JP3240570B2 - Color image binarization processing device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a technique for processing and recognizing a color input image obtained by imaging an object in a solder inspection machine, an intelligent robot, and the like. The present invention relates to a color image binarization apparatus that generates a binary image by binarizing a color input image with a binarization threshold value calculated from the color input image.

[0002]

2. Description of the Related Art As shown in FIG. 25, for example, as a solder inspection machine for inspecting the quality of soldering, a color television camera 4 is located above a substrate 1 to be inspected, and a different elevation angle with respect to the inspection position. There is a configuration in which annular light sources 5R, 5G, and 5B that emit hue light (for example, red light, green light, and blue light) are installed.

This solder inspection machine simultaneously irradiates the substrate 1 to be inspected with red light, green light and blue light from the light sources 5R, 5G and 5B, and reflects each hue reflected on the surface of the soldering portion 3 of the component 2. The light is observed by the color television camera 4. The image obtained by the television camera 4 is taken into the image processing device 6, and an image pattern corresponding to the curved surface property of the soldering portion 3 is binarized for each hue to obtain a hue pattern of three primary colors. Then, each hue pattern is compared with a reference pattern to determine the quality of the soldering at the soldering portion 3.

[0004]

By the way, the surface of the substrate 1 to be inspected, the surface of the component 2 and the characters displayed on the surface of the component 2 have unique colors. Although the intensity is different from that of the reflected light image, the hue may be common. Further, the color input image from the television camera 4 may have color unevenness, and for example, a reddish portion, a bluish portion, or the like may be present in addition to the image portion of the soldering portion 3.

When such a color input image is binarized for each hue to generate each hue pattern, there is a possibility that not only the soldering portion 3 but also the patterns of other portions are mixed in each hue pattern. This makes it difficult to correctly determine the quality of soldering. In addition, when a color input image is subjected to a binarization process, conventionally, an optimal binary image is compared while comparing a color input image with a binary image obtained by binarizing the color input image.
The threshold for the binarization is determined, so that it takes a lot of trouble and time to obtain the optimal threshold for the binarization.
There is also a problem that.

The present invention has been made in view of the above-mentioned problem. Even when a color input image includes a portion having a common hue and a color hue, the optimum binarization threshold value is obtained. It is an object of the present invention to provide a color image binarization processing device capable of automatically setting a color pattern and extracting a hue pattern of only a target portion.

[0007]

According to a first aspect of the present invention, there is provided a binarization process for binarizing a color input image with a binarization threshold value calculated from the color input image to generate a binary image. A processing device, comprising: a pixel extraction unit for extracting pixels constituting a color input image for which the binarization threshold is calculated; and a binarization threshold from pixel data of each pixel extracted by the pixel extraction unit. A binary threshold value calculating means for calculating a value, and a binary image for generating a binary image by subjecting the color input image to a binary processing using the binary threshold value calculated by the binary threshold value calculating means Generating means.
The pixel extraction unit inputs pixel data of each pixel constituting the color input image and measures the intensity of each hue for each pixel, and a condition setting for setting a condition for the intensity of each hue. Means for extracting only pixels whose intensity of each hue measured by the measuring means satisfies the condition set by the condition setting means, and outputting the pixel data to the binary threshold calculating means And

[0008]

According to a second aspect of the present invention, there is provided a color image binarization processing apparatus for inputting pixel data of each pixel constituting the color input image and measuring the intensity of each hue for each pixel. Measuring means, second measuring means for inputting pixel data of each pixel constituting the color input image and measuring brightness for each pixel, and hue of each pixel measured by the first measuring means . First threshold value calculating means for calculating a first binary threshold value from the intensity, and second binary value conversion from the brightness of each pixel measured by the second measuring means. A second threshold value calculating means for calculating a threshold value, a specifying means for specifying a hue to be subjected to the binarization processing, a first binary threshold value for the hue specified by the specifying means, A binarization process is performed using the second binarization threshold value to obtain 2
A binary image generating means for generating a value image.

[0010]

In the binarization processing device according to the first aspect of the present invention,
A color input image is captured by setting conditions for the intensity of each hue, and the intensity of each hue is measured from pixel data of each pixel constituting the color input image. Only pixels whose measured values satisfy the set conditions are extracted, and a binarization threshold is calculated using only the extracted pixel data of each pixel.

[0011]

[0012] In the binarization processing apparatus according to a second aspect of the invention, when the color input image is captured, the color intensity than the pixel data of each pixel constituting the color input image is measured, the hue of each pixel Binarization threshold from the intensity of
A value is calculated. Further, the brightness is measured for each pixel from the pixel data of each pixel forming the color input image, and a second binarization threshold is calculated from the brightness for each pixel. When the hue to be binarized is specified, the binarization is performed using the first and second binarization thresholds for the specified hue. Thus, a binary image is generated.

In any of the binarization processing devices, even if an image portion and an uneven color portion of a portion having the same hue as the target portion are mixed in the color image, the difference between the target portion and the target portion occurs due to the difference in hue intensity. The hue pattern of only the part to be extracted is extracted.

[0014]

FIG. 1 shows a schematic configuration of a color image binarization processing apparatus 10 according to a first embodiment of the present invention. This 2
The binarization processing device 10 includes a pixel extraction unit 11, a threshold value calculation unit 12, and a binary image generation unit 13. The output of the color television camera 14 is supplied to the pixel extraction unit 11, and the binary image generation unit 13 Are connected to a display unit 15 such as a CRT.

