JP2002163650A - Color extracting image processing device and color extracting method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To decide whether it is the same color or not with high reproducibility regardless of the presence of gloss. SOLUTION: A color image picked up by a color TV camera 11 is displayed on a display device 16. When a reference area including a color to be extracted is designated within the display device 16 and when the inside of the reference area has gloss, the range of the same color is determined by an RGB color system and when the inside of the reference area does not have gloss, the range of the same color is determined by hue, saturation and lightness. Further, when an examination area is designated in the color image, corresponding to whether the color selected in the reference area has gloss or not, any one range of the same color is used and a pixel within the relevant range is extracted from the examination area and displayed on the display device 16.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a color extraction image processing apparatus and a color extraction method.

[0002]

2. Description of the Related Art Generally, a color in a color image is specified by using a tristimulus value which is a mixed amount of original stimuli having the same color as the color of each pixel of the color image. The color image is R (70
0.0 nm), G (546.1 nm), B (435.8)
Since the three colors (nm) are used as the primary stimuli, an RGB color system that represents the colors by the tristimulus values of R, G, and B is used to specify the colors. In the following, the primary stimulus is [R] [G]
Symbols with brackets, such as [B], represent color stimuli, and symbols without brackets, such as R, G, B, represent tristimulus values. That is, since the tristimulus values that match the color [F] are represented by the following equations, the tristimulus values R, G, and B
Can specify the color. [F] = R [R] + G [G] + B [B] By the way, many color images are obtained by imaging an object with an imaging device such as a color TV camera. The tristimulus values R, G, and B may vary depending on the unevenness in the amount of light applied to the target. As described above, it is not appropriate if the same color is not determined because the three stimulus values R, G, and B are different from each other. Therefore, when specifying the color in the color image, the unevenness of the irradiation light amount to the object is considered. It is considered that a range is set for each of the tristimulus values R, G, and B, and if all of the tristimulus values R, G, and B do not exceed the set ranges, it is considered that the colors are the same. However, when a range is set for the tristimulus values, an unnecessary color area is also included, and a problem arises in that the color is determined to be the same even when the color is not the same.

In order to solve the above-mentioned problem, the technique disclosed in Japanese Patent Application Laid-Open No. 11-134503 does not include values related to tristimulus values R, G, and B of a pixel to be extracted. The range of R, G, B to be extracted is set within the range. Also, in addition to setting ranges for the tristimulus values R, G, and B, ranges for difference values such as RG values, BG values, and RB values are added, and the tristimulus values R , G, B
The publication also discloses a technique of converting into hue, saturation, and lightness and setting a range for these values, and setting a range for a difference value between these values.

As a technique for solving the above-mentioned problem, as described in Japanese Patent Publication No. Hei 7-92401, tristimulus values R, G, and B obtained from a color image are converted into hue, saturation, lightness, and color tone. The color in the range is determined to be the same color by converting into the angle and the color tone length and setting the range for at least three components including the hue and the color tone angle among these five components. It also describes that an L ^* a ^* b ^* color system can be adopted as a color system. The above publication describes that the color tone angle and color tone length correspond to the angle and length of a color tone vector on a color tone surface composed of lightness and chroma, respectively.

[0005]

As described above, the color of an object in a color image changes according to the amount of light applied to the object, and in particular, chips which have been increasingly miniaturized in recent years. When determining the quality of solder plating on parts or performing an appearance inspection such as dirt on relay terminals, the target object should be gold,
Since it is a metal such as silver or copper and has luster, the color of the object in the color image may be large depending on the relationship between the position where the object is imaged and the position of the light source and the brightness of the light source. A change occurs, and it is difficult to determine whether or not the colors are the same using the conventional technology that does not consider the gloss.

The present invention has been made in view of the above circumstances, and an object of the present invention is to determine with good reproducibility whether or not not only a region having no gloss but also a region having gloss has the same color. It is an object of the present invention to provide a color extraction image processing device and a color extraction method which can realize the above.

[0007]

According to the first aspect of the present invention, an RG
A reference region setting means for specifying a reference region including a color to be extracted from a color image having tristimulus values of the B color system as values of respective pixels; and RGB when the reference region has gloss.
First extraction range setting means for determining the same color range in the color system, and second extraction range setting for determining the same color range based on hue, saturation, and brightness when the reference region has no gloss Means, an inspection area setting means for specifying an inspection area for a color image from which pixels of the same color as the reference area are to be extracted,
Using the same color range determined by one of the first extraction range setting means and the second extraction range setting means depending on whether the color to be extracted from the inspection area has gloss, pixels in the range are determined. Color extraction means for extracting from the inspection area.

According to a second aspect of the present invention, in the first aspect of the invention, the first extraction range setting means sets the R / B and G /
The range of the same color is determined around each average value of the values of B and B.

According to a third aspect of the present invention, in the first aspect of the present invention, the first extraction range setting means sets RB and G-
The range of the same color is determined around each average value of the values of B and B.

According to a fourth aspect of the present invention, in the first to third aspects, the average value of the R / B and G / B values of all the pixels in the reference area or the total value of the R / B and G / B values in the reference area is obtained. A gloss judging unit for judging whether or not the reference area has gloss based on a magnitude relationship between each average value of the RB and GB values of the pixel and a preset judgment value; Is provided with a determination result presenting means for presenting the determination result according to.

