JP3621937B2 - Agricultural product quality judgment device and agricultural product quality judgment method - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an agricultural product quality determination apparatus and an agricultural product quality determination method, which are suitable for use in an insulator.
[0002]
[Prior art]
Various automatic fruit selection devices that automatically determine the quality of agricultural products and classify them have been developed. In the automatic fruit selection apparatus, first, the agricultural product is imaged, and the quality of the agricultural product is determined from the color, gloss, shape, etc. of the agricultural product based on the acquired image.
In order to accurately determine the quality of the agricultural products such as the eggplant with the stickers, the acquired images of the agricultural products are used for quality analysis after the scraps are removed. For this reason, it is important to accurately detect the material.
[0003]
Conventionally, a portion having a certain ratio with respect to the total length of the agricultural product has been virtually set. For example, if the agricultural product is an insulator, as shown in FIG. 16, a portion within a range of a length L1 (= Lt × R1) obtained by multiplying the total length Lt of the insulator 1 by a predetermined constant R1 is set as an actual sticker 1a. On the other hand, it is assumed to be a virtual set 1a '. In this case, it is necessary to identify the orientation (whether it is on the right or left in FIG. 16) in advance by another method or align it in advance. A detection device for detecting the orientation of the crops is disclosed in, for example, Japanese Patent Application Laid-Open No. 10-76233.
[0004]
[Problems to be solved by the invention]
However, even if it is the same kind of agricultural product, naturally, the ratio of the stagnation to the total length is not constant, so the conventional technology described above so that there is a difference between the imaginary stalk 1a 'and the actual stalk 1a in FIG. Then, it is impossible to accurately detect the material, and as a result, there is a problem that the quality of the agricultural product cannot be accurately determined.
[0005]
The present invention has been devised in view of such a problem, and can accurately remove the scraps from the acquired image of the agricultural product, so that the quality of the crop can be accurately determined based on the image from which the scraps are removed. An object of the present invention is to provide an agricultural product quality determination device and a quality determination method.
[0006]
[Means for Solving the Problems]
For this reason, the agricultural product quality determination device of the present invention (Claim 1) includes an image acquisition unit that acquires an image of an agricultural product with stagnation, and an image of the agricultural product acquired by the image acquisition unit. Binarized differential image creating means for creating a binarized differential image by turning on or off a pixel depending on whether or not a luminance change with respect to an adjacent pixel is larger than a predetermined threshold, and the binarized differential A crop extraction means for extracting a region corresponding to the splash from the binarized differential image created by the image creation means based on an on / off distribution of pixels, and an image of the agricultural product obtained by the image acquisition means From the image of the main body part of the agricultural product from which the area corresponding to the above-mentioned spatula extracted by the above-mentioned spatula extracting means is removed, , The quality A constant, is characterized by being configured to include the quality determining means.
[0007]
In this case, when the binarized differential image creation means is the crop where the crop is expected to have a brightness change larger than the main body portion, the brightness change is less than a predetermined threshold value. Create a binary differential image in which pixels are turned off and pixels whose luminance change is greater than or equal to the threshold value are turned on, and conversely, the crop has a larger luminance change in the main body portion than in the stalk In the case of crops for which a change in brightness is expected, a binary differential image in which pixels whose luminance change is less than a predetermined threshold value is turned on and pixels whose luminance change is equal to or higher than the threshold value is turned off is created, The squeezing means is divided so as to divide the binarized differential image into the side with and without the squeezing, and the squeezing means for enclosing the stalk with pixels in the ON state to clarify the contour. For each segment, an image in a closed area consisting of pixels in the on state A filling means for performing a filling process for turning on all of the pixels, and a filling judgment means for judging the area having the maximum number of pixels among the area where the pixels in the ON state are gathered as the collection (claim) Item 2).
[0008]
Further, it is preferable that the squeezing means is further provided with a contraction / expansion means for performing a contraction process on the binarized differential image subjected to the hole filling process by the hole filling means and then performing an expansion process ( Claim 3).
In particular, it is preferable to apply it to the eggplant as the agricultural product (claim 4).
A method for determining the quality of agricultural products according to the present invention (Claim 5) is a method for determining the quality of agricultural products, in which an image of a crop with crops is acquired and the quality of the agricultural product is determined based on the acquired image of the agricultural product. A first binarized differential image is generated by turning on or off each pixel of the acquired agricultural product image depending on whether or not a luminance change with respect to an adjacent pixel is larger than a predetermined threshold value. A second step of extracting a portion of the stem from the binarized differential image based on an on / off distribution of pixels; and the step extracted from the acquired image of the agricultural product in the second step. And a fourth step of determining the quality from the image of the agricultural product from which the region corresponding to the above-mentioned scraper has been removed. .
