Nut quality detection and analysis method based on visual technology

Abstract

The invention relates to the technical field of nut quality detection, and particularly discloses a nut quality detection analysis method based on a visual technology, and the method comprises the steps of nut external information collection, nut external quality evaluation, detection group nut weight screening, nut internal quality analysis and nut comprehensive quality evaluation. According to the method, whether the comprehensive quality of the nuts is qualified or not is analyzed from the two aspects of the external quality and the internal quality of the nuts, the appearance, the internal quality and the overall quality of the nuts can be comprehensively evaluated, the quality of the nuts is accurately classified and judged through multi-dimensional feature extraction and analysis, automation of nut quality detection is achieved, and the detection efficiency is improved. Compared with the prior art, the nut detection device has the advantages that the detection efficiency is improved, the labor cost and subjective errors of manual detection are reduced, meanwhile, nuts with unqualified quality are sorted through the mechanical arm, nuts with unqualified weight are screened, nuts with abnormal quality are screened in time, and an effective data basis is provided for subsequent comprehensive quality analysis of nuts.

Description

Technical Field

[0001] The present invention relates to the technical field of nut quality detection, and in particular to a nut quality detection and analysis method based on vision technology. Background Art

[0002] Currently, the quality of nuts during production, processing, and sales relies primarily on manual sampling inspections. This method suffers from low efficiency, high subjectivity, fatigue, and difficulty ensuring consistent quality. Furthermore, existing automated inspection technologies suffer from insufficient accuracy and inability to comprehensively assess quality for nuts, a product with diverse surface colors and internal structures requiring comprehensive evaluation. Therefore, an efficient, accurate, and comprehensive vision-based nut quality inspection and analysis method is needed.

[0003] The following problems still exist in the existing technology: 1. The existing technology only evaluates the quality based on the external signs of nuts, and does not combine the external and internal characteristics of nuts to evaluate the quality as a whole, which reduces the accuracy and rationality of the comprehensive quality assessment of nuts and cannot effectively protect the legitimate rights and interests of consumers.

[0004] 2. The existing technology simply judges the maturity based on the overall color of the nut shell, which is unable to accurately assess the maturity of the nuts, reducing the objectivity and consistency of nut maturity detection. Summary of the Invention

[0005] In view of this, in order to solve the problems raised in the above background technology, a nut quality detection and analysis method based on visual technology is proposed.

[0006] The purpose of the present invention can be achieved through the following technical solutions: The present invention provides a nut quality detection and analysis method based on visual technology, comprising the following steps: S1, collection of external information of nuts: dividing the target batch of nuts into several detection groups according to a preset ratio, transporting the nuts of each detection group through a detection conveyor belt, and using visual acquisition equipment to obtain the circumference, area and surface image data of all nuts in each detection group.

[0007] S2. Nut external quality assessment: Based on the surface image data, the preset appearance parameter threshold is used to determine whether the appearance quality of the nuts is qualified. If there are nuts with unqualified appearance, the sorting device is triggered to remove the unqualified nuts.

[0008] S3. Weight screening of nuts in the test group: Count the number of remaining nuts in each test group, obtain the total mass of each test group through a weighing device, and compare it with the set standard nut weight range. If the total mass does not meet the standard, the nuts in the test group will be processed.

[0009] S4. Analysis of the intrinsic quality of nuts: The internal structure of the target nuts after screening is imaged, and the internal defect parameters of each target nut are calculated using image analysis to generate the intrinsic quality qualification coefficient and count the number of internally unqualified nuts.

[0010] S5. Comprehensive quality assessment of nuts: Evaluate whether the comprehensive quality of nuts in the target batch is qualified based on the number of unqualified nuts. If unqualified, feedback will be provided immediately.

[0011] Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects: (1) The present invention can comprehensively evaluate the appearance, internal and overall quality of nuts by analyzing whether the comprehensive quality of nuts is qualified from two aspects: external quality and internal quality of nuts. Through multi-dimensional feature extraction and analysis, the quality of nuts can be accurately classified and judged, thereby realizing the automation of nut quality detection, improving detection efficiency, reducing labor costs and subjective errors in manual detection, and performing quality analysis from two levels at the same time, which will make the implementation of quality standards clear and effectively protect the legitimate rights and interests of consumers.

[0012] (2) The present invention uses a robotic arm to sort out nuts that do not meet external quality standards and screen nuts that do not meet weight standards, and promptly screens nuts with abnormal quality, providing effective data basis for comprehensive quality analysis of subsequent target batches of nuts. At the same time, automated screening can improve the quality and production efficiency of nut products.

