A method and system for non-destructive testing of durian kernels
By using CT scans and image processing technology, the durian seed area was segmented, and the number and content of seeds were calculated, solving the problem of inaccurate classification of durian seed quality and achieving accurate classification of durian quality.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- ZHEJIANG BOSHI NETWORK TECHNOLOGY CO LTD
- Filing Date
- 2026-03-23
- Publication Date
- 2026-07-07
AI Technical Summary
In existing technologies, classifying durians by weight, appearance, and internal defects cannot accurately classify the quality of the durian seeds.
CT tomography was used to acquire durian images. The seed area was segmented using image processing and edge detection algorithms. The number and content of seeds were calculated, and the seed quality score was calculated by combining the weight values to achieve seed quality classification.
This improves the accuracy of durian seed quality classification, enabling direct acquisition of the internal seed attributes and achieving more precise quality evaluation.
Smart Images
Figure CN122345628A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of non-destructive testing technology for fruits, specifically to a non-destructive testing method and system for durian seeds. Background Technology
[0002] In existing technologies, the quality inspection of durians usually involves testing the weight, appearance, and internal defects of the durians to grade the durians based on their quality, appearance, and internal defects, thus classifying durians into different grades.
[0003] However, the size and shape of the seeds of durians of different varieties or from different places of origin vary, and their appearance is also different. The appearance of durians is not related to the quality of the seeds. Therefore, classifying durians solely by weight, appearance, and internal defects cannot classify the quality of durian seeds, leading to inaccurate classification of durian seed quality. Summary of the Invention
[0004] To address the technical problems in existing technologies, such as inaccurate quality classification of durian seeds, this invention provides a non-destructive testing method and system for durian seeds.
[0005] The technical solution of the present invention to solve the above-mentioned technical problems is as follows: A non-destructive testing method for durian seeds includes the following steps: Acquire CT scan images of durian; The number of durian seeds was determined based on the CT scan images; Predict the seed content of the durian based on the CT scan images; The weight score of the durian seeds is calculated based on the number of seeds and the seed content. The weight of the durian seed is determined based on the seed weight score.
[0006] The beneficial effects of this invention are: by determining the number of seeds and predicting the content of seeds to evaluate the quality of seeds, the quality of durian can be determined by directly obtaining the attributes of the seeds inside the durian, thereby improving the accuracy of durian quality classification.
[0007] Based on the above technical solution, the present invention can be further improved as follows.
[0008] Furthermore, determining the number of durian seeds based on the CT scan images includes the following steps: The CT tomographic image is processed to obtain a grayscale image; The grayscale image is subjected to image enhancement processing to obtain an enhanced grayscale image; Determine the region of interest containing the fruit kernel in the enhanced grayscale image; An edge-based image segmentation method is used to segment the kernel image of the region of interest, resulting in multiple kernel segmentation images. The number of fruit kernels is obtained by counting the number of segmented images of multiple fruit kernels.
[0009] Furthermore, the image enhancement processing is specifically a grayscale transformation image processing method.
[0010] Furthermore, the grayscale transformation image processing method is a contrast enhancement method, a contrast stretching method, or a grayscale level layering method.
[0011] Further, determining the region of interest where the fruit kernel is located in the enhanced grayscale image includes the following steps: The Canny edge detection algorithm is used to detect the edges of the fruit core region in the enhanced grayscale image; The region of the edge envelope of the fruit pit region in the enhanced grayscale image is determined as the region of interest.
[0012] Furthermore, an edge-based image segmentation method is used to segment the kernel image of the region of interest, resulting in multiple kernel-segmented images, including the following steps: Extract the image of the region of interest to obtain the image of the fruit pit region; The image of the fruit pit region is subjected to Gaussian filtering to remove noise, resulting in a preprocessed image of the fruit pit region. The Canny operator was used to detect the closed boundaries of each kernel in the preprocessed kernel region image; The findContours function is used to extract the contours of the closed boundaries of all the fruit kernels; Image filling is performed inside the contour of each closed boundary to obtain multiple kernel segmentation images.
[0013] Furthermore, predicting the seed content of the durian based on the CT scan image includes the following steps: The contour of the closed boundary of the durian pulp in the CT tomographic image was detected based on the boundary extraction method. The cross-sectional area of the durian is calculated based on the contour of the closed boundary of the durian flesh. The cross-sectional area of each fruit pit is calculated based on the contour of the closed boundary of each fruit pit; Calculate the sum of the cross-sectional areas of all the fruit pits to obtain the equivalent cross-sectional area value of the fruit pits; The ratio of the equivalent cross-sectional area of the fruit pit to the cross-sectional area of the durian is used to obtain the fruit pit content value.
