Agricultural product quality control detection management system based on X-ray DR imaging technology

Abstract

The invention discloses an agricultural product quality control detection management system based on an X-ray DR imaging technology, relates to the technical field of agricultural product quality control detection, and mainly solves the technical problems that in the prior art, the defect recognition dimension is single, most of the defects are judged only according to gray values, insect damage holes and natural cavities are difficult to distinguish, and early mildew and normal chromatic aberration are difficult to distinguish. According to the method, through the full-process design of quantitative acquisition, parameter adaptation, optimal screening, feature analysis and quality control traceability, multiple core beneficial effects are achieved: based on accurate quantitative parameters, the problem of traditional X-ray imaging parameter empirical is solved, the imaging quality and the parameter adaptation are improved, multi-dimensional feature extraction and thresholding analysis are achieved, and the quality control traceability of the X-ray imaging system is improved. Accurate recognition and classification of nut hidden defects are achieved, a database and a traceability mechanism are linked, efficient management of data and dynamic optimization of a production line are achieved, and both detection precision and production efficiency are considered.

Description

Technical Field

[0001] This invention relates to the field of agricultural product quality control and testing technology, specifically to an agricultural product quality control and testing management system based on X-ray DR imaging technology. Background Technology

[0002] In the current field of quality control and testing of agricultural products (especially nuts), traditional methods rely heavily on manual screening and simple physical testing, which are inefficient, subjective, and have limited accuracy. However, with the large-scale production and consumption upgrading of agricultural products, the market has increasingly strict requirements for the quality of nuts. It is necessary not only to pay attention to the integrity of the appearance, but also to detect whether there are hidden defects such as insect damage, mold, and cavities inside.

[0003] Currently, X-ray DR imaging technology has been gradually applied to agricultural product quality control due to its advantages such as non-destructiveness and penetrability. However, significant technical bottlenecks still exist. On the one hand, the setting of imaging parameters relies on experience and is not specifically adapted to the quantitative characteristics of agricultural products (such as shell thickness and kernel density). This results in some thick-shelled nuts being imaged blurry or thin-shelled nuts being damaged due to excessive radiation dose. Furthermore, the defect identification dimension is singular, mostly relying on grayscale values, making it difficult to distinguish between insect-damaged holes and natural cavities, or early mold and normal color differences. On the other hand, data management is fragmented. The detection parameters, defect information, and agricultural product types have not formed a linked retrieval system, making it impossible to achieve production line traceability and dynamic parameter optimization. Consequently, the detection process is poorly adapted to the production line rhythm, and some imaging takes too long, making it difficult to meet the efficiency requirements of large-scale production.

[0004] To address the aforementioned technical shortcomings, a solution is proposed. Summary of the Invention

[0005] The purpose of this invention is to solve the problems mentioned above by proposing an agricultural product quality control and testing management system based on X-ray DR imaging technology.

[0006] The objective of this invention can be achieved through the following technical solution: an agricultural product quality control and testing management system based on X-ray DR imaging technology, including a quality control and testing management center, and a communication connection between the quality control and testing management center and the system. The quantitative data acquisition unit collects data based on the physical characteristics of agricultural products, thereby obtaining quantitative characteristic parameters of the agricultural products. The parameter combination test unit, upon receiving the parameter combination test signal, sets the X-ray DR imaging parameters based on the quantitative characteristic parameters of agricultural products, thus obtaining a matching combination. The parameter filtering unit filters the imaging electrical parameters of agricultural products corresponding to matching combinations. The database build and update unit performs database construction. The feature extraction unit extracts features from agricultural products on the real-time production line based on the parameter settings of the current database. The feature processing and analysis unit analyzes and processes the feature parameters.

[0007] Furthermore, the process of quantizing the acquisition unit is as follows: Different varieties of agricultural products were selected, and the number of samples was set. The average thickness and thickness range of the shells were measured using high-precision vernier calipers. The data were recorded and classified into thin-shelled products and thick-shelled products. The volume of the kernels was measured using the water displacement method, and the mass was measured using an electronic balance to calculate the kernel density. Average thickness, thickness range, and kernel density are uniformly labeled as quantitative characteristic parameters of agricultural products and then summarized.

[0008] Furthermore, the process of parameter combination testing unit is as follows: Voltage and current gradients are set and uniformly labeled as electrical parameters; the number of parameter combinations is obtained based on the actual set number of voltage and current gradients; batch imaging tests are performed on the collected sample agricultural products, and the imaging time of agricultural products during the imaging process is obtained based on the imaging test, while the radiation dose required for imaging during the imaging process is also obtained.