[0013] (3) The present invention calculates the maturity of each nut in each test group in the target batch by combining the RGB values ​​corresponding to each pixel in the surface image of the nut. By analyzing the RGB value of each pixel, rich color information can be obtained instead of simply judging the maturity based on the overall color. This allows for a more accurate assessment of the maturity of the nut, adapts to complex surface color changes, and improves the objectivity and consistency of nut maturity detection.

[0014] (4) The present invention weighs the nuts before the intrinsic quality inspection to preliminarily screen out nuts that do not meet the weight standards, thereby reducing the amount of nuts entering the intrinsic quality inspection link, improving the efficiency of the nut intrinsic quality inspection, and reducing the inspection cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0016] Figure 1Schematic diagram of the process steps of the present invention.

[0017] Figure 2 This is a flow chart of the external quality assessment of nuts of the present invention.

[0018] Figure 3 This is a flowchart of nut weight assessment according to the present invention. DETAILED DESCRIPTION

[0019] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0020] See also Figure 1 As shown, the present invention provides a nut quality detection and analysis method based on visual technology, including: S1, nut external information collection: dividing the target batch of nuts into several detection groups according to a preset ratio, transporting the nuts in each detection group through a detection conveyor belt, and using a visual acquisition device to obtain the circumference, area and surface image data of all nuts in each detection group.

[0021] It should be noted that the circumference, area and surface image of each nut in each inspection group in the target batch are collected as follows: 1) Image acquisition process: When the nut moves with the conveyor belt to the imaging area below the camera, the camera is triggered to collect the image. The triggering mechanism can be realized by a photoelectric sensor. When the nut blocks the light of the sensor, the sensor sends a signal to the camera, and the camera immediately takes a picture. This ensures that each nut can be accurately photographed, and the captured image can accurately reflect the actual shape of the nut, thereby obtaining the surface image of the nut; 2) Image processing and circumference and area calculation: First, the collected image is grayscaled to convert the color image into a grayscale image, and then filtered to remove noise in the image. The edge detection algorithm is used to extract the outline of the nut. After obtaining the outline of the nut, the mathematical method in the image processing software can be used to calculate its circumference and area.

[0022] S2. Nut external quality assessment: Based on the surface image data, the preset appearance parameter threshold is used to determine whether the appearance quality of the nuts is qualified. If there are nuts with unqualified appearance, the sorting device is triggered to remove the unqualified nuts.

[0023] The specific process of judging whether the appearance quality of the nuts is qualified is: performing a roundness analysis on the perimeter and area of ​​each nut in each test group, and determining the roundness compliance coefficient of each nut based on the roundness.

[0024] In a specific embodiment of the present invention, the specific process of calculating the roundness compliance coefficient of each nut is as follows: the circumference and area of ​​each nut in each detection group are calculated using the roundness formula to obtain the roundness of each nut, the standard roundness of the target nut is extracted from the database, and the roundness of each nut is analyzed for deviation from the corresponding standard roundness to obtain the roundness compliance coefficient of each nut.

[0025] The circularity formula is Here, π represents the ratio of the circumference of a circle to the area of ​​a circle, which is approximately 3.14. C and S represent the circumference and area, respectively.

[0026] The roundness of each nut meets the coefficient analysis formula: Among them, Δε represents the roundness deviation of the nut set as the reference, ε 标 is the standard roundness of the target nut, ε ij is the roundness of each nut in each test group. ij -ε 标 | Part: Calculate the absolute deviation between the actual circularity and the standard value and deduct it from the maximum allowable deviation to reflect the degree of shape compliance. Normalize it and limit the result range to [0,1].

[0027] The RGB channel color mean was located from the image of each nut in each test group, and compared with the RGB channel color range value when the nut was mature to obtain the maturity of each nut.

[0028] The specific process of determining the maturity of each nut is as follows: locating the R value, G value, and B value corresponding to each pixel point from the surface image of each nut in each detection group, and obtaining the color mean of the R, G, and B channels respectively through mean calculation.

[0029] The color means of the R, G, and B channels are respectively compared with the color range values ​​of the R, G, and B channels of the target nuts when they are mature and stored in the database. If the color means of the R, G, and B channels of a nut are all within the color range values ​​of the R, G, and B channels of the target nuts when they are mature, the maturity of the nut is recorded as τ1.