[0009]
In this case, in the first step, when the crop is a crop that is expected to have a brightness change larger than that of the main body, the pixels whose brightness change is less than a predetermined threshold value are turned off. And a binary differential image in which pixels whose luminance change is equal to or greater than the threshold value is turned on, and conversely, the crop is expected to have a larger luminance change in the main body portion than in the stalk. In the case of an agricultural product, a binary differential image in which a pixel whose luminance change is less than a predetermined threshold value is turned on and a pixel whose luminance change is equal to or greater than the threshold value is turned off is generated, and the second step Divide the binarized differential image into a side having the gap and a side not having the gap, and Turn on all pixels in the closed area consisting of pixels in the on state That filling steps, among the pixels in the ON state is set region, the region of the maximum number of pixels, may be configured to include a calyx determination step of determining a calyx (claim 6).
[0010]
Further, it is preferable that the second step further includes a step of performing an expansion process after performing a contraction process on the binarized differential image between the hole filling step and the settling step. (Claim 7).
In particular, it is preferable to apply it to the eggplant as the agricultural product (claim 8).
[0011]
DETAILED DESCRIPTION OF THE INVENTION
Embodiments of the present invention will be described below with reference to the drawings.
FIGS. 1-15 is a figure shown about the quality determination apparatus of agricultural products and the quality determination method of agricultural products as one Embodiment of this invention, FIG. 1 is a schematic diagram which shows the structure of a quality determination apparatus, FIG. (C) is a schematic diagram for explaining the quality judgment method, FIG. 3 is a schematic diagram for explaining the quality judgment method, (a) is a histogram relating to a luminance differential value in the magenta, (b) is a main body part. FIG. 4 is a schematic flowchart for explaining the functional configuration of the squeezing extraction unit, and FIGS. 5 to 15 are images for explaining the image processing of the squeezing extraction unit.
[0012]
The agricultural product quality determination apparatus 10 of this embodiment determines the quality of the insulator as an agricultural product, and is configured as shown in FIG. 1. The insulator 1 placed on the tray 2 and conveyed by the conveyor 3 is imaged. The quality of the insulator 1 is determined on the basis of an image (camera image) of the insulator 1 captured by the CCD camera (or TV camera) 11 as an acquisition means and acquired by the CCD camera 11. The quality of the insulator 1 is specifically the color, gloss and shape. The insulators 1 whose quality has been determined are sorted according to their quality on the downstream side (not shown).
[0013]
By arranging the trays 2 on the conveyor so that the long sides thereof coincide with the left-right direction of the CCD camera 11, the posture of the insulator 1 on the conveyor 3 is the right half or the left in the captured image. It is aligned so that the half is located in only one of the halves.
The quality determination apparatus 10 includes the CCD camera 11, the squeeze extraction unit 12, the stalk removal unit 13, and the quality determination unit 14. From the camera image obtained by the CCD camera 11 as shown in FIG. 2 (a), the splash extractor 12 extracts the splash 1a imagewise as shown in FIG. 2B is removed from the camera image of FIG. 2A, and the resulting insulator main body (hereinafter referred to as the main body portion or the actual body) as shown in FIG. The quality of the insulator is determined by the determination unit 14 based on the image of only 1b. Note that the stem extracting unit 12, the stem removing unit 13, and the determining unit 14 are configured in a computer, for example.
[0014]
The insulator 1 has no significant difference in color between the sticker 1a and the main body portion 1b, but there is a difference in the degree of gloss. That is, the main body portion 1b has a gloss on the surface and a relatively uniform luminance, while the heta has a stab (hair) -like projection on the surface, is not very glossy, and has a relatively uneven luminance. For example, when a histogram is created with the horizontal axis representing the luminance change (luminance differential value) and the vertical axis representing the frequency (number of measurement points), the rough 1a is as shown in FIG. 3A, and the main body portion 1b is as shown in FIG. 3B. It becomes like this.
[0015]
Utilizing this difference in luminance change, the sketch extracting unit 12 creates a differential image related to the brightness from the camera image, and extracts an image area having a large brightness change from the differential image as a stick.