[0009] Furthermore, based on the voltage and current set for imaging agricultural products, the imaging time and radiation dose were analyzed: If the imaging time exceeds the time threshold satisfied by the current processing line, or the radiation dose exceeds the dose threshold set by the imaging criteria, it is inferred that the quantitative characteristic parameters of the current sample agricultural product do not match the corresponding electrical parameters. The corresponding parameter values ​​are then set as a mismatch combination and sent to the quality control and testing management center. If the imaging time does not exceed the time threshold satisfied by the current processing line, and the radiation dose does not exceed the dose threshold set by the imaging criteria, it is inferred that the quantitative characteristic parameters of the current sample agricultural product match the corresponding electrical parameters. The corresponding parameter values ​​are then set as a match combination and sent to the quality control and testing management center.

[0010] Furthermore, the parameter filtering process is as follows: Imaging analysis is performed on the matching combinations. Data is collected based on the agricultural product images to obtain the lowest specification of defects identified in the collected agricultural product images, and the defect specification is marked as a sharpness parameter. The grayscale difference between the outer shell and the kernel of the agricultural product in the collected images is obtained, and the grayscale difference is marked as grayscale discrimination. Parameters are then filtered based on the sharpness parameter and the grayscale discrimination.

[0011] Furthermore, if the clarity parameter is lower than the set specification threshold and the grayscale discrimination exceeds the set grayscale threshold under the current matched quantitative characteristic parameters of agricultural products and corresponding electrical parameters, then it is inferred that the electrical parameters matched by the current quantitative characteristic parameters of agricultural products are marked as the optimal electrical parameters. If the clarity parameter is not lower than the set specification threshold or the grayscale discrimination does not exceed the set grayscale threshold under the current matched quantitative characteristic parameters of agricultural products and corresponding electrical parameters, then it is inferred that the electrical parameters matched by the current quantitative characteristic parameters of agricultural products are marked as non-optimal electrical parameters. Gradient division is performed based on the quantitative characteristic parameters of agricultural products. Based on the optimal and non-optimal electrical parameters of the corresponding parameters, the optimal electrical parameter range is constructed for the current quantitative characteristic parameters of agricultural products. The construction method is to determine the range of the optimal electrical parameters and shorten the range based on the non-optimal electrical parameters. The quantitative characteristic parameters of each type of agricultural product and the corresponding optimal electrical parameter range are sent to the quality control monitoring and management center.

[0012] Furthermore, the database construction and update unit builds a database based on the collected agricultural product types, corresponding quantitative characteristic parameters, and optimal electrical parameter ranges. The agricultural product type, quantitative characteristic parameters, and optimal electrical parameter range are all set as searchable tags. That is, when any data is used as a search tag for retrieval, the two types of data other than the search tag can be displayed as search results.

[0013] Furthermore, the process of the feature extraction unit is as follows: The characteristic parameters of agricultural products are obtained by imaging them, including porosity, grayscale standard deviation, roundness, area ratio, and local minimum grayscale difference. The porosity is calculated as follows: number of pixels in the defect / cavity area ÷ number of pixels in the overall area of ​​the agricultural product × 100%. Gray-scale standard deviation calculation method: Statistical analysis of gray-scale values ​​in the defective area, and calculation of the standard deviation; Roundness calculation method: Area percentage: Calculation method: Defective area area ÷ Cross-sectional area of ​​agricultural product × 100%; Local minimum grayscale difference: Calculation method: local minimum grayscale value of defective area - grayscale value of healthy kernel area; The feature parameters collected in real time are sent to the feature processing and analysis unit.

[0014] Furthermore, the feature processing and analysis unit analyzes the feature parameters of the agricultural product corresponding to the current production line. If the porosity is within the set porosity range, it is inferred that the porosity of the current agricultural product is qualified; if the porosity is not within the set porosity range, a pest signal is generated and sent to the quality control and testing management center along with the current agricultural product number. If the grayscale standard deviation is lower than the set standard deviation threshold, it indicates that the grayscale of the current agricultural product is uniform; conversely, if the grayscale standard deviation is not lower than the set standard deviation threshold, an early moldy signal is generated and sent to the quality control and testing management center along with the current agricultural product number. If the roundness exceeds the set roundness threshold, it is inferred that the cavity of the agricultural product is close to a regular circle, that is, the roundness is qualified; if the roundness does not exceed the set roundness threshold, a hole deformation signal is generated and sent to the quality control and testing management center along with the current agricultural product number. If the area ratio is lower than the area ratio threshold, it is inferred that the area ratio of the agricultural product is qualified; if the area ratio is not lower than the area ratio threshold, a defect signal is generated and sent to the quality control and testing management center along with the current agricultural product number; if the local minimum grayscale difference is lower than the set minimum grayscale difference threshold, it is inferred that the grayscale difference test of the agricultural product is qualified; if the local minimum grayscale difference is not lower than the set minimum grayscale difference threshold, a mold aggravation signal is generated and sent to the quality control and testing management center along with the current agricultural product number.