[0030] If the color mean values ​​of the R, G, and B channels of a nut are not within the color range values ​​of the R, G, and B channels of the target nut when it is mature, the maturity of the nut is recorded as τ2.

[0031] If any of the color means of the R, G, and B channels of a nut is not within the color range of the R, G, and B channels of the target nut when it is mature, the maturity of the nut is recorded as τ3.

[0032] In summary, the maturity of each nut in each test group was statistically analyzed.

[0033] In a specific embodiment of the present invention, the set value of τ1 is 1, the set value of τ3 is 0.5, and the set value of τ2 is 0.

[0034] It's also important to note that in the RGB color space, each pixel is represented by the values ​​of three channels: red (R), green (G), and blue (B). For mature nuts, the values ​​of the R, G, and B channels will fall within a certain range. For example, in the RGB color space, the R channel value of a mature hazelnut shell might be between 120-150, the G channel value between 90-120, and the B channel value between 60-90.

[0035] The embodiment of the present invention calculates the maturity of each nut in each test group in the target batch by combining the RGB values ​​corresponding to each pixel in the surface image of the nut. By analyzing the RGB value of each pixel, rich color information can be obtained instead of simply judging the maturity based on the overall color. This allows for a more accurate assessment of the maturity of the nut, can adapt to complex surface color changes, and improves the objectivity and consistency of nut maturity detection.

[0036] The external quality coefficient of each nut in each test group in the target batch is obtained by weighted average of the roundness compliance coefficient and maturity of each nut. For example, the external quality coefficient of each nut in each test group in the target batch is δ ij , Among them, a1 and a2 represent the set circularity compliance coefficient and maturity corresponding external quality assessment weight, a1+a2=1, χ ij is the maturity of each nut in each test group.

[0037] In a specific embodiment of the present invention, the setting value of a1 is 0.5, and the setting value of a2 is 0.5. The external quality evaluation corresponding to the roundness compliance coefficient and maturity is not independent of each other, but complements each other. When comprehensively evaluating the external quality coefficient of nuts, it is necessary to comprehensively consider these two factors and weigh their relative importance according to the specific evaluation purpose and application scenario.

[0038] See also Figure 2 As shown, the external quality coefficient of each nut in each test group in the target batch is compared with the external quality coefficient of the set reference. If the external quality coefficient of a nut in a certain test group is greater than the external quality coefficient of the set reference, it indicates that the external quality of the nut in the test group is qualified; otherwise, it indicates that the external quality of the nut in the test group is unqualified.

[0039] The specific method of triggering the sorting device to remove unqualified products is as follows: once it is detected that the external quality of a nut in each inspection group is unqualified, the system will send a signal to the robotic arm sorting device, and the system will send the position information of the nut with unqualified external quality to the robotic arm control system. The robotic arm control system controls the robotic arm to perform a grabbing operation at an appropriate position based on the received position information, so that the nuts with unqualified external quality in each inspection group are removed from the inspection conveyor belt and placed in the unqualified external quality area.

[0040] S3. Weight screening of nuts in the test group: Count the number of remaining nuts in each test group, obtain the total mass of each test group through a weighing device, and compare it with the set standard nut weight range. If the total mass does not meet the standard, the nuts in the test group will be processed.

[0041] It should be noted that the total weight of nuts in each test group in the target batch is collected by a weight sensor placed under the test conveyor belt.

[0042] The specific process of the content of step S3 is: multiplying the number of remaining nuts in each detection group by the weight range corresponding to the single target nut extracted from the database to obtain the standard nut weight range corresponding to each detection group.

[0043] See also Figure 3 As shown, the total weight of nuts in each test group in the target batch is compared with the standard nut weight range corresponding to each test group in the target batch. If the total weight of nuts in a test group in the target batch is within the standard nut weight range corresponding to the test group, it indicates that the nut weight of the test group in the target batch meets the standard; otherwise, it indicates that the nut weight of the test group in the target batch does not meet the standard.

[0044] It's important to note that accurately weighing nuts indirectly assesses their flesh quality. Full nuts are generally heavier, while shrunken or under-fleshed nuts are lighter. This method is simple and straightforward, but it only provides a rough estimate. Therefore, further research will be conducted to understand the internal details of the nut flesh.

[0045] In a specific embodiment of the present invention, the specific method of processing the nuts in the inspection group is: weighing each nut in the inspection group where the nut weight does not meet the standard in turn, obtaining the weight of each nut in the inspection group where the nut weight does not meet the standard, and comparing it with the weight range corresponding to the single target nut; if the weight of a nut in the inspection group where the nut weight does not meet the standard is not within the weight range corresponding to the single target nut, the nut in the inspection group where the nut weight does not meet the standard is screened.