According to a fifth aspect of the present invention, in the first to third aspects of the present invention, the average value of the R / B and G / B values of all the pixels in the reference area or the total value of the R / B and G / B values in the reference area is obtained. A gloss judging unit for judging whether or not the reference area has gloss based on a magnitude relationship between each average value of the RB and GB values of the pixel and a preset judgment value; And a selection means for selecting a first extraction range setting means and a second extraction range setting means using the result registered by the reference color registration means. .

According to a sixth aspect of the present invention, in the first to third aspects, the average value of the R / B and G / B values of all the pixels in the reference area or the total value of the R / B and G / B values in the reference area is obtained. A gloss judging unit for judging whether or not the reference area has gloss based on a magnitude relationship between each average value of the RB and GB values of the pixel and a preset judgment value; The first extraction range setting means and the second
Automatic selection means for selecting the extraction range setting means.

According to a seventh aspect of the present invention, when a reference area designated to include a color to be extracted from a color image having tristimulus values of the RGB color system as the values of each pixel includes a color to be extracted, if the reference area has a gloss, the RGB is used. When the same color range is determined in the color system, and when there is no gloss, the same color range is determined in a color system different from the RGB color system, and then the pixels of the same color as the reference area are extracted. The method is characterized in that one of the two ranges is selected according to whether or not the color to be extracted from the inspection area set has gloss, and pixels within the range are extracted from the inspection area.

According to an eighth aspect of the present invention, in accordance with the seventh aspect of the present invention, the magnitude relationship between the average value of the R / B and G / B values of all the pixels in the reference area and a predetermined judgment value is determined. It is characterized in that it is determined whether or not the reference area has gloss.

According to a ninth aspect of the present invention, in accordance with the seventh aspect of the present invention, the magnitude relationship between each average value of RB and GB values of all pixels in the reference area and a predetermined judgment value is determined. It is characterized in that it is determined whether or not the reference area has gloss.

[0016]

DESCRIPTION OF THE PREFERRED EMBODIMENTS (First Embodiment) In this embodiment described below, the image processing apparatus shown in FIG. 1 is used. This image processing device is a color TV as an imaging device.
The camera 11 captures an image of a space area including an object, and extracts a pixel of a specified color from a color image obtained by the color TV camera 11. An input device 18 including a keyboard, a mouse, and the like is used to specify a color.
The color image obtained by the color TV camera 11 is converted into a digital signal by the input circuit 12, and the signal values of each color corresponding to the tristimulus values R, G, and B according to the RGB color system are separately stored in the image capture memory 13R, 13G and 13B. The tristimulus values R, G, and B stored in the image capture memories 13R, 13G, and 13B are converted into a color system using a color conversion formula shown in Expression 1. Here, in Equation 1, the tristimulus values are X,
The first term on the right side is a primary conversion of the tristimulus values R, G, and B of the RGB color system, but includes a nonlinear correction term as the second term on the right side. The correction is made to the tristimulus values of the XYZ color system. This correction term makes it possible to adapt to a sensation measure for human color. Since the color system used in the present embodiment has a form obtained by adding a correction to the XYZ color system, it will be referred to as a corrected XYZ color system. In the expression 1, a to r represent the color TV camera 11
Is a coefficient corresponding to the spectral sensitivity of However, if the color TV camera 11 is determined, it can be treated as a constant.

[0017]

(Equation 1)

By applying Equation 2 to the tristimulus values X, Y, and Z of the corrected XYZ color system obtained from the tristimulus values R, G, and B of the RGB color system, the hue (H) and the saturation (C) , Lightness (L). Hue (H), saturation (C), lightness (L)
Calculates the tristimulus values X, Y, and Z in the CIELAB color space (L ^* a
^(* b ^* color space), it can be obtained based on the coordinate values L ^* , a ^* , b ^* in the CIELAB color space.

[0019]

(Equation 2)

Here, according to equations (1) and (2), the hue (H), saturation (C), and lightness (L) are obtained by using only the tristimulus values R, G, and B of the RGB color system as variables. , The tristimulus values R, G, B of the RGB color system and the hue (H),
A table for associating the saturation (C) and the lightness (L) can be created. Therefore, by providing such a table as the lookup table 14, the hue (H), saturation (C), saturation (C), and the like can be obtained from the signal values for each pixel stored in the image capture memories 13R, 13G, and 13B. The brightness (L) can be obtained immediately. This process is CP
U10 performs the operation and stores the values and operation results in the operation memory 15. That is, the hue (H), saturation (C), and lightness (L) of each pixel are stored in the calculation memory 15.

In the present embodiment, a display device 16 such as a CRT is used, and an image is displayed on the display device 16 through a display circuit 17. The input device 18 has a function of designating a portion to be displayed on the display device 16. In the present embodiment, when extracting a specific color range from a color image captured by the color TV camera 11, it is possible to select a color having gloss or a color having no gloss. In other words, the color to be extracted is determined by performing “reference color registration / selection processing” for selecting the color to be extracted from the colors displayed on the screen of the display device 16, and then it can be regarded as the same color as the determined color. A “color extraction process” is performed in which an area including a color within the range is automatically extracted and only the extracted area is displayed on the display device 16. As shown in FIG. 2, a menu in which “1.1 glossy color” and “1.2 non-glossy color” can be selected is displayed on the screen of the display device 16 to determine which area is to be extracted. Instructions are available.