The color television camera 14 captures an image of an object under illumination by a predetermined light source to generate a color input image. In the case of a solder inspection machine, the object is a component on a substrate to be inspected, and its illumination source is, for example, an annular light source that emits red light, green light, and blue light (see FIG. 25).
The pixel data of each pixel constituting the color input image is three primary colors, that is, red, green, and blue luminance data R,
G and B.

The pixel extracting section 11 extracts pixels constituting a color input image for which a binarization threshold is to be calculated. The threshold value calculation unit 12 calculates an optimal binarization threshold value from the pixel data of each pixel extracted by the pixel extraction unit 11. The binary image generation unit 13 includes a threshold value calculation unit 12
The color input image is divided into two by the binarization threshold value calculated in
The binarization process is performed to generate a binary image.

The pixel extraction unit 11 includes a hue degree calculation unit 1
6, a condition setting unit 17, a condition determination unit 18, and a gate unit 19. The hue degree calculator 16 scans the color input image, and uses the pixel data (red, green, and blue luminance data R, G, and B) of each pixel read in accordance with the scanning to perform the following (1) to (1). The calculation of equation (3) is executed to calculate the intensity (hereinafter, referred to as “hue degree”) r, g, and b of each hue for each pixel.

[0018]

(Equation 1)

[0019]

(Equation 2)

[0020]

(Equation 3)

The hue degrees r, g, and b are proportional to the degree of recognition of the three primary colors in human vision.
If the hue degree g is 0 or less, green is not recognized, and if the hue degree b is 0 or less, blue is not recognized.

The condition setting section 17 is composed of, for example, a keyboard, and sets conditions for each hue degree r, g, b of each pixel to be subjected to threshold value calculation, that is, each hue degree r,
This is for setting the range of g and b by the upper limit and the lower limit. The setting of the condition is not limited to a method of operating a keyboard to input a value by a key, but may be, for example, a method of reading from a file.

FIG. 2 shows a procedure of a key input operation for setting a condition from step 1 (“ST1” in the figure) to step 1
The reference numeral 8 is used. In step 1 of the figure, a guidance character (message) is displayed on the display unit 15 so as to input the upper limit value TH (r) ₁ of the red hue degree r by key. When the upper limit value TH (r) ₁ is inputted by a key in the next step 2, the key input data is fetched and stored in a storage means such as a RAM (step 3).

Next, at step 4, a guide character is displayed on the display unit 15 so as to input a lower limit value TH (r) ₂ of the red hue degree r by a key. At step 5, the value is input by a key. The key input data is captured and stored in the storage means (step 6).

The upper limit TH (g) ₁ and the lower limit TH (g) of the green hue g in steps 7 to 12 will be described below.
_In steps 13 to 18, the upper limit value TH (b) ₁ and the lower limit value TH (b) ₂ of the blue hue b are key-inputted and stored in the same procedure.

If it is desired to binarize the red image portion, the upper limit value TH (r) ₁ for the red hue r is set to "67", the lower limit value TH (r) _{2 is set} to "-33", and The upper limit value TH (g) ₁ of the hue degree g is “0”, the lower limit value TH (g) ₂ is “−33”, and the upper limit value TH (b) ₁ of the blue hue degree b is “0”. , Lower limit value TH (b) ₂
Is set to “−33”.

Returning to FIG. 1, the condition judging section 18 calculates the hue degrees r, g, and h of each pixel calculated by the hue degree calculating section 16.
Whether or not b satisfies the condition set by the condition setting unit 17, that is, the green hue falls within a range where the red hue degree r is defined by an upper limit value TH (r) ₁ and a lower limit value TH (r) _2. In a range where the degree g is defined by the upper limit value TH (g) ₁ and the lower limit value TH (g) ₂ , the hue degree b of blue is defined by the upper limit value TH (b) ₁ and the lower limit value TH (b) _2. It is determined whether each of the ranges is included.

If all the conditions are satisfied, the condition judging unit 18 opens the gate unit 19, and if any of the conditions is not satisfied, closes the gate unit 19. The pixel data of each pixel constituting the color input image is input to the gate unit 19, and only the pixels satisfying the above conditions are extracted by opening and closing the gate unit 19, and the pixel data is used as the threshold value calculation unit 12 Output to The functions of the hue degree calculation unit 16, the condition determination unit 18, and the gate unit 19 may be realized by software, or may be realized by a hardware configuration.

FIG. 3 is a diagram showing the above-mentioned two steps for generating a binary image.
Step 1 to Step 1
Indicated by 7. First, in step 1, the condition setting unit 17 sets the upper and lower limits for each of the hue degrees r, g, and b. In step 2, the color TV camera 14 converts the color input image from the binary image data into It is taken in. The hue degree calculation unit 16 of the image extraction unit 11 starts scanning the color input image, and first, pixel data of the pixel at the position of X = 0, Y = 0 is sampled, and each of the hue degrees r of red, green, and blue is sampled. , G and b are calculated (steps 3 to 3).
5).

In the next step 6, the condition judging section 18 determines that each of the hue degrees r, g, b is TH (r) ₁ >r> TH (r) ₂ ,
TH (g) ₁ >g> TH (g) ₂ : TH (b) ₁ >b>
It is determined whether each condition of TH (b) ₂ is satisfied.