[0046] The embodiment of the present invention weighs the nuts before the intrinsic quality inspection of the nuts to preliminarily screen out nuts that do not meet the weight standards, thereby reducing the amount of nuts entering the intrinsic quality inspection link, improving the efficiency of the nut intrinsic quality inspection, and reducing the inspection cost.

[0047] The embodiment of the present invention uses a robotic arm to sort nuts that do not meet the external quality standards and screen nuts that do not meet the weight standards, and promptly screens nuts with abnormal quality, providing effective data basis for the comprehensive quality analysis of subsequent target batches of nuts. At the same time, automated screening can improve the quality and production efficiency of nut products.

[0048] S4. Analysis of the intrinsic quality of nuts: The internal structure of the target nuts after screening is imaged, and the internal defect parameters of each target nut are calculated using image analysis to generate the intrinsic quality qualification coefficient and count the number of internally unqualified nuts.

[0049] It should be noted that the internal images of each target nut in each inspection group in the target batch are acquired by an X-ray detector placed near the inspection conveyor belt.

[0050] The intrinsic quality qualification coefficient generation process is: using the sieved test group nuts as target nuts, and locating the number of grayscale areas and the grayscale values ​​corresponding to each grayscale area from the internal image of each target nut in each test group.

[0051] The grayscale values ​​corresponding to each grayscale area of ​​each target nut in each detection group are compared with the normal grayscale value range corresponding to the flesh of the target nuts stored in the database. If the grayscale value corresponding to a certain grayscale area of ​​a target nut in a certain detection group is within the normal grayscale value range corresponding to the flesh of the target nut, then the grayscale area is recorded as the normal flesh area, and the number of normal flesh areas of each target nut in each detection group in the target batch is counted.

[0052] The ratio deviation analysis of the number of normal areas in the flesh of each target nut and the number of grayscale areas of the corresponding nut was performed to obtain the intrinsic quality qualification coefficient of each target nut.

[0053] The analytical formula for the intrinsic quality qualification coefficient of each target nut is: Among them, K represents the proportion of the normal area of ​​the pulp set as the reference, e represents the natural constant, σ ig is the number of normal flesh areas of each target nut in each test group, is the number of grayscale areas for each target nut in each test group. This formula is used to quantify the intrinsic quality qualification coefficient of the target nut. The core logic is to compare the proportion of the normal flesh area with the set standard, and then use an exponential function to highlight the difference, reflecting the degree of quality deviation.

[0054] In a specific embodiment of the present invention, the method for counting the number of intrinsically unqualified nuts is: comparing the intrinsic quality qualification coefficient of each target nut in each detection group with the intrinsic quality qualification coefficient of the set reference; if the intrinsic quality qualification coefficient of a target nut in a certain detection group is less than the intrinsic quality qualification coefficient of the set reference, then the target nut is recorded as an intrinsically unqualified nut, and the number of intrinsically unqualified nuts in each detection group is counted.

[0055] S5. Comprehensive quality assessment of nuts: Evaluate whether the comprehensive quality of nuts in the target batch is qualified based on the number of unqualified nuts. If unqualified, feedback will be provided immediately.

[0056] In a specific embodiment of the present invention, the specific process of evaluating whether the comprehensive quality of nuts in the target batch is qualified is: counting the number of nuts with unqualified external quality, the number of nuts that do not meet the nut weight standard, and the number of nuts with unqualified internal quality in each inspection group, and adding them up to obtain the number of nuts with abnormal quality in each inspection group in the target batch.

[0057] If the number of nuts with abnormal quality in a test group is 0, then the test group is recorded as a test group with normal quality. The number of test groups with normal quality in the target batch is counted and recorded as

[0058] The ratio analysis of the number of test groups with normal quality and the number of test groups in the target batch is performed to obtain the comprehensive quality coefficient θ of the nuts in the target batch. The calculation formula of the comprehensive quality coefficient of the nuts in the target batch is: Among them, K1 represents the proportion of the number of normal quality test groups set as the reference, and n represents the number of test groups. This formula is used to quantify the comprehensive quality level of the target batch of nuts. The core logic is to compare the difference between the actual proportion of the normal quality test group and the set standard, and then use the exponential function to calculate the difference. Amplify the differences and finally adjust the numerical range using the natural logarithm to reflect the overall quality of the batch.

