JPH0989539A - Testing device for fruit and vegetable - Google Patents

Abstract

PROBLEM TO BE SOLVED: To accurate test even when a subject for testing has portions where illumination undergoes total reflections by calculating an area using the coordinates of the contour of the subject for testing. SOLUTION: A subject T for testing is scanned by a scan means 13 in two directions from the right and left of the direction of X-axis in the order of scanning lines, and its coordinates are stored in a coordinate storage means 18 as the contour points of the subject T for testing. After the scans are complete, the area within the contour Ta of the subject T for testing is calculated by a size data calculation means 19 from the coordinates stored in the coordinate storage means 18, and this area value is used as data regarding size. The photographed part of the subject T for testing is thus recognized. In this case, since the area is calculated from the coordinates of the contour Ta, even if the subject T for testing has portions called highlights Tb produced by light, where illumination undergoes total reflections, these highlights Tb are also involved in calculation of the area for the correctness of the area value.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus for testing fruits and vegetables for carrying out size inspection and color inspection of fruits and vegetables such as apples imaged by a color camera.

[0002]

2. Description of the Related Art Conventionally, as a test device of this type, for example, an object to be tested for fruits and vegetables is imaged in color, and red, green, and blue component intensity data of the imaged image are stored for each pixel. There is one that calculates data relating to the size and color of the test object based on the storage result. In this conventional verification device, when obtaining the size, all the captured images are scanned, and the object is binary image from any one of the red component image, the green component image, and the blue component image or the brightness image. The object is extracted by using a means such as conversion, and the number of pixels is counted.

When calculating color data,
The entire image is scanned, and the brightness obtained for each pixel by the sum of the red, green, and blue color components, and the chromaticity obtained by the ratio of each component to the brightness of that pixel Whether or not the object is the object is discriminated by the degree, and the average value of the pixels forming the object is calculated.

[0004]

By the way, in the above-described conventional inspection apparatus, when the size is obtained, extraction is performed for the portion where the illumination called the highlight is totally reflected on the inspection object. Therefore, there is a problem that an accurate test cannot be performed because the measurement is made smaller than the actual area. In addition, size and color data,
Since each is obtained by scanning the entire image,
All the scans have to be performed twice, which takes time, and there is a problem that quick determination cannot be performed.

The present invention has been made in view of the above problems, and makes it possible to accurately determine the size, and to perform data related to the size and color without performing the entire scanning twice. It is an object of the present invention to provide an assay device capable of obtaining the above-mentioned value, shortening the assay time, and improving the efficiency.

[0006]

In order to achieve such an object, an apparatus for assaying fruits and vegetables according to the present invention includes an imager for color-imaging an object to be inspected for fruits and vegetables, a red image picked up by the imager, Image data storage means for storing each component intensity data of green and blue for each pixel, and data relating to the size and color of the imaged part of the test object based on the image data stored in the image data storage means In the verification apparatus for fruits and vegetables for calculating, the scanning means scans the image for each scanning line from the outside to the inside of the contour line of the test object, and for each pixel stored in the image data storage means by the scanning of the scanning means. Of the red, green, and blue component intensities, the image data reading means for reading out the required intensity in the above scanning order, and the test value for each pixel relating to the predetermined relationship of the intensities read by the image data reading means In order of reading A test value calculating means for calculating, a comparing means for comparing the test value calculated by the test value calculating means with a predetermined reference value in the order of the above calculation, and a test value by the comparison of the comparing means with respect to the reference value. Scan stop transition means for stopping the scan of the scanning means for the scan line and shifting to the scan of the next scan line when there is a predetermined relationship, and the coordinates of the stop position of the scan line stopped by the scan stop transition means. And a coordinate storage means for reading and storing the contour points of the test object, and an area within the contour line of the test object calculated from the coordinates stored in the coordinate storage means, and the area value is used as data relating to the size. And a color data calculating means for rescanning only the inside of the contour line of the test object from the coordinates stored in the coordinate storage means to calculate data relating to the color of the test object. Are

Rescanning means for rescanning only the inside of the contour line of the test object from the coordinates stored in the coordinate storage means, if necessary,
Color image data read-out means for reading out the necessary intensities of the red, green, and blue component intensities for each pixel stored in the image data storage means by the rescanning means, and the color image data. A color test value calculating means for calculating the color test value for each pixel relating to a predetermined relationship of the intensity read by the reading means in the order of reading, a color test value calculated by the color test value calculating means, and a predetermined color reference. The value is compared with the color value comparison means in the order of the above calculation, and the number of pixels when the color test value has a predetermined relationship with the color reference value by the comparison of the color value comparison means And a pixel number accumulating means for calculating.