As shown in FIG. 1, the shear extracting unit 12 includes a binarized differential image creating unit 12a, a contour clarifying unit 12b, a hole filling processing unit 12c, a contraction / expansion processing unit 12d, and a stitch determining unit 12e. Hereinafter, the wetting extraction unit 12 will be described in detail with reference to the flowchart of FIG.
[0016]
First, at the step A5, the camera image (see FIG. 5) 20 acquired by the CCD camera 11 is input to the material extracting unit 12. From the camera image 20, a binarized differential image creating unit 12a creates a binarized differential image related to luminance.
That is, in step A10, luminance information for each pixel is read from the camera image 20, and a multi-value differential image (see FIG. 6) 21 relating to luminance is created. The differential image is obtained by differentiating the camera image, and the luminance of each pixel of the differential image 21 in FIG. 6 is the magnitude of the luminance change in the pixel at the same coordinate of the camera image 20 in FIG. The difference in luminance (brightness differential value)] is shown, and the coordinates having a larger luminance change in the camera image 20 are displayed in the differential image 21 with higher luminance.
[0017]
This differential image is further binarized at a predetermined threshold C1 in step A20. That is, in each pixel of the differential image 21, if the luminance (= the luminance change amount in each pixel of the camera image 20) is less than the threshold C1, it is set to “0” (off) [the luminance is set to zero. If the luminance is equal to or higher than the threshold value C1, it is set to “1” (ON) [the luminance is set to a predetermined value and displayed in white].
[0018]
As described above, assuming that the histogram of the luminance differential value as shown in FIG. 3A is obtained for the gap and the histogram of the luminance differential value as shown in FIG. 3B is obtained for the main body portion, FIG. If the luminance differential value indicated by the alternate long and short dash line in (a) and (b) is set as the threshold value C1, it is possible to accurately distinguish the hatter from the main body. The threshold value C1 is appropriately set according to the type of the agricultural product, and varies depending on the brightness of the light source that irradiates the agricultural product, the aperture of the lens of the CCD camera, etc., but is about 2 when the crop is an insulator. .
[0019]
Further, image processing is performed in step A30 and step A40 in parallel with step A10 and step A20. In step A30, a mask image 22 (see FIG. 7) is created from the camera image. The mask image is an image obtained by removing the background (the region other than the insulator) from the camera image, and the pixel forming the background is “0” and the pixel forming the insulator is “1”. Further, in step A40, a boundary pixel image 23 (see FIG. 8) is created with the pixel forming the boundary between the background and the insulator set to “1” and the other pixels set to “0” for the mask image 22 (see FIG. 8). That is, the contour of the insulator is taken out as an image).
[0020]
In step A50, the binarized image obtained in step A20 and the mask image 22 obtained in step A30 are combined by AND processing. The AND processing means that if the corresponding pixels (pixels with the same coordinates) of the images to be combined are both “1”, the pixel of this coordinate is set to “1”, otherwise it is set to “0”. This is image processing for creating an image. When the binarized image and the mask image 22 are ANDed, the background is removed from the binarized image, and an image 24 as shown in FIG. 9 is obtained. Thereby, the image processing by the binarized differential image creation unit 12a is completed.
[0021]
In step A60 and step A70, the binarized differential image 24 created by the binarized differential image creating unit 12a is then clarified by the contour clarifying unit 12b to clarify the outline of the entire insulator and the outline of the punch. It has become.
That is, in step A60, the binarized image 24 from which the background has been removed and the boundary pixel image 23 (see FIG. 8) obtained in step A40 are synthesized by OR processing. The OR processing is image processing for creating a composite image by setting a pixel of this coordinate as “1” if at least one of the pixels corresponding to each other in the composite image is “1”. The binarized image 24 in FIG. 9 is a pixel in which the luminance change of the camera image 20 in FIG. 5 is “1”. In the binarized image 24, the contour of the insulator becomes discontinuous as shown in the figure. Although it is not always clear, the boundary pixel image 23 of FIG. 8 is superimposed on the binarized image 24 of FIG. 9 when the binarized image and the boundary pixel image are ORed in this way. As shown in FIG. 10, a binary image 25 in which the contour of the insulator is clarified is obtained.