[0059] Compare the comprehensive quality coefficient of nuts in the target batch with the comprehensive quality coefficient of nuts in the set reference. If the comprehensive quality coefficient of nuts in the target batch is greater than or equal to the comprehensive quality coefficient of nuts in the set reference, it indicates that the comprehensive quality of nuts in the target batch is qualified; otherwise, it indicates that the comprehensive quality of nuts in the target batch is unqualified.

[0060] The embodiment of the present invention analyzes whether the comprehensive quality of nuts is qualified from two levels: external quality and internal quality of nuts. It can comprehensively evaluate the appearance, internal and overall quality of nuts. Through multi-dimensional feature extraction and analysis, the quality of nuts can be accurately classified and judged, thereby realizing the automation of nut quality detection, improving detection efficiency, reducing labor costs and subjective errors in manual detection, and performing quality analysis from two levels at the same time, which will make the implementation of quality standards clear and effectively protect the legitimate rights and interests of consumers.

[0061] The above contents are merely examples and explanations of the concept of the present invention. Those skilled in the art may make various modifications or additions to the described specific embodiments or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, they should all fall within the scope of protection of the present invention.

Claims

1. A nut quality detection and analysis method based on visual technology, characterized in that: The steps include: S1. Nut external information collection: Divide the target batch of nuts into several inspection groups according to a preset ratio. The nuts in each inspection group are transported via an inspection conveyor belt. The perimeter, area, and surface image data of all nuts in each inspection group are obtained using a visual acquisition device. S2. Nut external quality assessment: Based on the surface image data, the preset appearance parameter threshold is used to determine whether the appearance quality of the nuts is qualified. If there are nuts with unqualified appearance, the sorting device is triggered to remove the unqualified nuts; S3. Weight screening of nuts in the test group: Count the number of remaining nuts in each test group, obtain the total mass of each test group using a weighing device, and compare it with the set standard nut weight range. If the total mass does not meet the standard, the nuts in the test group will be processed; S4. Nut internal quality analysis: The internal structure of the sieved target nuts is imaged, and the internal defect parameters of each target nut are calculated using image analysis to generate an internal quality qualification coefficient and count the number of internally unqualified nuts. S5. Comprehensive quality assessment of nuts: Evaluate whether the comprehensive quality of nuts in the target batch is qualified based on the number of unqualified nuts. If unqualified, feedback will be provided immediately.

2. The nut quality detection and analysis method based on visual technology according to claim 1, characterized in that: The specific process of judging whether the appearance quality of nuts is qualified is as follows: The circumference and area of ​​each nut in each test group were analyzed for circularity, and the circularity compliance coefficient of each nut was determined based on the circularity; The RGB channel color mean is located from the image of each nut in each test group, and is compared with the RGB channel color range value when the nut is mature to obtain the maturity of each nut; The roundness compliance coefficient and maturity of each nut are weighted and averaged to obtain the external quality coefficient of each nut in each test group in the target batch; The external quality coefficient of each nut in each test group in the target batch is compared with the external quality coefficient of the set reference. If the external quality coefficient of a nut in a test group is greater than the external quality coefficient of the set reference, it indicates that the external quality of the nut in the test group is qualified; otherwise, it indicates that the external quality of the nut in the test group is unqualified.

3. The nut quality detection and analysis method based on visual technology according to claim 2, characterized in that: The specific process of calculating the roundness compliance coefficient of each nut is as follows: The circumference and area of ​​each nut in each test group are calculated using the roundness formula to obtain the roundness of each nut. The standard roundness of the target nuts is extracted from the database. The roundness of each nut is analyzed for deviation from the corresponding standard roundness to obtain the roundness compliance coefficient of each nut.

4. The method for detecting and analyzing nut quality based on visual technology according to claim 2, characterized in that: The specific process of the maturity of each nut is as follows: From the surface images of each nut in each test group, the R value, G value, and B value corresponding to each pixel are located, and the color mean of the R, G, and B channels are obtained by mean calculation; Compare the color mean values ​​of the R, G, and B channels with the color range values ​​of the R, G, and B channels of the target nuts when they are mature, stored in the database. If the color mean values ​​of the R, G, and B channels of a nut are all within the color range values ​​of the R, G, and B channels of the target nuts when they are mature, the maturity of the nut is recorded as τ1; If the color mean values ​​of the R, G, and B channels of a nut are not within the color range of the R, G, and B channels of the target nut when it is mature, the maturity of the nut is recorded as τ2; If any of the color mean values ​​of the R, G, and B channels of a nut is not within the color range of the R, G, and B channels of the target nut when it is mature, the maturity of the nut is recorded as τ3; In summary, the maturity of each nut in each test group was statistically analyzed.