When an instruction to extract a glossy color is issued, the processing is performed according to the procedure shown in FIG. First, when an object is imaged by the color TV camera 11, the image is stored in the image capturing memories 13R, 13G, and 13B, and a color image is displayed on the display device 16 through the display circuit 17. Here, the user specifies an area to be extracted by using the input device 18. For example, as shown in FIG.
The object 1 is displayed on the screen of the display device 16,
When there is an area D1 to be extracted in the object 1 and an area D2 that does not need to be extracted, a reference area Db for extracting the color of the area D1 to be extracted is designated using a mouse or the like (S1). That is, the display device 16 and the input device 18 constitute reference area setting means.
The reference area Db is specified within a range that can be regarded as substantially the same color.
When the reference area Db is designated, the CPU 10 obtains two values of R / B and G / B for each pixel in the reference area Db (S2). After that, the average values RBm and GBm of the values of the respective types are obtained, and the obtained average values RBm and GB are obtained.
An upper limit value and a lower limit value to be extracted are set by adding and subtracting prescribed margin values α and β to m, respectively. That is, for R / B, the upper limit value t1 = RBm + α and the lower limit value t2 = RBm−α
And the upper limit s1 = GBm + β and the lower limit s2 = GBm−β for G / B. Obtained upper limit values t1, s1
And the lower limit values t2 and s2 are stored in the calculation memory 15 (S3). Upper limit value t1, s1 and lower limit value t2, s2
Is determined, the range to be extracted for the value of R / B and the value of G / B is RBm−α <R / B <RBm + αG
R / which satisfies the condition of Bm-β <G / B <GBm + β
That is, the value of B and the value of G / B. Here, if the variance of each type of value R / B and G / B in the area specified in the screen of the display device 16 is used as the margin values α and β, the upper limit value and the lower limit value are automatically set. Can be determined. Alternatively, the variance may be corrected according to a certain rule, and the values may be used as margin values α and β. As described above, the operation of the CPU 10 implements a first extraction range setting unit.

After the range of colors to be extracted is determined as described above, pixels belonging to the specified color range are extracted from the image displayed on the screen of the display device 16 and the screen of the display device 16 is extracted. Perform the "color extraction process" displayed in. The color extraction process has a procedure shown in FIG. 5, and the user specifies an area with the input device 18 while the target object is displayed on the display device 16. However, as shown in FIG. 6, the designated inspection area Da is set to include the area D1 to be extracted. The inspection area setting means for setting the inspection area Da is realized by the display device 16 and the input device 18. Inspection area D
When “a” is designated, as shown in FIG. 5, the CPU 10 performs R / B and R / B conversion for each pixel in the inspection area Da.
Two values of G / B are obtained (S1). When both of the obtained two types of values satisfy the condition of being within the range defined by the upper limit values t1, s1 and the lower limit values t2, s2 set in the “reference color registration / selection process” (S2),
The pixel is extracted (S3), and if the condition is not satisfied, the pixel is not extracted (S4). If such processing is performed for all the pixels in the inspection area Da (S5), a group of pixels within the same color range as the color of the pixels extracted from the reference area Db can be extracted from the inspection area Da. The pixels extracted from the inspection area Da in this way are displayed on the screen of the display device 16 as shown in FIG. 7, and the pixels not extracted from the inspection area Da and the pixels outside the inspection area Da Also R = G
= B = 0 and is not displayed on the display device 16. In short, only the pixels extracted from the color image are displayed on the screen of the display device 16.

The above-described processing is for the case of extracting a glossy area. When extracting a non-glossy area, the processing is performed in the following procedure. In this case, similarly to the case where a glossy area is to be extracted, first, “reference color registration / selection processing” is performed. In this process, FIG.
As shown in (1), the user first specifies the reference area Db including the color of the area D1 to be extracted from the color image displayed on the display device 16 by the input device 18 (S1). The reference area Db is specified in a range that can be regarded as substantially the same color. When the reference area Db is specified, the CPU 10 determines the hue (H), the saturation (C), and the brightness (L) for each pixel in the reference area Db. (S2). That is, the tristimulus values R,
G and B are converted into hue (H), saturation (C), and lightness (L). Thereafter, average values Hm, Cm, and Lm are determined for the hue (H), saturation (C), and lightness (L), respectively, and the determined average values Hm, Cm, and Lm are respectively replaced with prescribed margin values γ, δ, and ε. An upper limit value and a lower limit value to be extracted by addition and subtraction are set. That is, for the hue (H), the upper limit h
1 = Hm + γ, lower limit h2 = Hm−γ, and upper limit c1 = Cm + δ and lower limit c2 =
Cm−δ, and the upper limit l1 = L for the lightness (L)
m + ε, and the lower limit l2 = Lm−ε. The calculated upper limit h
1, c1, l1 and the lower limit values h2, c2, l2 are stored in the operation memory 15 (S3). Note that the margin values γ,
In δ and ε, hue (H), saturation (C), and lightness (L) in the reference area Db specified in the screen of the display device 16 are shown.
The upper limit and the lower limit may be automatically determined using the variance, or the upper limit and the lower limit may be set by correcting the variance. As described above, the CPU 10 implements a second extraction range setting unit.