[0008]

BEST MODE FOR CARRYING OUT THE INVENTION A vegetable and fruit testing device according to an embodiment of the present invention will be described below with reference to the accompanying drawings. This test device is a device for continuously testing apples as fruits and vegetables. In the fruit and vegetable inspection device according to the embodiment of the present invention shown in FIGS. 1 and 2, 1 is a belt conveyor that conveys an inspection object T of apples as fruits and vegetables, which are arranged at equal intervals along the traveling direction. There is provided a plurality of holders 2 for holding an apple, which is an object T to be inspected, which has the axis of its core vertical and the connection side to the tree is up.

Reference numeral 3 is an image pickup device such as a CCD camera which is provided on the way of conveyance of the belt conveyer 1 and color-images the test object T held by the holder 2 from directly above. Reference numeral 4 is a light that illuminates the test object T from diagonally above. Imager 3 above
The light 4 and the light 4 are provided in the light shielding box 5. The light shielding box 5 covers the test object T to be tested and blocks light rays other than the light 4.
At the entrance and exit of the light shielding box 5, for example, a comb-shaped portion (not shown) through which the test object T can pass is provided. A timing sensor 6 is for causing the image pickup device 3 to take an image when the test object T is located at the image pickup position of the image pickup device 3.

Reference numeral 10 denotes a control unit having functions such as a CPU,
Reference numeral 11 denotes a display unit such as a CRT that displays various data such as images. In the control unit 10, as shown in FIG.
Is an image data storage means for storing, for each pixel, component intensity data of red component intensity (R), green component intensity (G), and blue component intensity (B) of the image captured by the image pickup device 3. The image is a discrete image of 512 pixels in the vertical direction and 512 pixels in the horizontal direction, for example. Reference numeral 13 denotes a scanning unit that scans the above-mentioned image for each scanning line from the outside to the inside of the contour line Ta of the test object T, and, for example, scans in two or four directions from the boundary of the image. In the case of two directions, there are two directions from the left side and the right side in the X-axis direction in FIG. 1, and in the case of four directions, the above-mentioned two directions in the X direction and the upper side and the lower side in the Y-axis direction. There are 4 directions, 2 directions from. In this example, two directions from the left and right in the X-axis direction (forward scan and reverse scan)
The case of scanning in will be described.

Reference numeral 14 is an image data reading means for reading out necessary intensities of red, green, and blue component intensities for each pixel stored in the image data storage means 12 by the scanning of the scanning means 13, and 15 is an image data reading means. It is a verification value calculation means for calculating a verification value M for each pixel relating to a predetermined intensity relationship read out by the image data reading means 14 in the above-mentioned reading order. The test value M (M _{1 to} M _{i to} M _n , where n is the total number of pixels) for each pixel is, for example, the ratio of two components of each of red, green, and blue components. For example, the intensity R (R ₁ ~
The ratio (R / G) of R _{i to} R _n ) and the intensity G (G _{1 to} G _{i to} G _n ) of the green component is used (M _i = R _i / G _i ). Therefore,
The image data reading means 14 controls the intensity R (R _{1 to} R) of the red component.
_{i to} R _n ) and the green component intensity G (G _{1 to} G _{i to} G _n ) are read out.

Reference numeral 16 is a comparison means for comparing the test value M _i calculated by the test value calculation means 15 with a predetermined reference value M ₀ in the above calculation order. The reference value M ₀ is, for example, the ratio (R ₀ / G) of the reference intensity R ₀ of the red component and the reference intensity G ₀ of the green component.
₀ ) is used. As an example, for example, M ₀ = R ₀ /
G ₀ = 2.7. Reference numeral 17 denotes a scanning stop for stopping the scanning of the scanning unit 13 for the scanning line and shifting to the scanning of the next scanning line when the verification value M has a predetermined relationship with the reference value M ₀ by the comparison of the comparison unit 16. It is a means of transition. The predetermined relationship is defined as follows. Test value M _i
When <M ₀ , it is determined that the test object T does not exist, and scanning is advanced to the next pixel. When the test value M _i ≧ M ₀ , it is determined that the test object T is present, and the scan line is stopped and the scan of the next scan line is started.