If all the conditions are satisfied, the determinations in steps 7 to 9 are "YES" and the process proceeds to step 10, where the pixel data of the pixel at the position of X = 0, Y = 0 is stored in the gate unit. 19, and is output to the threshold value calculation unit 12. If any one of the conditions is not satisfied, one of the determinations in Steps 7 to 9 becomes “NO”, and
The pixel data does not pass through the gate unit 19, and
0 is skipped.

Next, at step 11, the same procedure as described above is executed by designating the pixel at the next X coordinate position (X = 1 in this case), and all the pixels (one horizontal scan) on the same horizontal scan line are executed. When the processing for (the number of pixels on the line is 256) is completed, the determination in step 12 is “YE
S ", and shifts to processing for each pixel on the next horizontal scanning line (step 13).

When the processing for all the horizontal scanning lines (the number of horizontal scanning lines is 256) is completed, the determination in step 14 becomes "YES" and the process proceeds to step 15, where the threshold value calculation unit 12 An optimum binarization threshold is calculated from the pixel data of each pixel extracted by the extraction unit 11. Next, in step 16, the binary image generation unit 13 binarizes the color input image using the binarization threshold to generate a binary image, and the binary image is displayed on the display unit 15. (Step 17).

FIG. 4 shows an example of a specific circuit configuration of the threshold value calculator 12. In the illustrated example, the threshold value calculation unit 12
This method calculates an optimal binarization threshold value from an image quality evaluation value. The details of this method are described in JP-A-64-515.
No. 86, here, the overall configuration and its principle are outlined.

In the illustrated example, the threshold value calculating section 12 is mainly controlled by a microcomputer 20 and has a window scanning section 2.
1, illegal pattern detection unit 22, illegal pattern histogram generation unit 23, brightness histogram generation unit 24, image quality histogram generation unit 25, histogram smoothing processing unit 2
6, an optimum threshold value search unit 27 is provided.

The window scanning unit 21 includes an address generator 29 connected to the microcomputer 20 via a CPU bus 28, and an address generator 29.
An image memory 31 connected to the image bus 30 via an image bus 30;
FIF connected to the data bus 33 of the image bus 30
A first shift register 35 of the O (first-in first-out) system and a first group of latch circuits 36 to 39, and a second shift register 40 of the same FIFO system connected to the last-stage latch circuit 39 And the second group of latch circuits 4
1 to 44.

The window scanning unit 21 in the illustrated example forms a rectangular image as shown in FIG. 5 on the color input image written in the image memory 31, that is, the color input image that has passed through the gate unit 19 of the pixel extracting unit 11. set the window 50 be used to raster scan, the pixel data d ₀ to d ₇ of predetermined pixels a ₀ to a ₇ per image in the window 50 is taken out at the same time. In the figure, the crosses indicate pixels that are not to be processed, and pixels that have not been extracted by the pixel extraction unit 11 are also treated as pixels that are not to be processed.

The address generator 29 is an image memory 3
1 is sequentially generated, and the pixel data of the pixel corresponding to the address is first generated via the image bus 30.
To the shift register 35. The image bus 30
Of these, an address bus 31 stores an address and a data bus 3
3 transmits pixel data, and the control bus 33 transmits an address strobe signal and a read / write signal.

The first shift register 35 stores the image memory 3
Assuming that the number of pixels in the horizontal direction of 1 is n (see FIG. 5), this n
The number of stages of the first group of latch circuits 36 to 39 is subtracted from the number of stages (n−4), and the output of pixel data is delayed by the number of stages. First group of four-stage latch circuits 36
To 39 perform the operation of the shift register as a whole, and the output of the final-stage latch circuit 39 is the second shift register 40.
Output to This second shift register 40 is also (n-
4) The output of the pixel data is delayed by the number of stages and transmitted to the second group of four-stage latch circuits 41 to 44.

In this way, pixel data is simultaneously extracted from the first and second group of latch circuits 36 to 39, 41 to 44 and the second shift register 40.
Pixel data d ₀ to d ₂ pixels A ₀ to A ₂ in the window 50 by the outputs of the first group of latch circuits 36 to 38
And second shift register 40 and second shift register 40
The output of each of the group latch circuits 41 to 44 causes a window 5
Pixel data d ₃ to d ₇ of the pixel A ₃ to A ₇ in the 0 is given.

The pixel data d _{0 to} d ₇ of each pixel A _{0 to} A ₇
Together are output to illegal pattern detection unit 22, is also output to the illegal pattern histogram generation unit 23 and the brightness histogram generation unit 24 for the pixel data d ₁ specific pixel A _1. The illegal pattern detection unit 22 is for determining whether the image in the window 50 is a legal pattern or an illegal pattern,
If it is an illegal pattern, a flag of "1" is outputted, and if it is a legal pattern, a flag of "0" is outputted.

[0042] The illegal pattern detection unit 22, the pixel data d of each pixel A ₀ and A ₂ to A ₇ of the window 50 the pixel data d ₁ specific pixel A ₁ as a threshold
_For binarizing ₀ and d _{2 to} d ₇ ,
Each pixel data d _i (i = 0, 2, 3, ‥‥, 7)
There "1" when the d _i ≧ d _1, when the d _i <d ₁ output the respective binary data of "0". in this case,
As the pixel data d _{0 to} d ₇ , luminance data R, G, and B of a hue to be binarized may be used, or a total value (R + G + B) of the luminance data may be used. Note that the specific pixel A ₁
Is always "1".