After the range of colors to be extracted is determined as described above, pixels belonging to the specified range of colors are extracted from the image displayed on the screen of the display device 16 and the screen of the display device 16 is extracted. Perform the "color extraction process" displayed in. That is, the CPU 10 implements a color extracting unit. The color extraction processing for an object having no luster is shown in FIG.
First, the user designates the inspection area Da so as to cover the entire area D1 to be extracted by the input device 18 while the object is displayed on the display device 16. When the inspection area Da is designated, the CPU
In step 10, the hue (H), saturation (C), and lightness (L) are obtained for each pixel in the inspection area Da using the lookup table 14 (S1). For each pixel, the hue (H), saturation (C), and brightness (L) are all the upper limit values h1, c1, and l1 and the lower limit values h2, c2, and l2.
When (S2), the color difference is further obtained by using Equation 3, and when the color difference is within a specified range (S3),
The pixel is extracted (S4). In addition, when any of the hue (H), the saturation (C), and the lightness (L) deviate from the range defined by the upper limit h1, c1, l1 and the lower limit h2, c2, l2 in step S2, If the color difference exceeds the specified range in Equation 3, no pixel is extracted (S5). In Equation 3, ΔH ^* , ΔL ^* , and ΔC ^* are the hue (H), saturation (C), and brightness (L) of each pixel, respectively.
And the average values Hm, Cm, Lm (hue difference, lightness difference,
Saturation difference).

[0026]

(Equation 3)

If such processing is performed for all the pixels in the inspection area Da (S6), a pixel group within the same color range as the color of the pixels extracted from the reference area Db is extracted from the inspection area Da. Can be. The pixels extracted from the inspection area Da in this manner are displayed on the screen of the display device 16 as they are, and both the pixels not extracted from the inspection area Da and the pixels outside the inspection area Da are R
= G = B = 0. As a result, only the pixels extracted from the color image are displayed on the screen of the display device 16.

It should be noted that instead of obtaining the color difference from Equation 3,
4, the color difference may be obtained. Number 4 is CIELAB color sky
Color difference between ΔL^*, Δa^*, Δb^*Each
The brightness index L of each pixel^*, Color coordinates a^*, B^*And each
Average value L^*m, a^*m, b ^*m.

[0029]

(Equation 4)

As described above, since the method of extracting the difference between a user having a glossy color and a color having no glossiness by the user is different, the reproducibility of the color having the glossiness can be improved. Good colors can be extracted.

(Second Embodiment) In this embodiment, as in the first embodiment, the user selects a method for extracting a glossy color and a glossy color. . First
In the embodiment, R / B and G / B are used as variables of interest for extracting a glossy color. However, in this embodiment, RB and GB are used. That is, FIG.
As shown in (1), after the user designates a reference region Db for extracting the color of the region D1 to be extracted by using a mouse or the like (S1), the CPU performs processing on each pixel in the reference region Db.
10 finds two values, RB and GB, respectively (S2). After that, the average value RBm, GBm of each type of value
Is determined, and upper and lower limits to be extracted are set by adding and subtracting predetermined margin values α and β to the calculated average values RBm and GBm, respectively. That is, for RB, the upper limit value t1 = R
Bm + α, lower limit t2 = RBm−α, and upper limit s1 = GBm + β, lower limit s2 = GBm− for GB.
Becomes β. The obtained upper limit value t1, s1 and lower limit value t2,
s2 is stored in the operation memory 15 (S3). When the upper limit value t1, s1 and the lower limit value t2, s2 are determined, R
The ranges to be extracted with respect to the value of -B and the value of GB are RB value and GB value such that RBm-α <RB <RBm + α GBm-β <GB <GBm + β become.

After the range of colors to be extracted is determined as described above, pixels belonging to the specified range of colors are extracted from the image displayed on the screen of the display device 16 and the screen of the display device 16 is extracted. Perform the "color extraction process" displayed in. The color extraction process has a procedure shown in FIG.
The user specifies the inspection area Da with the input device 18 while the target object is displayed on the display device 16. When the inspection area Da is specified, the CPU 10 determines whether each pixel in the inspection area Da has RB, G-
Two kinds of values of B are obtained (S1). When the two types of values satisfy the condition that both values are within the range defined by the upper limit values t1 and s1 and the lower limit values t2 and s2 set in the “reference color registration / selection process” (S2), the pixel of interest is determined. Is extracted (S3), and if the condition is not satisfied, the pixel is not extracted (S4). If such processing is performed for all pixels in the inspection area Da, (S
5) A pixel group in the same color range as the color of the pixel extracted from the reference area Db can be extracted from the inspection area Da. The pixels extracted from the inspection area Da are displayed on the screen of the display device 16 as they are,
Pixels not extracted from inside a and pixels outside the inspection area Da are both treated as pixels of R = G = B = 0. As a result, only the pixels extracted from the color image are displayed on the screen of the display device 16.