[0014] Furthermore, the weight score of the durian seeds is calculated based on the number of seeds and the seed content, including the following steps: Weight values are assigned to the number of fruit pits and the content of fruit pits respectively, and the corresponding quantity weight value and content weight value are obtained; The quality score of the fruit kernels is calculated based on the number of kernels, the quantity weight value, the kernel content value, and the content weight value.
[0015] Furthermore, the quality score of the fruit pits is calculated based on the number of pits, the quantity weight value, the content value of the fruit pits, and the content weight value, using the following formula: ; in, This indicates the weight score of the fruit pit. This indicates the number of fruit pits. This represents the quantity weight value. This indicates the content value of the fruit kernel. This represents the content weight value.
[0016] To address the aforementioned technical problems, this invention also provides a non-destructive testing system for durian seeds, the specific technical details of which are as follows: A non-destructive testing system for durian seeds includes an image acquisition module, an image processing module, a calculation module, and a quality judgment module; The image acquisition module is used to acquire CT scan images of durian; The image processing module is used to determine the number of durian seeds based on the CT scan image; and to predict the seed content value of the durian based on the CT scan image. The calculation module is used to calculate the fruit seed quality score of the durian based on the number of fruit seeds and the fruit seed content value; The quality judgment module is used to determine the quality of the durian's seeds based on the seed quality score. Attached Figure Description
[0017] Figure 1 This is a flowchart of a non-destructive testing system for durian seeds according to an embodiment of the present invention; Figure 2 The image shown is a CT scan of a durian in an embodiment of the present invention. Figure 3 This is a schematic diagram of a non-destructive testing system for durian seeds according to an embodiment of the present invention. Detailed Implementation
[0018] The principles and features of the present invention are described below with reference to the accompanying drawings. The examples given are only for explaining the present invention and are not intended to limit the scope of the present invention.
[0019] like Figure 1 As shown, this embodiment provides a non-destructive testing method for durian seeds, including the following steps: S1. Acquire CT tomographic images of durian; A CT scan device was installed on the durian weighing and testing equipment to obtain data such as... Figure 2 The image shown is a CT scan of a durian, used to classify durians by quality.
[0020] S2. Determine the number of durian seeds based on the CT scan images; Determining the number of durian seeds based on the CT scan images includes the following steps: The CT tomographic image is processed to obtain a grayscale image; The grayscale image is subjected to image enhancement processing to obtain an enhanced grayscale image; the image enhancement processing is specifically a grayscale transformation image processing method. The grayscale transformation image processing method is a contrast enhancement method, a contrast stretching method, or a grayscale level layering method.
[0021] Contrast enhancement method: When the slope is greater than 1, the grayscale range of the image is expanded, making dark areas darker and bright areas brighter, thereby enhancing details. For example, multiplying the grayscale value by 1.5 can improve contrast.
[0022] Contrast stretching method: The grayscale range [a, b] of the original image is linearly mapped to [c, d], expanding the dynamic range of the range, while the grayscale values outside the range are cropped or kept unchanged, thereby improving the overall contrast.
[0023] Gray-level layering method: Highlights a specific gray-level range (such as the seed area of a durian in this invention) while suppressing other irrelevant gray levels, and is often used for target detection.
[0024] Determine the region of interest containing the fruit kernel in the enhanced grayscale image; Determining the region of interest containing the fruit kernel in the enhanced grayscale image includes the following steps: The Canny edge detection algorithm is used to detect the edges of the kernel region in the enhanced grayscale image. Based on the OpenCV library, Gaussian filtering is used for image smoothing to reduce noise. The image gradient is calculated using the Canny operator and non-maximum suppression is applied to refine the edges. A dual-threshold detection method is used to distinguish between strong and weak edges, and continuous edges are preserved through edge concatenation. Finally, a binarized edge image is output for subsequent image analysis or processing.
[0025] The region of the edge envelope of the fruit pit region in the enhanced grayscale image is determined as the region of interest.
[0026] An edge-based image segmentation method is used to segment the kernel image of the region of interest, resulting in multiple kernel segmentation images. An edge-based image segmentation method is used to segment the kernel image of the region of interest, resulting in multiple kernel-segmented images, including the following steps: Extract the image of the region of interest to obtain the image of the fruit pit region; The image of the fruit pit region is subjected to Gaussian filtering to remove noise, resulting in a preprocessed image of the fruit pit region. The Canny operator was used to detect the closed boundaries of each kernel in the preprocessed kernel region image; The findContours function is used to extract the contours of the closed boundaries of all the fruit kernels; Image filling is performed inside the contour of each closed boundary to obtain multiple kernel segmentation images.