FIG. 6 shows each of the pixels A ₀ to A ₀ in the window 50.
Showing a process of pixel data d ₀ to d ₇ of A ₇ is converted is binarized into binary data f ₀ ~f _7. This binary data f
_i (i = 0, 2, 3,..., 7) is determined by a logical operation whether or not the image in the window 50 is an illegal pattern. If the pattern is an illegal pattern, a flag of “1” is set.
If it is a legal pattern, a flag of "0" is output. Here, the legal pattern means a binary pattern frequently appearing on a smooth image as shown in FIGS. 7 (1) to 7 (6), and the illegal pattern frequently appears on an unsmooth image in FIG. Means a binary pattern as shown in FIG. 7 and 8, hatched portions indicate pixels having binary data of "1". In the binary pattern of FIG. 8, a hole 51 of "0" is formed inside the pixel region of "1". .

Each time an illegal pattern is detected and a flag of "1" is output as described above, the illegal pattern histogram generator 23 counts the frequency of occurrence of this flag and outputs a histogram f _hist (k ). Here, f _hist (k) means the frequency of occurrence of illegal patterns at the threshold value k.

The brightness histogram 24 is a brightness histogram hist representing the frequency of occurrence of the pixel of the pixel data k.
(K), and after the generation of the illegal pattern histogram and the brightness histogram is completed, the image quality histogram generation unit 25 sets the image quality histogram q
Generate _hist (k). This image quality histogram q
_hist (k) gives an image quality evaluation criterion, in this embodiment, a criterion indicating poor image quality. When hist (k) = 0, the following equation (4) is used. of
Each is generated by equation (5). In equations (4) and (5),
Q ₀ is a constant (maximum value).

[0046]

(Equation 4)

[0047]

(Equation 5)

[0048] Then the histogram smoothing processing section 26 is for smoothing the image quality histogram q _hist (k) optionally, as a result, as shown in FIG. 9, the image quality histogram sq _hist smoothed (K) is generated. In FIG. 9, the horizontal axis is k (threshold), and the vertical axis is sq.
_hist (k) (bad image quality), and the bad image quality is minimum when the threshold value k _opt is used.

The optimum threshold value search section 27 searches for this threshold value k _opt as an optimum threshold value, and
The microcomputer 20 is notified through the PU bus 28.

FIG. 10 shows a schematic configuration of a color image binarization processing device 60 according to a second embodiment of the present invention. The binarization processing device 60 in the illustrated example includes a hue degree calculator 61, a threshold calculator 62, a binarization target designator 63, a threshold selector 64, and a binary image generator 65. The output of the color TV camera 14 is sent to the hue degree calculator 61,
The output of the binary image generation unit 65 is connected to the display unit 15.

The hue degree calculator 61 scans a color input image input from the color television camera 14, and stores pixel data (red, green,
The above (1) to (1) to (5) are used by using each of the blue luminance data R, G, B).
By executing the calculation of equation (3), the hue degrees r, g, and b are calculated for each pixel. These hue degrees r, g, and b are provided to the threshold value calculation unit 62 and the binary image generation unit 65.

The threshold value calculating section 62 calculates optimum binarized threshold values TH _R , TH _G , TH for each hue from the hue degrees r, g, b for each pixel inputted from the hue degree calculating section 61. _B
, And a method of obtaining an evaluation value of image quality and determining an optimal binarization threshold value is used as in the first embodiment. Therefore, the configuration of the threshold value calculation unit 62 is the same as that shown in FIG. 4, and illustration and description thereof are omitted here.

The binarization target designating section 63 is composed of a keyboard, a mouse, and the like, and has a hue to be binarized, that is, red, for an image area to be binarized.
This is for designating which hue of green or blue, or any combination of these colors.

The threshold value selecting section 64 stores the two-color image data corresponding to the hue specified by the binarized target specifying section 63.
The binarization threshold TH is selected from any of the binarization thresholds TH _R , TH _G , and TH _B calculated by the threshold calculator 62, and the selected binarization threshold TH is set to 2 Whether to use as the upper limit or the lower limit in the binarization process is set.

The binary image generating section 65 calculates any one of the hue degrees r, g, b calculated by the hue degree calculating section 61.
Is binarized with the binarization threshold value TH selected by the threshold value calculation unit 64 to generate a binary image.

FIG. 11 shows steps 1 to 7 of the control procedure of the binarization processing device 60 for generating a binary image. First, in step 1, when a color input image is captured by the color television camera 14 into the binarization processing device 60, first, the hue degree calculation unit 61 scans the color input image and scans each hue degree r, g, b for each pixel. Is calculated, the threshold value calculation unit 62 calculates the hue degrees r, g,
b, an image quality histogram is generated for each hue, and an optimum binarization threshold T
H _R , TH _G , and TH _B are obtained (steps 2 and 3).

In the next step 4, when the hue to be binarized is specified by the binarization target specifying section 63 in accordance with the procedure shown in FIG. 13, the binary threshold value TH calculated by the threshold value calculation unit 62 according to the control procedure shown in FIG.
Select one of _R , TH _G and TH _B.