The processing for extracting pixels having no gloss is the same as in the first embodiment, and the other configurations and operations are the same as those in the first embodiment, so description thereof will be omitted.

(Third Embodiment) In the above-described embodiment, the user determines whether or not the reference area Db is a glossy area. However, in the present embodiment, as shown in FIG. When the user designates Db (S1), it is determined whether the reference area Db should be processed in a glossy area or a non-glossy processing (that is, provided with a gloss determination means), and the determination result is used. This function is provided with a function of presenting to a user (that is, determination result presenting means). In the present embodiment, in order to determine whether or not the reference area Db has gloss, two types of values of R / B and G / B are used to determine whether the reference area Db has gloss or has no gloss. Is set in advance. Thus, for each pixel in the reference area Db, R
/ B, G / B are obtained (S2), and the average values RBm, GBm of the obtained R / B, G / B values are stored in the arithmetic memory 14, and the difference between the average values RBm, GBm and the judgment value is calculated by
It is determined whether or not the reference area Db is a glossy area (S
3). In short, the smaller of the difference between the average values RBm, GBm and the determination value is adopted as the determination result.

In order to make the above determination, in the present embodiment, as shown in FIG.
A menu screen having three options of “1. Register and select reference color”, “2. Color with gloss (for inspection)”, and “3. Color without gloss (for inspection)” is displayed. In the “reference color registration / selection processing”, the reference area Db is specified as described above, and it is determined whether or not the reference area Db is a glossy area. For example, if the reference area Db is an area having no gloss, a message such as “Please select a program with no gloss” is presented on the screen of the display device 16 as shown in FIG. (S4, S5).

After such a judgment result is output,
According to the judgment result, “2. Glossy color (for inspection)”
Alternatively, select “3. Glossless color (for inspection)”. Each process is performed according to the first embodiment or the second embodiment.
This is the same as selecting “1.1 glossy color” or selecting “1.2 non-glossy color” on the menu screen shown in FIG. 2 in the embodiment. However, since the reference region Db has already been selected, the process of selecting the reference region Db can be omitted. Other configurations and operations are the same as those of the first embodiment or the second embodiment.

(Fourth Embodiment) This embodiment is a third embodiment.
By using the same method as that of the first embodiment, it is determined whether the reference region Db specified by the user is a glossy region or a non-glossy region (provided with a mouth determination means). Instead of indicating to the user, subsequent processing is determined based on the determination result. That is, if the determination result is a glossy region, the processing is performed from step S3 of the first embodiment shown in FIG. 2 to determine the range of colors to be extracted, and the determination result indicates that the region has no gloss. If so, the process is performed from step S2 of the first embodiment shown in FIG. 8 to determine the range of colors to be extracted. Therefore, in the present embodiment, as shown in FIG. 15, a menu having two types of options, “1. Reference color registration / selection” and “2. Inspection”, is displayed on the screen of the display device 16.

In this embodiment, when "reference color registration / selection processing" is selected, the user designates a reference area Db from which a color is to be extracted, as shown in FIG.
1) R / B and G / B are obtained for each pixel in the reference area Db (S2), and the average value RB of the obtained R / B and G / B values
m, GBm are stored in the arithmetic memory 14, and the difference between the average values RBm, GBm and the determination value is used to determine the reference area Db.
It is determined whether or not is a glossy area (S3). When the result of the determination is a glossy region, as described in the first embodiment, the average value R / B and G / B values R
Upper limits t1 and c1 for providing glossy areas by giving margins α and β around Bm and GBm.
And the lower limit values t2 and s2 are set and stored in the operation memory 14 (S4). On the other hand, when the determination result is a region having no gloss, the hue (H) of each pixel in the reference region Db,
The average values of the saturation (C) and the lightness (L) are obtained, and the margins γ, δ, and ε are provided around the respective average values, so that the upper limit values h1, c1, and l1 for the region having no gloss are obtained. And the lower limit values h2, c2, and l2 are set and stored in the operation memory 14 (S5). As described above, the range of colors to be extracted can be set based on the pixels in the reference area Db (this procedure implements a reference color registration unit).

Thereafter, when "2. Inspection" is selected, the first
First, the inspection area Da is specified as in the embodiment. Here, depending on whether the reference area Db has been determined to be a glossy area or a non-glossy area, the processing of the glossy area or the non-glossy area is selected and executed. (The selection means is realized by this procedure). That is, the processing of FIG. 5 or 11 is performed, or the processing of FIG. 9 is performed. Other configurations and operations are the same as those of the first embodiment.

(Fifth Embodiment) In each of the above-described embodiments, a process for determining a range of colors to be extracted from the reference region Db, a "reference color registration / selection process", and a process for selecting a desired color in the inspection region Da The process of extracting the pixels of “color extraction process” is not performed continuously, and the user starts the “color extraction process” after the “reference color registration / selection process”. After the “reference color registration / selection processing”, the “color extraction processing” is automatically activated without any operation by the user.