[0027] The number of fruit kernels is obtained by counting the number of segmented images of multiple fruit kernels.
[0028] S3. Predict the seed content of the durian based on the CT scan image; Predicting the seed content of the durian based on the CT scan image includes the following steps: The contour of the closed boundary of the durian pulp in the CT tomographic image was detected based on the boundary extraction method. The cross-sectional area of the durian is calculated based on the contour of the closed boundary of the durian flesh. Area calculation: For each found closed contour, the cv2.contourArea() function is called to calculate the pixel area of the region enclosed by each closed contour.
[0029] The cross-sectional area of each fruit pit is calculated based on the contour of the closed boundary of each fruit pit; Calculate the sum of the cross-sectional areas of all the fruit pits to obtain the equivalent cross-sectional area value of the fruit pits; The ratio of the equivalent cross-sectional area of the fruit pit to the cross-sectional area of the durian is used to obtain the fruit pit content value.
[0030] S4. Calculate the durian seed quality score based on the number of seeds and the seed content value; calculating the durian seed quality score based on the number of seeds and the seed content value includes the following steps: Weight values are assigned to the number of fruit pits and the content of fruit pits respectively, and the corresponding quantity weight value and content weight value are obtained; The quality score of the fruit kernels is calculated based on the number of kernels, the quantity weight value, the kernel content value, and the content weight value.
[0031] Furthermore, the quality score of the fruit pits is calculated based on the number of pits, the quantity weight value, the content value of the fruit pits, and the content weight value, using the following formula: ; in, This indicates the weight score of the fruit pit. This indicates the number of fruit pits. This represents the quantity weight value. This indicates the content value of the fruit kernel. This represents the content weight value. The weight can be set according to actual classification requirements.
[0032] For example, when the classification standard is based primarily on the number of fruit pits, the quantity weight value can be assigned to 0.7; then the content weight value can be assigned to 0.3.
[0033] S5. Determine the weight of the durian kernel based on the kernel weight score.
[0034] Multiple classification intervals can be constructed based on experience, with each interval corresponding to a classification criterion.
[0035] The first classification interval corresponds to the low-quality category. If the seed quality score falls within the first interval, the durian seed is determined to be of low quality.
[0036] A second classification interval is defined to correspond to the medium quality category. If the seed quality score falls within the second interval, the durian seed quality is determined to be medium.
[0037] A third category is defined to correspond to the high-quality category. If the seed quality score falls within the third category, the durian seed quality is determined to be high-quality.
[0038] This invention, through determining the number and predicted content of fruit seeds, calculating a weighted score for each seed, and then evaluating the seed quality based on that score, directly obtains the attributes of the seeds inside the durian to determine its quality, thereby improving the accuracy of durian quality classification.
[0039] like Figure 3 As shown, in some other embodiments, a non-destructive testing system for durian seeds is also provided, including an image acquisition module, an image processing module, a calculation module, and a quality judgment module; The image acquisition module is used to acquire CT scan images of durian; The image processing module is used to determine the number of durian seeds based on the CT scan image; and to predict the seed content value of the durian based on the CT scan image. The calculation module is used to calculate the fruit seed quality score of the durian based on the number of fruit seeds and the fruit seed content value; The quality judgment module is used to determine the quality of the durian seed based on the seed quality score. The image acquisition module, image processing module, calculation module, and quality judgment module are all program modules or computing devices.
[0040] In some other embodiments, a storage medium is also provided, which stores a computer program or computer instructions that, when executed by a computer's processor, implement the steps of the above-described non-destructive testing method for durian seeds.
[0041] The storage medium can be an internal storage unit of any data processing device described in any of the foregoing embodiments, such as a hard disk or memory. The storage medium can also be an external storage device of any data processing device, such as a plug-in hard disk, smart memory card, SD card, flash memory card, etc., mounted on the device. Furthermore, the storage medium can include both internal storage units and external storage devices of any data processing device. The computer-readable storage medium is used to store the computer program and other programs and data required by the data processing device, and can also be used to temporarily store data that has been output or will be output.
[0042] In other embodiments, a computer is also provided, including a memory and one or more processors, wherein executable code is stored in the memory, and when the one or more processors execute the executable code, the steps of the above-described non-destructive testing method for durian seeds are implemented.