Next, in step 6, the binary image generation unit 65
Is used to binarize any one of the hue degrees r, g, and b calculated by the hue degree calculation unit 61 in accordance with the control procedure shown in FIG. 14 with the binarization threshold value TH selected by the threshold value calculation unit 64. Processing is performed to generate a binary image, and in the next step 7, the display unit 15 displays the binary image.

FIG. 12 shows the procedure of the designation operation by the binarized target designation section 63 in steps 1 to 9. In step 1 of the figure, a guide character is displayed on the display unit 15 so as to designate any one of red, green, and blue hues. When a desired color is designated by a key operation in the next step 2, the determination becomes "YES", and in step 3, key input data for the designated color is fetched.

In the next step 4, it is determined whether or not the designated hue is red. If the determination is "YES", a red designation flag _Fr indicating the designation of red is set in a flag setting area such as a RAM. Is set to If it green specifies the determination in Step 6, the blue-designation indicating the determination in step 8 if blue is specified, are "YES", the green indication flag F _g or blue designation indicating a green designation The flag _Fb is set in the flag setting area (steps 7 and 9).

FIG. 13 and FIG. 14 show the control procedure for selecting the threshold value in the
Indicated at 22. In step 1 of the figure, first, the number s of designated colors is calculated from the number of sessions of each of the designated flags F _r , F _g , and F _b for red, green, and blue.
, One, three, or zero, and in accordance with the number, a binarization threshold is selected in the following steps, and the binarization threshold is used for binarization processing. Is determined as the upper limit or the lower limit.

If the number s of designated colors is two, the determination in step 2 is “YES”, and in step 4 the upper limit value flag FF indicating that the binarization threshold is set to the upper limit value is set to “1”. After that, in subsequent steps 5 and 6, it is determined which color is not specified.

If the red color is not specified, the determination of "F _r = 0" in step 5 becomes "YES" and the routine proceeds to step 7, where the binary threshold value for red is set as the binary threshold value TH. TH _R is selected, and the red selection flag F (R) is set to “1” (step 8).

If green is not designated, the determination of "F _g = 0" in step 6 becomes "YES" and the routine proceeds to step 9, where the binary threshold for green is set as the binary threshold TH. select TH _G, and sets the green selection flag F (G) is "1" (step 10).

If the blue color is out of the designated range, step 5,
Any determination of 6 also proceeds to step 11 is "NO", select the binary threshold value TH _B for blue as binary threshold value TH, blue selection flag F a (B) to "1" Set (step 12).

Next, if the number s of designated colors is one, the determination in step 2 is “NO”, and the determination in step 3 is “YE”.
S ", the process proceeds to the control procedure shown in FIG. 14, and in subsequent steps 13 and 14, it is determined which color has been designated.

If red is designated, step 1
The determination of “F _r = 1” of No. 3 becomes “YES” and the routine proceeds to step 15, where the binarization threshold TH _R for red is selected as the binarization threshold TH, and the red selection flag F
(R) is set to "1" (step 16).

If green is designated, step 1
The determination of “F _g = 1” of No. 4 is “YES”, the process proceeds to step 17, the binarization threshold TH _G for green is selected as the binarization threshold TH, and the green selection flag F
(G) is set to "1" (step 18).

If blue is designated, step 1
Steps 1 and 3 are both "NO".
9 and as the binarization threshold TH, 2 for blue
Select valued threshold TH _B, blue selection flag F (B)
Is set to "1" (step 20).

In any case, in step 21, 2
An upper limit flag FF indicating that the binarization threshold is set to an upper limit for binarization processing is set to “0”, and the binarization threshold is set as a lower limit for binarization processing. If the number s of designated colors is 0 or 3, abnormal processing such as displaying that the designation is not appropriate is performed (step 22).

FIGS. 15 to 17 show the binary image generating section 6.
Steps 1 to 22 show a control procedure of the binarization processing by the step 5. In Steps 1 and 2 of FIG. 3, the pixel at the image position where X = 0 and Y = 0 is specified, and in the next Step 2, it is determined whether or not the upper limit value flag FF is set to “1”. If the determination is “YES”, the RAM 2
For example, binary data “0” representing a white pixel is set in the value data storage area UP, and binary data “1” representing a black pixel is set in the other binary data storage area LOW.

If the upper limit value flag FF is "0", the determination in step 2 is "NO", and binary data "1" representing, for example, a black pixel is stored in the storage area UP. For example, binary data “0” representing a white pixel
Are set respectively.

The next step 5 is to set the red selection flag F
(R) is set to “1”, and the following step 6 determines whether or not the green selection flag F (G) is set to “1”. If "YES", the red hue degree r is sampled for the pixel at the coordinate position (X, Y), and it is determined whether or not the hue degree r is larger than the red binarization threshold TH _R (step 7, 8).

If the determination in step 8 is "YES", the storage area UP is referred to and the 2
The value data is read and the pixel is binarized (step 9). If "NO", the storage area LOW is referred to, the binary data set in the area is read, and the pixel is binarized. (Step 10).

If the green selection flag F (G) is set to "1", the determination in step 6 becomes "YES", and the green hue g of the pixel at the coordinate position (X, Y) is set.
Is sampled, and it is determined whether or not the hue degree g is larger than the green binarization threshold value TH _G (step 1).
1, 12).