[0022]
In step A70, the dilation image 25 to which the contour is added is subjected to a dilation process a predetermined number of times. The expansion process is a process of setting a pixel adjacent to this pixel to “1” if a pixel is “1”, and is a process of thickening an image composed of the pixel of “1” by one pixel. For example, if an expansion process is performed on an image as shown in FIG. 11B, an image as shown in FIG. 11C is obtained. In FIGS. 11B and 11C, each cell represents a pixel. As a result, as shown in FIG. 11 (a), an image 26 in which the outline (boundary between the spatula and the insulator body) on the insulator main body side (left side in FIG. 10) is clarified is obtained. It has become.
[0023]
Thus, in step A60, the outline of the entire insulator is clarified, and in step A70, the boundary between the stalk and the insulator body is clarified, that is, the outline of the stalk is clarified. The image processing by the unit 12b is completed. Clarification of the outline of the material is to make the material a closed region surrounded by “1” pixels, which leads to stable filling of the material by a filling process described later.
[0024]
The image 26 subjected to the image processing by the contour clarifying unit 12b is then subjected to image processing by the hole filling processing unit 12c, and the left half 26a is subjected to hole filling processing in step A80 for the image 26. As a result, an image 27 as shown in FIG. 12 is obtained. Further, the right half 27b of the image 27 is subjected to hole filling processing in step A90, and an image 28 as shown in FIG. 13 is obtained.
[0025]
The hole filling process is a process for setting all the pixels in the closed region to “1” when there is a closed region surrounded by “1” pixels in the image. If the entire image is subjected to the hole filling process, the entire insulator is filled, but only the right side is filled by performing the hole filling process for each of the left and right halves of the image. As described above, in FIG. 12 and FIG. 13, the insulators are arranged on the conveyor so that the left and right sides are positioned on the conveyor. The image is divided into the non-exposed side and the hole filling process is performed, so that it is possible to reliably fill only the material.
[0026]
The image 28 subjected to the image processing by the hole filling processing unit 12c is further subjected to image processing by the contraction / expansion processing unit 12d. That is, in step A100, after the contraction process is repeated a predetermined number of times (k times) for the image 28, the expansion process is repeated a predetermined number of times (m times). In contrast to the expansion process, the contraction process is a process for thinning an image composed of “1” pixels by one pixel. For example, if a contraction process is performed on an image as shown in FIG. 14B, an image as shown in FIG. 14C is obtained. Then, if an expansion process is performed on this image, an image as shown in FIG. 14D is obtained. By combining the contraction process and the expansion process in this way, only fine dots 41a are removed from the image 41 as can be seen by comparing the image 41 in FIG. 14B and the image 42 in FIG. 14D. It becomes possible.
[0027]
Therefore, the contraction / expansion processing unit 12d repeats the contraction process for the image 28 shown in FIG. 13 a predetermined number of times, and then repeats the expansion process a predetermined number of times, so that the contour line of the insulator main body as shown in FIG. An image 29 is obtained in which dots scattered in the insulator main body are removed.
Even if the contraction process and the expansion process are performed as described above, an unnecessary image (an aggregate of consecutive “1” pixels) other than the stutter may remain as shown in FIG. In A110, an image having the largest area (consisting of the largest number of pixels) is determined as a set by the set determination unit 12e, and as a result, an image 30 having only the set is obtained as shown in FIG. It becomes.
[0028]
Since the agricultural product quality determination apparatus according to the embodiment of the present invention is configured as described above, it is possible to accurately extract the material from the camera image according to the distribution of the luminance differential value. Therefore, there is an advantage in that it is possible to accurately determine the quality of the insulator based on the insulator image (that is, the image of only the insulator body) from which the insulator has been removed with high accuracy by removing the insulator from the insulator image (camera image).
[0029]
In addition, the quality determination apparatus and quality determination method of the agricultural product of this invention are not limited to embodiment mentioned above, It can change variously in the range which does not deviate from the meaning of this invention.
For example, in the above-described embodiment, the posture of the insulator 1 on the conveyor 3 is aligned in advance so that only the right half or the left half is positioned in the captured image. It is also possible to perform quality judgment without aligning the posture on one conveyor 3 in advance. In other words, (1) the insulator is regarded as an inertia ellipsoid, the inertia main axis (long axis) direction is obtained from the binarized image, and (2) the binarized image is aligned so that the main axis direction coincides with the horizontal direction of the image. The region showing the {circle over (3)} insulator is divided into a left half and a right half with the center of gravity as the boundary, and a hole filling process is performed for each left half and right half as in the above embodiment.