[0043] The memory can be an internal storage unit of any data processing device described in any of the foregoing embodiments, such as a hard disk or RAM. The memory can also be an external storage device of any data processing device, such as a plug-in hard disk, smart memory card, SD card, flash memory card, etc., mounted on the device. Furthermore, the memory can include both internal storage units and external storage devices of any data processing device. The memory is used to store the computer program and other programs and data required by the data processing device, and can also be used to temporarily store data that has been output or will be output.
[0044] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the concept and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A non-destructive testing method for durian seeds, characterized in that, Includes the following steps: Acquire CT scan images of durian; The number of durian seeds was determined based on the CT scan images; Predict the seed content of the durian based on the CT scan images; The weight score of the durian seeds is calculated based on the number of seeds and the seed content. The weight of the durian seed is determined based on the seed weight score.
2. The non-destructive testing method for durian seeds according to claim 1, characterized in that, Determining the number of durian seeds based on the CT scan images includes the following steps: The CT tomographic image is processed to obtain a grayscale image; The grayscale image is subjected to image enhancement processing to obtain an enhanced grayscale image; Determine the region of interest containing the fruit kernel in the enhanced grayscale image; An edge-based image segmentation method is used to segment the kernel image of the region of interest, resulting in multiple kernel segmentation images. The number of fruit kernels is obtained by counting the number of segmented images of multiple fruit kernels.
3. The non-destructive testing method for durian seeds according to claim 2, characterized in that, The image enhancement processing is specifically a grayscale transformation image processing method.
4. The non-destructive testing method for durian seeds according to claim 3, characterized in that, The grayscale transformation image processing method is either contrast enhancement, contrast stretching, or grayscale layering.
5. The non-destructive testing method for durian seeds according to claim 2, characterized in that, Determining the region of interest containing the fruit kernel in the enhanced grayscale image includes the following steps: The Canny edge detection algorithm is used to detect the edges of the fruit core region in the enhanced grayscale image; The region of the edge envelope of the fruit pit region in the enhanced grayscale image is determined as the region of interest.
6. The non-destructive testing method for durian seeds according to claim 2, characterized in that, An edge-based image segmentation method is used to segment the kernel image of the region of interest, resulting in multiple kernel-segmented images, including the following steps: Extract the image of the region of interest to obtain the image of the fruit pit region; The image of the fruit pit region is subjected to Gaussian filtering to remove noise, resulting in a preprocessed image of the fruit pit region. The Canny operator was used to detect the closed boundaries of each kernel in the preprocessed kernel region image; The findContours function is used to extract the contours of the closed boundaries of all the fruit kernels; Image filling is performed inside the contour of each closed boundary to obtain multiple kernel segmentation images.
7. The non-destructive testing method for durian seeds according to claim 6, characterized in that, Predicting the seed content of the durian based on the CT scan image includes the following steps: The contour of the closed boundary of the durian pulp in the CT tomographic image was detected based on the boundary extraction method. The cross-sectional area of the durian is calculated based on the contour of the closed boundary of the durian flesh. The cross-sectional area of each fruit pit is calculated based on the contour of the closed boundary of each fruit pit; Calculate the sum of the cross-sectional areas of all the fruit pits to obtain the equivalent cross-sectional area value of the fruit pits; The ratio of the equivalent cross-sectional area of the fruit pit to the cross-sectional area of the durian is used to obtain the fruit pit content value.
8. The non-destructive testing method for durian seeds according to claim 1, characterized in that, The durian seed quality score is calculated based on the number of seeds and the seed content, including the following steps: Weight values are assigned to the number of fruit pits and the content of fruit pits respectively, and the corresponding quantity weight value and content weight value are obtained; The quality score of the fruit kernels is calculated based on the number of kernels, the quantity weight value, the kernel content value, and the content weight value.
9. The non-destructive testing method for durian seeds according to claim 8, characterized in that, The quality score of the fruit pits is calculated based on the number of fruit pits, the quantity weight value, the fruit pit content value, and the content weight value, using the following formula: ; in, This indicates the weight score of the fruit pit. This indicates the number of fruit pits. This represents the quantity weight value. This indicates the content value of the fruit kernel. This represents the content weight value.
10. A system employing the non-destructive testing method for durian seeds as described in any one of claims 1 to 9, characterized in that, It includes an image acquisition module, an image processing module, a calculation module, and a quality judgment module; The image acquisition module is used to acquire CT scan images of durian; The image processing module is used to determine the number of durian seeds based on the CT scan image; and to predict the seed content value of the durian based on the CT scan image. The calculation module is used to calculate the fruit seed quality score of the durian based on the number of fruit seeds and the fruit seed content value; The quality judgment module is used to determine the quality of the durian's seeds based on the seed quality score.