That is, the processing procedure of this embodiment is the same as that of FIG.
Or as shown in FIG. As is clear from the above-described embodiments, the “reference color registration / selection processing” requires the user to specify the reference area Db, and the “color extraction processing” requires the user to specify the inspection area Da. is there. Therefore, in order to prevent the user from performing an operation between the “reference color registration / selection process” and the “color extraction process”, the process shown in FIG.
As shown by 0, it is necessary to specify both the reference area Db and the inspection area Da.

As shown in FIG. 17, first, both the reference area Db and the inspection area Da are designated in a state where the color image picked up by the color TV camera 11 is displayed on the display device 16 (S1). Next, R / B and G / B are obtained for each pixel of the reference area Db (S2), and the obtained R is calculated.
The average values RBm and GBm of the values / B and G / B are stored in the arithmetic memory 14, and the difference between the average values RBm and GBm and a predetermined determination value determines whether the reference region Db is a glossy region. A determination is made (S3). This processing implements a gloss determination unit. When the determination result is a glossy region, as described in the first embodiment, the widths of the margin values α and β around the average values RBm and GBm of the R / B and G / B values are set. By giving
Upper limit values t1 and s1 and lower limit value t to be a glossy region
2, s2 are set and stored in the operation memory 14 (S4),
The process proceeds to the glossy color extraction process shown in FIG. 18 (S
5). On the other hand, when the determination result is a region having no gloss, the values of the tristimulus values R, G, and B of each pixel in the reference region Db are converted into hue (H), saturation (C), and brightness (L). (S
6) The upper limit value h1 for the region having no gloss by giving the widths of the margins γ, δ, ε around the average values of the obtained hue (H), saturation (C), and lightness (L). , C1,
l1 and the lower limit values h2, c2, and l2 are set and stored in the operation memory 14 (S7), and the process proceeds to the glossless color extraction process shown in FIG. 19 (S8). In short, an appropriate color extraction process is automatically selected based on the determination result of the gloss determination unit, and this process implements an automatic selection unit.

Incidentally, in step S4, the upper limit value t
1, s1 and lower limit values t2, s2,
The processing automatically shifts to the processing in FIG. The processing shown in FIG. 18 is the same as the color extraction processing in the case of having gloss in the first embodiment shown in FIG.
The two types of values R / B and G / B are obtained for each pixel in (S9). When satisfying the condition that both of the obtained values are within the range defined by the upper limit values t1 and s1 and the lower limit values t2 and s2 (S10),
The pixel is extracted (S11), and if the condition is not satisfied, the pixel is not extracted (S12). If such processing is performed for all the pixels in the inspection area Da (S13), it is possible to extract from the inspection area Da a pixel group that is within the same color range as the color of the pixel extracted from the reference area Db. Pixels extracted from the inspection area Da in this way are, as shown in FIG.
And the pixels not extracted from the inspection area Da and the pixels outside the inspection area Da are both treated as R = G = B = 0 pixels, and as a result, are extracted from the color image. Only the pixels are displayed on the screen of the display device 16.

On the other hand, in step S7, the upper limit value h1,
When c1, l1 and the lower limit values h2, c2, l2 are set, the processing automatically shifts to the processing in FIG. The process shown in FIG. 19 is the same as the color extraction process in the case where there is no gloss in the first embodiment shown in FIG. Is used to determine hue (H), saturation (C), and lightness (L) (S14). When all of the hue (H), saturation (C), and lightness (L) of each pixel are between the above-described upper limit values h1, c1, and l1 and the lower limit values h2, c2, and l2 (S15), The color difference is obtained in the same manner as in the first embodiment, and when the color difference is within a specified range (S16), the pixel is extracted (S17). In step S15, any one of the hue (H), the saturation (C), and the lightness (L) is set to the upper limit h1, c1, l1 and the lower limit h.
If the color difference deviates from the range defined by 2, 2, and 2, or if the color difference exceeds the specified range in step S16, no pixel is extracted (S18). If such processing is performed for all the pixels in the inspection area Da (S19), it is possible to extract from the inspection area Da a pixel group that is in the same color range as the color of the pixel extracted from the reference area Db. The pixels extracted from the inspection area Da in this way are displayed as they are on the screen of the display device 16, and the pixels not extracted from the inspection area Da and the pixels outside the inspection area Da are both R = G = B = It is treated as a 0 pixel. As a result, only the pixels extracted from the color image are displayed on the display device 16 as shown in FIG.
Will be displayed on the screen.

Other structures and operations are the same as those of the first embodiment. In the third to fifth embodiments, when determining the range of colors to be extracted from the reference area Db when gloss is present, R / R
It is assumed that the values of B and G / B are used. However, as in the second embodiment, this process is performed using RB and G / B.
It goes without saying that the value of -B may be used. In addition, in the third to fifth embodiments, the reference region D
Although the average value of R / B and G / B is obtained for the pixel in b and compared with the determination value, the average value of R-B and GB is calculated and compared with the determination value. Good.

A region having a gloss using R / B and G / B has a condition for judging any of gold, silver, and copper, and has a gloss using RB and GB. Table 1 shows an example of a condition for determining whether the region is gold, silver, or copper. The determination condition means that when the condition in the cell is satisfied, the colors are gold, silver, and copper, respectively.