[0030]
In the above embodiment, an example in which the image acquisition means for acquiring an image of a crop is configured by a CCD camera has been described. However, an image captured by an external device is temporarily recorded on a recording medium such as a floppy (registered trademark) disk. Thereafter, an image may be read from the recording medium by a recording / reading device such as a floppy (registered trademark) disk drive. In this case, the recording / reading apparatus constitutes an image acquisition unit.
[0031]
In the above embodiment, an example has been described in which an insulator having a larger luminance change than the main body (actual) is handled as an agricultural product. Therefore, the binarized differential image includes pixels having a luminance change larger than a threshold value. A pixel with a luminance change smaller than the threshold value is set to “0” and “1” is created. However, when dealing with a crop whose actual luminance change is larger than the stagnation, the binary differential image is reversed. A pixel whose luminance change is larger than the threshold is set to “0” and a pixel whose luminance change is smaller than the threshold is set to “1”.
[0032]
Further, in the above-described embodiment, an example has been described in which an insulator in which the actual color and the color of the sputum are similar is handled as the crop. ) Is widely applicable.
[0033]
【The invention's effect】
As described above in detail, according to the present invention, even for a crop with a similar color to the real color, it was obtained based on the on / off distribution of pixels of the binary differential image related to luminance. There is an advantage that the grime can be accurately removed from the image of the agricultural product, and the quality of the crop can be accurately determined based on the image from which the grime has been removed.
[Brief description of the drawings]
FIG. 1 is a schematic diagram showing the configuration of an agricultural product quality determination apparatus according to an embodiment of the present invention.
FIGS. 2A to 2C are schematic diagrams for explaining a method for determining the quality of agricultural products according to an embodiment of the present invention.
FIGS. 3A and 3B are schematic diagrams for explaining a method for judging the quality of agricultural products according to an embodiment of the present invention, in which FIG. 3A is a histogram relating to a luminance differential value in a set, and FIG. 3B is a luminance differential value in a main body portion; It is a histogram regarding.
FIG. 4 is a schematic flowchart for explaining the functional configuration of the crop extraction unit of the agricultural product quality determination apparatus as one embodiment of the present invention (the agricultural product quality determination method as one embodiment of the present invention); .
FIG. 5 is an image for explaining image processing (a method for determining the quality of agricultural products as one embodiment of the present invention) of the stalk extractor according to one embodiment of the present invention.
FIG. 6 is an image for explaining image processing (a method for determining the quality of agricultural products as one embodiment of the present invention) of the stalk extractor according to one embodiment of the present invention.
FIG. 7 is an image for explaining image processing (a method for determining the quality of agricultural products as one embodiment of the present invention) of the patch extractor according to one embodiment of the present invention.
FIG. 8 is an image for explaining image processing (a method for determining the quality of agricultural products as one embodiment of the present invention) of the stalk extractor according to one embodiment of the present invention.
FIG. 9 is an image for explaining image processing (a method for determining the quality of agricultural products as one embodiment of the present invention) of the stalk extractor according to one embodiment of the present invention.
FIG. 10 is an image for explaining image processing (a method for determining the quality of agricultural products as one embodiment of the present invention) of the stalk extractor according to one embodiment of the present invention.
FIG. 11 is a diagram for explaining image processing (a method for judging quality of agricultural products as one embodiment of the present invention) of the stalk extractor according to one embodiment of the present invention, in which (a) is subjected to expansion processing; The binarized differential images (b) and (c) are schematic diagrams for explaining the expansion process.
FIG. 12 is an image for explaining image processing (a method for determining the quality of agricultural products as one embodiment of the present invention) of the stalk extractor according to one embodiment of the present invention.
FIG. 13 is an image for explaining image processing (a method for determining the quality of agricultural products as one embodiment of the present invention) of the stalk extractor according to one embodiment of the present invention.
FIG. 14 is a diagram for explaining image processing (a method for judging quality of agricultural products as one embodiment of the present invention) of the patch extraction unit according to one embodiment of the present invention, in which (a) is subjected to contraction / expansion processing; The binarized differential images (b) to (d) are schematic diagrams for explaining the contraction / expansion processing.
FIG. 15 is an image for explaining image processing (a method for determining the quality of agricultural products according to an embodiment of the present invention) of an image extraction unit according to an embodiment of the present invention.
FIG. 16 is a schematic diagram for explaining a method of removing a sticker from a conventional insulator image.