Reference numeral 18 is a coordinate storage means for reading the coordinates of the stop position of the scanning line stopped by the scanning stop transition means 17 and storing the coordinates as the contour points of the test object T. That is, the coordinates f _L (X,
Y) and the coordinates f _R (X, Y) of the contour point when scanned from the right side in the X-axis direction are stored. Reference numeral 19 is a coordinate storage means 18.
Is a size data calculating means for calculating the area within the contour line Ta of the test object T from the stored coordinates and using the area value as the data related to the size. The area S is calculated by the following formula. S = ΣΔ | f _L (X) −f _R (X) | That is, the area S is the number of pixels in the contour line Ta.

Reference numeral 20 is a color data calculation means for rescanning only the inside of the contour line Ta of the test object T from the coordinates stored in the coordinate storage means 18 to calculate data relating to the color of the test object T. In this color data calculation means 20,
Reference numeral 21 is a rescanning means for rescanning only the inside of the contour line Ta of the test object T from the coordinates stored in the coordinate storage means 18. This rescan is performed only in the forward direction. Reference numeral 22 denotes the image data storage means 1 which is scanned by the rescanning means 21.
2 is a color image data reading means for reading out the necessary intensities of the red, green, and blue component intensities stored for each pixel in the scanning order.

Reference numeral 23 is a color test value calculation means for calculating the color test value C for each pixel relating to a predetermined intensity relationship read by the color image data read means 22 in the above-mentioned reading order. The color test value C (C _{1 to} C _{i to} C _n , where n is the total number of pixels) is, for example, the ratio of two components of red, green, and blue components. For example, similar to the above, the intensity R (R _{1 to} R _{i to} R _n ) of the red component and the intensity G (G _{1 to} G) of the green component of each pixel.
_Using the ratio (R / G) of _{i to} G _n (C _i = R _i / G
_i ). Therefore, the color image data reading means 22 causes the intensity R (R _{1 to} R _{i to} R _n ) of the red component and the intensity G (G ₁ to R _n ) of the green component.
G _{i to} G _n ) are read out.

Reference numeral 24 is a color value comparison means for comparing the color test value C _i calculated by the color test value calculation means 23 with a predetermined color reference value C ₀ in the above calculation order. As the color reference value C ₀ , for example, the ratio (R ₀ / G ₀ ) of the reference intensity R ₀ of the red component and the reference intensity G ₀ of the green component is used, as in the above. As an example, for example, as in the above, C ₀ = R ₀ / G ₀ = 2.
7 is set.

Reference numeral 25 is a pixel number accumulating means for accumulating the number of pixels when the color test value C has a predetermined relationship with the color reference value by the comparison of the color value comparing means 24 and calculating the accumulated value N. . The predetermined relationship is defined as follows. When the color test value C _i <C ₀ , the color is bad and the colors are not accumulated. When the color test value C _i ≧ C ₀ , it is determined that the color is good, and the cumulative total value N is calculated. Reference numeral 26 is a color discrimination data calculation means for making data relating to color, and calculates the color discrimination value H (H = N / S) by taking the ratio of the area S and the cumulative value N. It can be determined that the larger H is, the better the color is.

Next, the operation of the apparatus for testing fruits and vegetables according to the embodiment of the present invention will be described with reference to the flow charts shown in FIGS. 3 and 4. As shown in FIG. 2, an inspection object T of apples as fruits and vegetables is held by a holder 2 of a belt conveyor 1 and sequentially conveyed. Then, when it comes directly below the image pickup device 3 of the light shielding box 5, the image pickup device 3 takes an image under the illumination of the light 4 (3-1). And
The red, green, and blue component intensity data of this captured image are stored in the image data storage means 12 for each pixel (3-2).

In this state, the scanning means 13 performs two directions from the left side and the right side in the X-axis direction (forward scanning and reverse scanning).
Scanning is performed in the order of scanning lines (3-3 to 3-1N).
In this process, one pixel is scanned to read out the necessary intensities (red component intensity R and green component intensity G) among the red, green, and blue component intensities stored in the image data storage means 12 for each pixel. (3-4), and the test value calculation means 15 calculates the test value M (M _i = R _i / G _i ) (3
5) The comparison means 16 compares the test value M _i calculated by the test value calculation means 15 with the predetermined reference value M ₀ (3-6).