[0047]

[Table 1]

[0048]

According to the first aspect of the present invention, there is provided a reference area setting means for designating a reference area including a color to be extracted from a color image having a tristimulus value of an RGB color system as a value of each pixel, and a reference area. First extraction range setting means for determining a range of the same color in the RGB color system when the inside has a gloss, and a range of the same color based on hue, saturation, and brightness when the reference region has no gloss. Second extraction range setting means for determining, inspection area setting means for specifying an inspection area for a color image from which pixels of the same color as the reference area are to be extracted, and whether the color to be extracted from the inspection area has gloss. A color extraction unit that uses a range of the same color determined by one of the first extraction range setting unit and the second extraction range setting unit depending on whether the pixel is within the range and extracts pixels within the range from the inspection area Color to be extracted is glossy The range of colors to be extracted from the inspection area is made different depending on whether or not it is extracted from the area that has the gloss, and the color extraction is reproduced by applying the technology appropriate for the case with and without the gloss. It can be done well. In addition, when there is gloss, colors are extracted using the RGB color system, and when there is no gloss, hue, saturation, and lightness are used. Therefore, it is possible to extract colors close to the human interval scale. .

According to a second aspect of the present invention, in the first aspect of the present invention, the first extraction range setting means sets R / B and G /
The range of the same color is determined centering on each average value of the values of B, and a highly sensitive color can be extracted by performing the division.

According to a third aspect of the present invention, in the first aspect of the invention, the first extraction range setting means sets the RB and G-
The range of the same color is determined centering on each average value of the values B and B. There is no need to consider significant figures or division by 0, and the calculation is easy.

According to a fourth aspect of the present invention, in the first to third aspects of the present invention, the average value of the R / B and G / B values of all the pixels in the reference area or the total value of the R / B and G / B values in the reference area is obtained. A gloss judging unit for judging whether or not the reference area has gloss based on a magnitude relationship between each of the average values of the RB and GB values of the pixel and a predetermined judgment value; It is provided with a judgment result presenting means for presenting the judgment result, and by presenting the judgment result on the presence or absence of gloss to the user, even a user with little knowledge of the color can extract the color. .

According to a fifth aspect of the present invention, in the first to third aspects, the average value of the R / B and G / B values of all the pixels in the reference area or the total value of the R / B and G / B values in the reference area is obtained. A gloss judging unit for judging whether or not the reference area has gloss based on a magnitude relationship between each of the average values of the RB and GB values of the pixel and a predetermined judgment value; Reference color registration means for registering the determination result, and selection means for selecting the first extraction range setting means and the second extraction range setting means using the result registered by the reference color registration means, By registering the presence or absence of gloss by the reference color registration unit, the appropriate one of the first extraction range setting unit and the second extraction range setting unit is automatically selected, and the usability is improved.

According to a sixth aspect of the present invention, in the first to third aspects of the present invention, the average value of the R / B and G / B values of all the pixels in the reference area or the total value of the R / B and G / B values in the reference area is obtained. A gloss judging unit for judging whether or not the reference area has gloss based on a magnitude relationship between each of the average values of the RB and GB values of the pixel and a predetermined judgment value; The apparatus includes an automatic selection unit that selects the first extraction range setting unit and the second extraction range setting unit based on the determination result. It is not necessary for the user to specify the glossiness. In spite of this, it is possible to automate the color extraction, and as a result, it is possible to cope with a case where a glossy member and a non-glossy member are conveyed together in a factory line or the like.

According to a seventh aspect of the present invention, when a reference area designated to include a color to be extracted for a color image having tristimulus values of the RGB color system as the values of respective pixels is included in the RGB, when there is gloss, When the same color range is determined in the color system, and when there is no gloss, the same color range is determined in a color system different from the RGB color system, and then the pixels of the same color as the reference area are extracted. The method is characterized in that one of the two types of ranges is selected according to whether or not the color to be extracted from the inspection area set has gloss, and pixels within the range are extracted from the inspection area. The range of colors to be extracted from the inspection area is made different depending on whether the color to be extracted is extracted from the glossy area, and the appropriate technology is applied to the case with or without the gloss. Color reproducibility It can be carried out. In addition, when there is gloss, colors are extracted using the RGB color system, and when there is no gloss, hue, saturation, and lightness are used. Therefore, it is possible to extract colors close to the human interval scale. .

According to an eighth aspect of the present invention, in accordance with the seventh aspect of the present invention, the magnitude relationship between the average value of the R / B and G / B values of all the pixels in the reference area and a predetermined judgment value is determined. It is characterized in that it is determined whether or not the reference area has gloss. By performing the division, a highly sensitive color can be extracted.

According to a ninth aspect of the present invention, in accordance with the seventh aspect of the present invention, the average value of RB and GB values of all the pixels in the reference area is determined by a magnitude relation between a predetermined judgment value and each average value. It is characterized in that it is determined whether or not the inside of the reference area has gloss, and there is no need to consider significant figures or division by 0, and the calculation is easy.

[Brief description of the drawings]

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

FIG. 2 is a diagram showing a menu screen of the above.

FIG. 3 is an operation explanatory diagram of the above.

FIG. 4 is a diagram illustrating a reference region in the above embodiment.

FIG. 5 is an operation explanatory view of the above.

FIG. 6 is a diagram illustrating an inspection region in the above.