5. The nut quality detection and analysis method based on visual technology according to claim 2, characterized in that: The specific method of triggering the sorting device to remove unqualified products is as follows: once it is detected that the external quality of a nut in each inspection group is unqualified, the system will send a signal to the robotic arm sorting device, and the system will send the position information of the nut with unqualified external quality to the robotic arm control system. The robotic arm control system controls the robotic arm to perform a grabbing operation at an appropriate position based on the received position information, so that the nuts with unqualified external quality in each inspection group are removed from the inspection conveyor belt and placed in the unqualified external quality area.

6. The nut quality detection and analysis method based on visual technology according to claim 5, characterized in that: The specific process of step S3 is as follows: Multiply the number of remaining nuts in each test group by the weight interval corresponding to the single target nut extracted from the database to obtain the standard nut weight interval corresponding to each test group; Compare the total weight of nuts in each test group with the corresponding standard nut weight range. If the total weight of nuts in a test group is within the corresponding standard nut weight range, it means that the nut weight of the test group meets the standard; otherwise, it means that the nut weight of the test group does not meet the standard.

7. The nut quality detection and analysis method based on visual technology according to claim 6, characterized in that: The specific method of processing the nuts in the inspection group is: weighing each nut in the inspection group whose nut weight does not meet the standard in turn, obtaining the weight of each nut in the inspection group whose nut weight does not meet the standard, and comparing it with the weight range corresponding to the single target nut; if the weight of a nut in the inspection group whose nut weight does not meet the standard is not within the weight range corresponding to the single target nut, the nut in the inspection group whose nut weight does not meet the standard is screened.

8. The method for detecting and analyzing nut quality based on visual technology according to claim 7, characterized in that: The intrinsic quality qualification coefficient generation process is: The sieved nuts in the test group are used as target nuts, and the number of grayscale areas and the grayscale values ​​corresponding to the grayscale areas are located from the internal images of the target nuts in each test group; Compare the grayscale values ​​corresponding to each grayscale area of ​​each target nut in each test group with the normal grayscale value interval corresponding to the flesh of the target nut stored in the database. If the grayscale value corresponding to a grayscale area of ​​a target nut in a certain test group is within the normal grayscale value interval corresponding to the flesh of the target nut, then record the grayscale area as the normal flesh area. Count the number of normal flesh areas of each target nut in each test group in the target batch. The ratio deviation analysis of the number of normal areas in the flesh of each target nut and the number of grayscale areas of the corresponding nut was performed to obtain the intrinsic quality qualification coefficient of each target nut.

9. The method for detecting and analyzing nut quality based on visual technology according to claim 8, characterized in that: The method for counting the number of intrinsically unqualified nuts is as follows: comparing the intrinsic quality qualification coefficient of each target nut in each testing group with the intrinsic quality qualification coefficient of the set reference; if the intrinsic quality qualification coefficient of a target nut in a certain testing group is less than the intrinsic quality qualification coefficient of the set reference, then the target nut is recorded as an intrinsically unqualified nut, and the number of intrinsically unqualified nuts in each testing group is counted.

10. The method for detecting and analyzing nut quality based on visual technology according to claim 9, characterized in that: The specific process of evaluating whether the comprehensive quality of nuts in the target batch is qualified is as follows: Count the number of nuts with unqualified external quality, the number of nuts that do not meet the nut weight standard, and the number of nuts with unqualified internal quality in each test group, and add them up to obtain the number of nuts with abnormal quality in each test group; If the number of nuts with abnormal quality in a test group is 0, then the test group is recorded as a test group with normal quality, and the number of test groups with normal quality in the target batch is counted; Perform a ratio analysis on the number of test groups with normal quality and the number of test groups in the target batch to obtain the comprehensive quality coefficient of nuts in the target batch; Compare the comprehensive quality coefficient of nuts in the target batch with the comprehensive quality coefficient of nuts in the set reference. If the comprehensive quality coefficient of nuts in the target batch is greater than or equal to the comprehensive quality coefficient of nuts in the set reference, it indicates that the comprehensive quality of nuts in the target batch is qualified; otherwise, it indicates that the comprehensive quality of nuts in the target batch is unqualified.

Images