If the determination in step 12 is "YES", the pixel is binarized into binary data set in the storage area UP (step 13), and if "NO", the pixel is stored in the storage area LOW. (Step 14).

If the blue selection flag F (B) is set to "1", the determinations at steps 5 and 6 are both "N".
O ", the hue degree b of blue is sampled for the pixel at the coordinate position (X, Y), and the hue degree b is 2
Determines binarization threshold TH _B greater than or (step 15, 16).

If the determination in step 16 is "YES", the pixel is binarized into binary data set in the storage area UP (step 17), and if "NO", the storage area LOW (Step 18).

Next, in step 19, the same procedure as described above is executed by designating the pixel at the next X coordinate position (X = 1 in this case), and all the pixels (1 horizontal scan) on the same horizontal scan line are executed. When the processing for the number of pixels on the line is 256) is completed, the determination in step 20 is “YE
S ", and shifts to processing for each pixel on the next horizontal scanning line (step 21).

When the processing for all the horizontal scanning lines (the number of horizontal scanning lines is 256) is repeatedly executed, the determination in step 22 becomes “YES”, and the binarization processing is completed. The image is displayed on the display unit 15.

FIG. 18 shows a schematic configuration of a color image binarization processing apparatus 70 according to the third embodiment of the present invention. The binarization processing device 70 in the illustrated example includes a hue degree calculation unit 71, a hue degree threshold value calculation unit 72, a brightness calculation unit 73, a brightness threshold value calculation unit 74, an extraction target designation unit 75, and a binary image generation unit. 76
The output of the color television camera 14 is output to the hue calculation unit 71, the brightness calculation unit 72, and the binary image generation unit 76, and the output of the binary image generation unit 76 is output to the display unit 15.
Each is connected.

The hue degree calculating section 71 scans the color input image input from the color television camera 14 within the target area specified by the extraction target specifying section 75, and captures pixel data (red, red, Using the green and blue luminance data R, G, and B),
The hue degrees r, g, and b are calculated for each pixel by executing the calculations of the equations (6) to (8). These hue degrees r, g, b
Is given to the hue degree threshold value calculation unit 72. (6) ~
Each hue degree r, g, b calculated by the equation (8) is obtained by the above (1)
There is no substantial difference from the value calculated by the formula (3).

[0083]

(Equation 6)

[0084]

(Equation 7)

[0085]

(Equation 8)

The hue degree threshold value calculator 72 calculates the hue degrees r, g, and h for each pixel input from the hue degree calculator 71.
b, the optimum binarization threshold values TH _R , TH _G ,
This is for calculating TH _B, and employs a method of obtaining an evaluation value of image quality and obtaining an optimal binarization threshold value, as in the first and second embodiments.

The brightness calculating section 73 extracts the color input image input from the color television camera 14 into the extraction target specifying section 7.
5 is scanned within the target area designated by 5, and the pixel data (red, green,
The brightness L is calculated for each pixel by executing the following equation (9) using the blue luminance data R, G, and B). This lightness L is provided to the lightness threshold value calculation unit 74.

[0088]

(Equation 9)

The lightness threshold value calculating section 74 converts each of the hue degrees r, g, b into a binary threshold value TH for each of the hue degrees.
_R TH _G and TH _B are compared with each other to calculate optimal binarization thresholds TH _LR , TH _LG and TH _LB for each hue from the brightness L of each target pixel only for pixels satisfying the conditions described later. In this case, the method of finding the optimum binarization threshold value by finding the evaluation value of the image quality is the same as in the first and second embodiments.

The extraction target specifying section 75 is constituted by a keyboard, a mouse, and the like, specifies a target area in the color input image to be binarized, and sets a hue to be binarized or excluded from the binarization processing, ie, a red hue. , Green, or blue.

The binary image generating section 75 generates a binary image by performing a binarizing process using one of the binary threshold values calculated by the brightness threshold value calculating section 74.

FIG. 19 shows the control procedure of the binarization processing unit 70 for generating a binary image in steps 1 to 10. First, in step 1, when a color input image is taken into the binarization processing device 70 from the color television camera 14, the color input image is stored in an image memory (not shown). In the next step 2, a target area to be binarized is specified by the extraction target specifying unit 75 according to the procedure shown in FIGS. 21 to 23 described later, and a hue to be binarized or not to be binarized is specified. .

Next, in step 3, the hue degree calculation section 71 scans the color input image in the target area to calculate the hue degrees r, g, and b for each pixel, and then calculates the hue degree threshold value calculation section. Reference numeral 72 inputs the respective hue degrees r, g, and b for each pixel from the hue degree calculator 71, generates an image quality histogram for each hue, and further obtains an optimum binarization threshold TH _R for each hue from each image quality histogram. , TH _G and TH _B are obtained (steps 4 and 5).

Next, at step 6, the brightness calculation section 7
3 scans the color input image in the target area and calculates the lightness L for each pixel, and then the lightness threshold value calculation unit 74 converts each of the hue degrees r, g, b into a binary value of the respective hue degree. Value TH
_R , TH _G , and TH _B are compared with each other, and the brightness L of each target pixel is calculated only for the pixels that satisfy a condition described later.
3, an image quality histogram is generated for each hue, and optimum binarization thresholds TH _LR , TH _LG , and TH _LB are determined for each hue from each image quality histogram (step 7,
8).