[Explanation of symbols]
1 Isogo (agricultural crops)
DESCRIPTION OF SYMBOLS 1a Scrap 1b insulator main body 2 tray 3 conveyor 10 quality judgment apparatus 11 CCD camera (image acquisition means)
12 Streaks Extraction Unit 12a Binary Differential Image Creation Unit 12b Outline Clarification Unit 12c Hole Filling Processing Unit 12d Shrinkage Expansion Processing Unit 12e Staging Determination Unit 13 Staging Removal Unit 14 Quality Determination Unit 20 Camera Image 21 Multilevel Differential Image 22 Mask Image 23 Boundary pixel image 24-30 Binary differential image

Claims (8)

Image acquisition means for acquiring images of crops with stickiness,
A binarized differential image with respect to the image of the agricultural product acquired by the image acquisition means, wherein each pixel of the image is turned on or off depending on whether or not the luminance change with respect to the adjacent pixel is larger than a predetermined threshold value. A binary differential image creating means for creating
A sampler extracting unit for extracting a region corresponding to the sampler based on the on / off distribution of pixels from the binary differential image generated by the binary differential image generating unit;
A remover for removing a region corresponding to the above-mentioned splash extracted by the splash extractor from the image of the agricultural product acquired by the image acquirer;
The agricultural product is characterized by comprising quality judging means for judging the quality from the image of the main part of the produce from which the area corresponding to the above-mentioned sticker has been removed by the sticker removing means. Quality judgment device. The binarized differential image creating means
If the crop is a crop that is expected to have a brightness change larger than that of the main portion, the pixel whose brightness change is less than a predetermined threshold value is turned off and the brightness change is the threshold value. Create a binary differential image with pixels above the value turned on,
Conversely, when the crop is a crop that is expected to have a larger luminance change in the main body portion than the stagnation, a pixel whose luminance change is less than a predetermined threshold value is turned on and the luminance change occurs. Create a binary differential image with the pixels above the threshold turned off,
The squeezing means is
A contour clarification means for surrounding the stalk with pixels in an ON state to clarify the contour;
The binarized differential image is divided so as to be divided into the side with the gap and the side without the gap, and for each of the divisions, a filling process is performed to turn on all the pixels in the closed region including the pixels in the on state. The hole filling means,
2. The agricultural product quality determination device according to claim 1, further comprising: a determination unit that determines a region having the maximum number of pixels as a set from among a group of on-state pixels. . The squeezing means is further provided with a contraction / expansion means for performing a contraction process on the binarized differential image subjected to the hole filling process by the hole filling means and then performing a expansion process. The agricultural product quality determination device according to claim 2. The agricultural product quality determination device according to any one of claims 1 to 3, wherein the agricultural product is an insulator. In the method for determining the quality of agricultural products, an image of crops with a sticker is acquired, and the quality of the agricultural products is determined based on the acquired images of the agricultural products.
A binarized differential image is created by turning on or off each pixel of the acquired agricultural product image depending on whether or not the luminance change with respect to an adjacent pixel is larger than a predetermined threshold value. The first step;
A second step of extracting from the binarized differential image a region corresponding to the gap based on pixel on / off distribution;
A third step of removing an area corresponding to the above-described stalk extracted in the second step from the acquired image of the agricultural product;
A method for determining the quality of agricultural products, comprising: a fourth step of determining the quality from the image of the agricultural products from which the area corresponding to the above-mentioned scrap has been removed. In the first step,
If the crop is a crop that is expected to have a brightness change larger than that of the main portion, the pixel whose brightness change is less than a predetermined threshold value is turned off and the brightness change is the threshold value. Create a binary differential image with pixels above the value turned on,
Conversely, when the crop is a crop that is expected to have a larger luminance change in the main body portion than the stagnation, a pixel whose luminance change is less than a predetermined threshold value is turned on and the luminance change occurs. Create a binary differential image with the pixels above the threshold turned off,
The second step is
Surrounding the stalk with on-state pixels to clarify the contour;
The binarized differential image is divided so as to be divided into the side with the gap and the side without the gap, and for each of the divisions, a filling step for turning on all the pixels in the closed region composed of the pixels in the on state,
6. The method for judging quality of agricultural products according to claim 5, further comprising: a step of determining a region having the maximum number of pixels among regions where pixels in an on state are gathered as a set. The second step is
7. The method according to claim 6, further comprising a step of performing an expansion process after the shrinkage process is performed on the binarized differential image between the hole filling step and the setter determination step. A method for judging the quality of agricultural products. The method for judging quality of agricultural products according to any one of claims 5 to 7, wherein the agricultural products are eggplants.

Images