Then, in the scan stop transition means 17,
When the test value M _i is M _i <M ₀ (3-7N), it is determined that the test object T does not exist, and scanning is advanced to the next pixel (3
-3 to 3-7N). If the test value M _i ≧ M ₀ (3-7
Y), assuming that the test object T is present, the scanning line is stopped and the scanning of the next scanning line is started (3-).
8). In this case, since the process moves to the next scanning line, it is not necessary to perform all the scanning, and the speed can be increased accordingly. And
The coordinate storage means 18 stores the coordinates f _L (X, Y) when scanned from the left side in the X-axis direction and the coordinates f _R (X, Y) when scanned from the right side in the X-axis direction (3 −
9). In this manner, scanning in two directions (forward scanning and reverse scanning) from the left side and the right side in the X-axis direction is performed in the scanning line order, and the coordinates corresponding to each scanning line are stored (3-
9).

Next, when all the scans are completed (3-10
Y), the size data calculation unit 19 calculates the area within the contour line Ta of the test object T from the coordinates stored in the coordinate storage unit 18, and the area value is used as the size-related data (3-11). . Thereby, the size of the imaged part of the test object T can be recognized. In this case, since the area is calculated by the coordinates of the contour line Ta, the light 4
Even if there is a portion of the test object T that is called the highlight Tb where the illumination is totally reflected, this highlight Tb portion is also included in the area calculation. Therefore, as compared with the conventional case where the highlight Tb portion is not extracted, As a result, the value becomes approximately the same as the actual area, and the area value becomes accurate, so that an accurate test can be performed.

Next, the color data calculating means 20 calculates data relating to the color of the test object T. In the color data calculation means 20, as shown in FIG.
The rescanning means 21 rescans only the inside of the contour line Ta of the test object T from the coordinates stored in the coordinate storage means 18 (4-1). This rescanning is performed only in the forward direction (4-1 to 4-7N). In this case, since scanning is performed only inside the contour line Ta, it is not necessary to perform scanning outside the contour line Ta. Therefore, it is not necessary to perform all scanning, so that the speed can be increased accordingly. .

In the course of this scanning, the color image data reading means 22 scans one pixel to scan the one pixel, and the necessary intensity (red, red, green, blue) of each component intensity stored in the image data storage means 12 for each pixel. The intensity R of the component and the intensity G of the green component are read (4-2), and the test value M (C _i = R _i / G _i ) is calculated by the color test value calculating means 23 (4-).
3), the color value comparison means 24 causes the color test value calculation means 23
The test value C _i calculated by is compared with a predetermined reference value C ₀ (4-4). Then, in the pixel number accumulating means 25, when the color test value C _i <C ₀ , it is determined that the color is bad and the pixels are not accumulated. When the color test value C _i ≧ C ₀ , it is determined that the color is good, the pixels are accumulated, and the accumulated value N is calculated (4-
6).

After that, when all the scans are completed (4-7
Y), the ratio of the area S and the cumulative value N is calculated by the color discrimination data calculation means 26, and the color discrimination value H (H = N) is calculated.
/ S) is calculated (4-8). As a result, the degree of color of the imaged portion of the test object T can be recognized. That is, it can be determined that the greater the color discrimination value H, the better the color fit.

In the above embodiment, the scanning means 1
Although the scanning of 3 is performed in two directions from the boundary of the image, it may be performed in four directions of scanning from two directions from the upper side and the lower side in the Y-axis direction, and may be appropriately changed. Further, the test value M, the color test value C, the reference value M _0, and the color reference value C ₀ are
Not limited to the setting of the ratio of the red and green components, it may be set in any manner and may be changed appropriately. Further, the reference value may be changed as appropriate. Further, in the above embodiment, an example using apples as fruits and vegetables has been described, but the present invention is not limited to this, and the present invention is applied to any kind of fruits and vegetables such as other fruits and vegetables. Of course you can.

[0026]

As described above, according to the inspection apparatus for fruits and vegetables of the present invention, in calculating the data relating to the size, the area is calculated by the coordinates of the contour line of the inspection object, and therefore the inspection object. Even if there is a part where the illumination is totally reflected in the area, it can be included in the area calculation including this highlight part, and the actual area compared to the case where the highlight part is not extracted as in the past And the area value becomes accurate, so that an accurate test can be performed. Further, in the calculation of the color-related data, re-scanning is performed only on the inside of the contour line of the test object, so that it is not necessary to perform scanning on the outside of the contour line. Since it is good, there is an effect that the speed can be increased and the time required for the verification can be significantly shortened. Further, both the size and the color data are calculated by using the same factor of the red, green, and blue component intensities of each pixel stored in the image data storage means. It is good, and also in this respect, it is possible to increase the speed.

