the background image can be extracted more accurately, and the color correction can be performed more accurately.

According to the eighth aspect of the present invention, since the background color of the color image is corrected based on the selected representative background color, the background color can be automatically corrected. Further, it is possible to further reduce the cost due to labor costs and shorten the working time.

[Brief description of the drawings]

FIG. 1 is a system configuration diagram of an image color processing apparatus according to an embodiment of the present invention.

FIG. 2 is a flowchart illustrating an entire image color process according to the embodiment;

FIG. 3 is a flowchart illustrating an entire image color process according to the embodiment;

FIG. 4 is a diagram showing a state of division of a target image in the embodiment.

FIG. 5 is a diagram showing a method for specifying a pixel.

FIG. 6 shows a parameter d of a target section in the embodiment.
It is a flowchart which shows the procedure which calculates | requires (i, j).

FIG. 7 is a flowchart showing a procedure for obtaining a non-edge section in the embodiment.

FIG. 8 is a flowchart showing a procedure for obtaining a maximum similar color area coefficient in the embodiment.

FIG. 9 is a diagram showing a manner of setting a similar color width coefficient of each section in the embodiment.

FIG. 10 is a diagram schematically illustrating determination of a similar color in the Lab color system.

FIG. 11 is a diagram illustrating a relationship between a Lab color space and an HSL color space.

[Explanation of symbols]

 REFERENCE SIGNS LIST 1 CPU 2 ROM (recording medium) 3 RAM 4 reading unit 4a recording medium 7 hard disk (recording medium) 100 image color processing device B section AB target section RP representative background color T parameter c (m, n) similar color width coefficient th1, th2 First and second thresholds

Continued on the front page F-term (reference) 5B057 CA01 CA08 CA12 CA16 CB01 CB08 CB12 CB16 CC03 CE09 CE17 CE18 CH07 DA08 DB02 DB06 DB09 DC04 DC16 DC19 DC25 5C077 LL17 LL18 MP08 PP32 PP35 PP37 PP47 PP52 PP53 PQ19 PQ23 LA10B12LA NA03 NA11 NA25 PA07 5L096 AA02 AA06 EA45 FA19 FA59 GA53 JA11

Claims (12)

[Claims] A dividing unit that divides a color image into a plurality of sections; a representative color extracting unit that extracts a representative color of each of the plurality of sections; and a non-edge section that does not include an edge image among the plurality of sections. A non-edge selecting means for extracting a non-edge section set, which is a set of the non-edge section sets; A similar color width determination for obtaining, as a similar color width, a spread of a non-edge partition having a similar color as a representative color that satisfies a predetermined similar color determination condition in a color space with respect to the representative color of the partition. Means; and a background color determining means for setting a representative color of the non-edge section having the largest similar color width in the non-edge section set as a representative background color of the color image. Image color processing apparatus according to. 2. The image color processing apparatus according to claim 1, wherein the similar color determination condition is whether or not a distance from a point represented by a representative color of the target section in a Lab color space is equal to or less than a predetermined threshold. An image color processing device, characterized in that the conditions are a determination. 3. The image color processing apparatus according to claim 1, further comprising: if the representative background color is substantially achromatic, the non-edge section has substantially an achromatic color. Achromatic color selection means for selecting an achromatic color section; and selecting, from among the representative colors of the selected achromatic color section, a substantially lightest representative color and a substantially darkest representative color as a brightest background color and a darkest background color, respectively. And an image color processing device. 4. The image color processing device according to claim 1, wherein, when the representative background color is substantially chromatic, a chromatic color having substantially chromatic color among the non-edge sections. Chromatic color selecting means for selecting a section; and light and dark for selecting, as the brightest background color and the darkest background color, the substantially brightest representative color and the substantially darkest representative color among the representative colors of the selected chromatic color section, respectively. An image color processing device comprising: a selection unit. 5. The image color processing apparatus according to claim 1, further comprising: a determination unit configured to determine whether the representative background color is substantially an achromatic color or a chromatic color; When the representative background color is determined to be substantially achromatic, achromatic color selecting means for selecting an achromatic section having substantially achromatic color among the non-edge sections; and the representative of the selected achromatic color section. First light / dark selecting means for selecting a lightest representative color and a darkest representative color as the lightest background color and the darkest background color, respectively, among the colors; and the representative background color is substantially chromatic by the determining means. When it is determined that there is a chromatic color section having a substantially chromatic color among the non-edge sections, a chromatic color selecting unit, and a substantially brightest representative color among the representative colors of the selected chromatic color section and Ho Darkest representative color, image color processing apparatus characterized by comprising: a second contrast selection means for selecting as the brightest background color and the cycler background color, respectively, the. 6. The image color processing device according to claim 3, wherein the achromatic color selection unit is configured such that a distance from an origin in an ab plane in a Lab color space is equal to or less than a predetermined threshold. An image color processing apparatus, wherein the non-edge section having a representative color is the achromatic section. 7. The image color processing device according to claim 3, wherein the chromatic color selection unit determines a difference between the representative background color and the representative background color for each of saturation and hue. An image color processing apparatus, wherein a non-edge section having a representative color whose difference falls within a predetermined threshold range is set as the chromatic section. 8. The image color processing apparatus according to claim 1, further comprising: correcting a background color of the color image based on the selected representative background color. An image color processing apparatus comprising: 9. A division step of dividing a color image into a plurality of sections; a representative color extraction step of extracting a representative color of each of the plurality of sections; and a non-edge section not including an edge image among the plurality of sections. A non-edge selection step of extracting a non-edge section set, which is a set of the non-edge section sets; A similar color width determination for obtaining, as a similar color width, a spread of a non-edge partition having a similar color as a representative color that satisfies a predetermined similar color determination condition in a color space with respect to the representative color of the partition. And a background color determining step of setting a representative color of the non-edge section having the largest similar color width in the non-edge section set as a representative background color of the color image. Image color processing method according to. 10. The image color processing method according to claim 9, further comprising: a judging step of judging whether the representative background color is substantially achromatic or a chromatic color; Is determined to be substantially achromatic, an achromatic color selecting step of selecting an achromatic section having a substantially achromatic color among the non-edge sections; and A first light / dark selection step of selecting a brightest representative color and a substantially darkest representative color as a brightest background color and a darkest background color, respectively; and the determination step determines that the representative background color is substantially a chromatic color. A chromatic color selection step of selecting a chromatic color section having substantially a chromatic color among the non-edge sections, and a substantially brightest representative color and a substantially darkest representative color among the representative colors of the selected chromatic color section. Image color processing method comprising: a color, a second contrast selection step of selecting as the brightest background color and the cycler background color, respectively, the. 11. The image color processing method according to claim 9, further comprising a step of correcting a background color of the color image based on the selected representative background color. An image color processing method comprising: 12. A recording medium storing a program for extracting a background color from a color image by a computer, a dividing function of dividing the color image into a plurality of sections, and extracting a representative color of each of the plurality of sections. A representative color extraction function, a non-edge selection function of extracting a non-edge section which is a set of non-edge sections not including an edge image among the plurality of sections, and sequentially paying attention to each non-edge section of the non-edge section set. Spreading of the section belonging to the non-edge section set having a similar color which is continuously present around the noticeable section and has a similar color visually regarded as similar to the representative color of the noticeable section as the representative color while forming the section. Is obtained for each non-edge section of the set of non-edge sections. A computer recording a program for realizing a function and a background color determining function of setting a representative color of the non-edge section having the largest similar color width in the non-edge section set as a representative background color of the color image A readable recording medium.