FIG. 7 is a diagram showing a display example of a result in the above.

FIG. 8 is an operation explanatory view of the above.

FIG. 9 is an operation explanatory view of the above.

FIG. 10 is an operation explanatory view showing a second embodiment of the present invention.

FIG. 11 is an operation explanatory diagram of the above.

FIG. 12 is a diagram showing a menu screen according to a third embodiment of the present invention.

FIG. 13 is a diagram showing a display example of a message in the above.

FIG. 14 is an operation explanatory view showing a fourth embodiment of the present invention.

FIG. 15 is an explanatory diagram of the operation of the above.

FIG. 16 is an operation explanatory view of the above.

FIG. 17 is an operation explanatory view showing a fifth embodiment of the present invention.

FIG. 18 is an operation explanatory view of the above.

FIG. 19 is a diagram illustrating the operation of the above.

FIG. 20 is a diagram illustrating a reference region and an inspection region in the above.

FIG. 21 is a diagram showing a display example of a result in the above.

[Explanation of symbols]

 11 color TV camera 12 input device 13R, 13G. 13B Image capture memory 14 Look-up table 15 Operation memory 16 Display device 17 Display circuit 18 Input device

──────────────────────────────────────────────────の Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat ゛ (Reference) G01N 21/94 G01N 21/94 5C066 H04N 17/02 H04N 17/02 Z 5L096 F-term (Reference) 2G020 AA04 DA02 DA03 DA04 DA05 DA06 DA13 DA52 2G051 AA73 AB14 AB20 BA08 CA04 CB01 EA11 EA14 EA17 EA23 EB01 EB02 ED01 ED04 ED07 ED30 2G059 AA05 BB16 EE02 EE13 FF01 HH02 KK04 LL04 MM03 MM01 MM05 MM05 MM05 AMM CH08 5C061 BB01 CC01 CC05 5C066 AA01 AA03 CA11 ED00 GA01 GA02 GA33 GB00 KD02 KD06 KE02 KE03 KE05 KE09 KG01 KP02 5L096 AA02 AA05 BA03 CA02 DA01 FA15 GA07

Claims (9)

[Claims] 1. A reference area setting means for designating a reference area including a color to be extracted from a color image having a tristimulus value of an RGB color system as a value of each pixel. First to determine the same color range in RGB color system
Extraction range setting means, second extraction range setting means for determining the same color range based on hue, saturation and lightness when the reference area has no gloss, and extracting pixels of the same color as the reference area. Inspection area setting means for specifying an inspection area for a color image to be extracted, and first and second extraction range setting means according to whether or not a color to be extracted from the inspection area has gloss. A color extraction unit that uses a range of the same color determined by one of the above and extracts pixels within the range from the inspection area. 2. The method according to claim 1, wherein the first extraction range setting means sets the R of all pixels.
2. The color extraction image processing apparatus according to claim 1, wherein a range of the same color is determined centering on each average value of the values of / B and G / B. 3. The first extraction range setting means sets R
2. The color extraction image processing apparatus according to claim 1, wherein the same color range is determined centering on each average value of -B and GB. 4. The R / B and G / R of all pixels in the reference area.
B and the average value of all the pixels in the reference area.
A gloss judging unit for judging whether or not the reference area has gloss based on a magnitude relation between each of the average values of -B and GB and a preset judgment value; The color extraction image processing apparatus according to any one of claims 1 to 3, further comprising a determination result presenting means for presenting. 5. The R / B and G / G of all pixels in the reference area.
B and the average value of all the pixels in the reference area.
A gloss judging unit for judging whether or not the reference area has gloss based on a magnitude relation between each of the average values of -B and GB and a preset judgment value; A reference color registration means for registering, and a selection means for selecting a first extraction range setting means and a second extraction range setting means using a result registered by the reference color registration means. Any one of claims 1 to 3
Item 10. The color extraction image processing device according to Item 1. 6. The R / B and G / G of all pixels in the reference area.
B and the average value of all the pixels in the reference area.
A gloss judging unit for judging whether or not the reference area has gloss based on a magnitude relation between each of the average values of -B and GB and a preset judgment value; The color extraction image processing according to any one of claims 1 to 3, further comprising an automatic selection unit that selects the first extraction range setting unit and the second extraction range setting unit using the selection unit. apparatus. 7. When a reference area designated to include a color to be extracted for a color image having a tristimulus value of the RGB color system as a value of each pixel includes a color to be extracted.
When the same color range is determined in the color system and there is no gloss, R
A range of the same color is determined in a color system different from the GB color system, and then the color to be extracted from the inspection region set for the color image from which a pixel of the same color as the reference region is to be extracted has gloss. A color extracting method, wherein one of the two types of ranges is selected depending on whether or not the pixels are selected, and pixels within the range are extracted from the inspection area. 8. The R / B and G / R of all pixels in the reference area
8. The color extraction method according to claim 7, wherein it is determined whether or not the reference area has a gloss based on a magnitude relationship between each average value of B and a predetermined determination value. 9. The RB and G-pixels of all pixels in the reference area.
8. The color extraction method according to claim 7, wherein it is determined whether or not the reference area has a gloss based on a magnitude relationship between each average value of B and a predetermined determination value.