[Brief description of drawings]

FIG. 1 is a block diagram showing a main part of an apparatus for testing fruits and vegetables according to an embodiment of the present invention.

FIG. 2 is an overall view showing a vegetable and fruit testing device according to an embodiment of the present invention.

FIG. 3 is a flow chart diagram showing operation steps of the fruit and vegetable inspection apparatus according to the embodiment of the present invention.

FIG. 4 is a flow chart diagram showing operation steps of the fruit and vegetable inspection apparatus according to the embodiment of the present invention.

[Explanation of symbols]

T Test object Ta Contour line Tb Highlight 1 Conveyor 2 Holder 3 Imager 4 Light 5 Light-blocking box 6 Timing sensor 10 Control unit 11 Display unit 12 Image data storage unit 13 Scanning unit 14 Image data reading unit 15 Test value calculating unit 16 Comparison means 17 Scan stop transition means 18 Coordinate storage means 19 Size data calculation means 20 Color data calculation means 21 Rescanning means 22 Color image data reading means 23 Color test value calculation means 24 Color value comparison means 25 Pixel accumulation means 26 colors Discrimination data calculation means M (M _{0 to} M _{i to} M _n ) Test value R (R _{1 to} R _{i to} R _n ) Red component intensity G (G _{1 to} G _{i to} G _n ) Green component intensity B Blue component intensity M ₀ reference value S area C (C _{1 to} C _{i to} C _n ) color test value C ₀ color reference value H color discrimination value

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification number Internal reference number FI Technical display location G06T 7/60 G06F 15/70 355 (72) Inventor Norimitsu Hanamatsu Aomori City, Aomori 8 Role Ashiya 202-4, Aomori Prefecture Industrial Technology Development Center (72) Inventor Fujitoshi Shino Aomori-shi, Aomori-city Yatsutsu Role Ashiya 202-4, Aomori Industrial Technology Development Center (72) Inventor Tomoyuki Kudo Aomori Prefecture Towa Denki Kogyo Co., Ltd., 1-5 Oita-Metalcho, Hirosaki-shi (72) Inventor Shunichi Mikami 5-5 Ohta-Metal-cho, Hirosaki City, Aomori Prefecture Towa Denki Co., Ltd

Claims (2)

[Claims] 1. An image pickup device for color-picking up an image of an object to be tested for fruits and vegetables, and image data storage means for storing, for each pixel, red, green, and blue component intensity data of an image picked up by the image pickup device. In a fruit and vegetable test apparatus for calculating data relating to the size and color of a part of a test object imaged based on the image data stored in the image data storage means, the image is obtained from outside the contour line of the test object. Scanning means for scanning each scanning line toward, and necessary intensity among the red, green, and blue component intensities of each pixel stored in the image data storage means by the scanning of the scanning means is read in the above scanning order. An image data read-out means, a test value calculation means for calculating a test value for each pixel relating to a predetermined intensity relationship read by the image data read-out order in the reading order, and a test value calculated by the test value calculation means. Predetermined Comparing means for comparing the quasi-value with the above-mentioned calculation order, and when the comparison value of the comparing means has a predetermined relationship with the reference value, the scanning of the scanning means is stopped for the scanning line and the next scanning is performed. Scanning stop transition means for shifting to line scanning, coordinate storage means for reading the coordinates of the stop position of the scanning line stopped by the scanning stop transition means, and storing the coordinates as contour points of the test object, and the coordinate storage means. A size data calculating means for calculating the area in the contour line of the test object from the stored coordinates and using the area value as data relating to the size, and the contour line of the test object from the coordinates stored in the coordinate storage means. And a color data calculating unit that rescans only the inside to calculate data related to the color of the test object. 2. Rescanning means for rescanning the color data calculating means only from the coordinates stored in the coordinate storage means only inside the contour line of the test object, and the image data storage by scanning of the rescanning means. There is a predetermined relationship between the color image data read-out means for reading out the necessary intensities of the red, green, and blue component intensities for each pixel stored in the means in the above-described scanning order and the intensity read out by the color image data read-out means. A color test value calculation means for calculating the color test value for each pixel in the above-mentioned reading order, and a color value for comparing the color test value calculated by the color test value calculation means with a predetermined color reference value in the above calculation order. Comparing means, and a pixel number accumulating means for accumulating the number of pixels when the color test value has a predetermined relationship with the color reference value by the comparison of the color value comparing means to calculate the accumulated value. It is constituted, It is characterized by the above-mentioned. Test apparatus of the placing of fruits and vegetables.