Next, in step 9, the binary image generation unit 76
The pixel data is binarized by using a binarization threshold corresponding to a designated color from the binarization thresholds TH _LR , TH _LG , and TH _LB calculated by the brightness threshold calculator 74. Then, the display unit 15 displays the binary image in the next step 10.

FIG. 20 shows a threshold value table TB created based on a key input from the extraction target designation section 75 and referred to in the binarization processing. In the figure, T
H _LR, TH _LG, TH _LB is binarization threshold of luminance for each hue, TH _R TH _G, TH _B is binary threshold value of the hue of the each hue. Also, FF _L , FF _r , FF _g , FF
_b is an upper limit flag indicating whether or not each of the binarization thresholds is to be an upper limit for the binarization process. If these flags are "1", it indicates that the upper limit is "0". "Indicates the lower limit. Further, S _r , S _g ,
S _b or is the hue of the binarization processing target, or a color designated identification data indicating which covered, with the value thereof hue if "1" is binarized processed A value of "0" indicates that the hue is out of target.

Thus, the binary data q by the binarization process
(Q = 1 or 0) indicates that the designated hue is red by the following logical expression (10), and if the designated hue is green, by the logical expression of the following expression (11): if there is,
Each is determined by the logical expression shown in Expression (12).

[0098]

(Equation 10)

[0099]

[Equation 11]

[0100]

(Equation 12)

FIGS. 21 to 23 show the details of step 2 in FIG. 19, that is, the steps of the designation operation by the extraction target designation unit 75 in steps 1 to 27. First, at step 1 of the figure, a guide character (message) is displayed on the display unit 15 so as to specify a target area, and then at step 2
A guide character is displayed to input the coordinates of the end points (start point and end point) defining the target area.

FIG. 24 shows a color input image 80, in which the coordinates (x1, y1) of the starting point P and the coordinates (x, y) of the ending point Q are shown.
2, y2) by the key input from the extraction target specifying unit 75, the rectangular target area 81 is specified.

Steps 3 to 7 in FIG. 21 show a procedure for specifying the target area. When this procedure is completed, a confirmation display for specifying the target area is displayed on the display unit 15, and then red and green are displayed. , Blue are designated (steps 8 and 9).

When a desired color is designated by key operation in the next step 10, the determination is "YES", and in steps 11 to 15, red and green are set as hues to be subjected to the binarization processing or not to be subjected to the binarization processing. , Blue are designated.

If red is specified as the hue to be subjected to the binarization processing, the determination in step 11 is "YES".
Becomes "1" as the identification data S _r red designation, also is "0" as the upper limit flag FF _r, are respectively set in the threshold value table TB (Step 16, 1
7).

If green is specified as the hue to be subjected to the binarization processing, the determination in step 12 is made, and if blue is specified, the determination in step 13 becomes "YES". Color designation identification data S _g , S _b
And the upper limit value flags FF _g and FF _b are set (steps 18 and 19 and steps 20 and 2).
1).

If red is specified as a hue to be excluded from the binarization processing, the determination in step 14 is “Y
ES ”and“ 0 ”as the identification data _Sr for red designation.
But also it is "1" as the upper limit flag FF _r, are respectively set in the threshold value table TB (Step 2
2, 23).

If green is specified as a hue to be excluded from the binarization processing, the determination in step 15 is “YE
S ”, and when blue is designated, the determination in step 15 is“ NO ”, and the same color designation identification data S _g ,
S _b and upper limit value flags FF _g and FF _b are set (steps 24 and 25 and steps 26 and 26, respectively).
27).

In the next step 28, a guide character is displayed on the display unit 15 so as to designate whether the image portion to be binarized is the brighter or the darker of the designated hues.

When the brightness is designated by a key operation in the next step 29, the determination is "YES", and in the next step 30, it is determined whether or not the brighter one is designated. if the determination is "YES" as the upper limit flag FF _L "0", if step 3
0 of the determination is "1" as the upper limit value and if "NO" flag FF _L, is set to the threshold value table TB, respectively.

The threshold value table TB is set as described above, and the binarization threshold value T of the hue degree for each hue is set.
H _R, TH _G, TH _B is calculated threshold value table T
After the brightness is set to B, the brightness calculation unit 73 calculates the brightness L for each pixel in the target area. If red is specified as the hue to be subjected to the binarization processing, the brightness calculation is performed. The threshold value calculation unit 74 generates an image quality histogram from the brightness of each target pixel only for pixels whose hue degree r of each pixel is greater than or equal to the binarization threshold TH _R , and converts the image quality histogram into an optimal binary image. determine the threshold value TH _LR.

If the brighter red color is designated, the binary image generator 76 extracts pixels whose lightness L is greater than or equal to the binarization threshold TH _LR. If specified, the brightness L of each pixel is 2
Extracting the binarization threshold TH _LR smaller pixels, so that each produce a binary image. Note that the same applies to the case where green or blue is designated, and a description thereof will be omitted here.

In each of the embodiments described above, the circuit configuration shown in FIG. 4 is used to generate the image quality histogram. However, the present invention is not limited to this, and a circuit configuration using fuzzy inference may be used. .

[0114]

As described above, according to the present invention, the pixel data of each pixel constituting the color input image is input, the intensity of each hue is measured for each pixel, and the intensity of each hue is binarized for each pixel. Since the threshold value is calculated and the binarization process of the color input image is performed, even if the image portion and the color unevenness portion of the portion having the same hue as the target portion are mixed in the color input image, There is a remarkable effect that a hue pattern of only a target portion can be extracted due to a difference in hue intensity.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a circuit configuration of a binarization processing device according to a first embodiment of the present invention.

FIG. 2 is a flowchart showing an operation procedure for setting conditions.

FIG. 3 is a flowchart illustrating a control procedure of the binarization processing device according to the first embodiment.

FIG. 4 is a circuit block diagram illustrating a specific example of a threshold value calculation unit.

FIG. 5 is an explanatory diagram illustrating the principle of a window scanning unit.

FIG. 6 is an explanatory diagram showing a process of converting each pixel data in a window into binary data.

FIG. 7 is an explanatory diagram showing an example of a legal pattern.

FIG. 8 is an explanatory diagram showing an example of an illegal pattern.

FIG. 9 is an explanatory diagram showing an image quality histogram.

FIG. 10 is a block diagram showing a circuit configuration of a binarization processing device according to a second embodiment of the present invention.

FIG. 11 is a flowchart illustrating a control procedure of the binarization processing device according to the second embodiment.

FIG. 12 is a flowchart illustrating an operation procedure performed by a binarization target unit.

FIG. 13 is a flowchart illustrating a control procedure of threshold selection in a threshold selector.

FIG. 14 is a flowchart illustrating a control procedure of threshold selection in a threshold selection unit.

FIG. 15 is a flowchart illustrating a control procedure of a binarization process performed by a binarization generation unit;

FIG. 16 is a flowchart illustrating a control procedure of a binarization process performed by a binarization generation unit;

FIG. 17 is a flowchart illustrating a control procedure of a binarization process performed by a binarization generation unit;

FIG. 18 is a block diagram showing a circuit configuration of a binarization processing device according to a third embodiment of the present invention.

FIG. 19 is a flowchart illustrating a control procedure of the binarization processing device according to the third embodiment.

FIG. 20 is an explanatory diagram showing a threshold value table.

FIG. 21 is a flowchart illustrating a procedure of a designation operation by an extraction target designation unit.

FIG. 22 is a flowchart illustrating a procedure of a designation operation by an extraction target designation unit.

FIG. 23 is a flowchart illustrating a procedure of a designation operation by an extraction target designation unit.

FIG. 24 is an explanatory diagram showing a method of specifying a target area.

FIG. 25 is an explanatory diagram showing a configuration of a solder inspection machine.

[Explanation of symbols]

 10, 60, 70 Binarization processing device 11 Pixel extraction unit 12, 62 Threshold value calculation unit 13, 65, 76 Binary image generation unit 14 Color television camera 16, 61, 71 Hue degree calculation unit 17 Condition setting unit 18 Condition determination unit 19 Game unit 63 Binary target designation unit 64 Threshold value selection unit 72 Hue degree threshold value calculation unit 73 Lightness calculation unit 74 Lightness threshold value calculation unit

Continued on the front page (72) Inventor Masato Ishiba 17 Science Center Building, Chudo-ji Minamimachi, Shimogyo-ku, Kyoto Inside OMRONLA IF Science Research Institute Inc. In-company (56) References JP-A-3-190469 (JP, A) JP-A-1-66781 (JP, A) JP-A-1-51586 (JP, A) (58) Fields investigated (Int. . ^7, DB name) G06T 5/00 200 G06T 1/00 305 G06T 1/00 460 H04N 1/46

Claims (2)

(57) [Claims] 1. A binarization processing device for binarizing a color input image with a binarization threshold value calculated from the color input image to generate a binary image, wherein the binarization threshold value Pixel extraction means for extracting pixels constituting a color input image to be calculated, and binarization threshold value calculation means for calculating a binarization threshold value from pixel data of each pixel extracted by the pixel extraction means. A binary image generating unit that generates a binary image by performing a binarization process on the color input image based on the binary threshold value calculated by the binary threshold calculating unit; Measuring means for inputting pixel data of each pixel constituting the color input image and measuring the intensity of each hue for each pixel; condition setting means for setting a condition regarding the intensity of each hue; Each color measured by means Binarization processing of a color image, comprising: a condition determination unit that extracts only pixels whose intensity satisfies the condition set by the condition setting unit and outputs the pixel data to the binarization threshold calculation unit. apparatus. 2. A binarization processing apparatus for binarizing a color input image with a binarization threshold value calculated from the color input image to generate a binary image, wherein the binarization processing apparatus comprises: First measuring means for inputting pixel data of each pixel to be measured and measuring the intensity of each hue for each pixel; and inputting pixel data of each pixel constituting the color input image and measuring brightness for each pixel A second measuring unit that performs hue intensity measurement for each pixel measured by the first measuring unit.
A first threshold value calculating means for calculating a first binary threshold value from the degree; and a second binary threshold value based on the brightness of each pixel measured by the second measuring means. A second threshold value calculating unit; a specifying unit that specifies a hue to be subjected to the binarization processing; a first binary threshold value and a second binary value related to the hue specified by the specifying unit A binarization processing apparatus for a color image, comprising: a binary image generating means for performing a binarization process using a binarization threshold to generate a binary image.