[0015] Furthermore, the quality control and testing management center screens agricultural products based on the agricultural products and corresponding type signals, and conducts traceability control based on the rejected agricultural products to reduce the number of rejected agricultural products in the production line, and packages and stores the remaining agricultural products.

[0016] Compared with the prior art, the beneficial effects of the present invention are: 1. The quantitative acquisition unit obtains quantitative characteristic parameters through high-precision tools, providing objective and accurate data for subsequent X-ray imaging parameter setting. This avoids the blindness of traditional experience-based parameter setting. It classifies and processes parameters according to shell thickness type, making subsequent parameter testing more targeted and ensuring that nuts with different characteristics can obtain suitable imaging conditions, thereby improving the stability of imaging quality from the source.

[0017] 2. The parameter combination test unit comprehensively covers possible imaging conditions through multiple parameter combinations, ensuring that no optimal matching scheme is missed. It judges parameter matching by using time and dose dual thresholds, which avoids excessive imaging time affecting production line efficiency and prevents excessive radiation dose from damaging nut quality, thus achieving a balance between detection efficiency and product safety. It can also significantly narrow down the range of subsequent parameter screening based on the elimination mechanism of mismatched combinations, thereby improving the efficiency of determining the optimal parameters.

[0018] 3. The parameter screening unit performs secondary screening on matching combinations using sharpness parameters (the lowest specification for defect identification) and grayscale discrimination (the grayscale difference between the shell and the kernel), and further divides the optimal electrical parameter range by combining the gradient of quantitative characteristic parameters. The use of dual indicators of sharpness and grayscale discrimination ensures that the screened electrical parameters simultaneously meet the requirements for defect identification accuracy and internal / external structural differentiation, improving the practicality of imaging. Constructing the optimal electrical parameter range according to the gradient of quantitative characteristic parameters achieves a precise mapping between "agricultural product characteristics" and "imaging parameters," providing customized imaging solutions for nuts of different specifications and avoiding image quality fluctuations caused by "one-size-fits-all" parameters. Furthermore, the range-based parameter settings offer greater flexibility than single parameters, adapting to the subtle differences in agricultural product characteristics in actual production.

[0019] 4. Compared with traditional single gray value detection, the feature extraction unit can more comprehensively and accurately reflect the internal quality of nuts with multi-dimensional parameters. Porosity is directly related to the degree of pest infestation, gray standard deviation corresponds to the risk of mold, roundness distinguishes the type of cavity, area ratio reflects the scale of defects, and local minimum gray difference helps to judge the degree of mold aggravation. This provides rich and three-dimensional data for subsequent defect identification and greatly reduces the probability of misjudgment by a single parameter. Meanwhile, thresholds are set for each characteristic parameter and judged one by one to generate accurate signals such as pests, early mold, and borer hole deformation, as well as corresponding agricultural product numbers. The advantages are as follows: First, the threshold-based judgment standard makes defect identification more objective and avoids the subjective error of human judgment. Second, the accurate classification of signals for different defect types enables precise binding of "defect type - agricultural product number", which facilitates the subsequent targeted removal of unqualified products. Third, the early warning of signals such as early mold and borer hole deformation can detect hidden defects in time, prevent unqualified products from entering the market, and improve the foresight and strictness of quality control. Attached Figure Description

[0020] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings.

[0021] Figure 1 This is a system principle block diagram of the present invention; Figure 2 This is a flowchart of the method of the present invention. Detailed Implementation

[0022] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0024] Please see Figures 1-2 As shown, the agricultural product quality control and testing management system based on X-ray DR imaging technology includes a quality control and testing management center. The quality control and testing management center is connected to a quantitative acquisition unit, a parameter combination testing unit, a parameter screening unit, and a database construction and update unit. Simultaneously, the quality control and testing management center is also connected to a feature extraction unit and a feature processing and analysis unit. In this application, the agricultural product being tested is specifically nut products. The quantitative acquisition unit, parameter combination testing unit, parameter screening unit, and database construction and update unit connected to the quality control and testing management center constitute the first stage; the feature extraction unit and feature processing and analysis unit constitute the second stage. The quality control and testing management center generates quantitative acquisition signals and sends them to the quantitative acquisition unit; After receiving the quantitative acquisition signal, the quantitative acquisition unit performs quantitative acquisition on the agricultural products based on their physical characteristics. Select different varieties of agricultural products and set the sample size. Use a high-precision vernier caliper (0.01mm accuracy) to measure the average thickness and thickness range of the shells (e.g., walnut shells are usually 1.2-2.5mm thick, hazelnut shells are 0.8-1.5mm thick). Record the data and classify them into thin-shelled products and thick-shelled products. The volume of the kernels was measured using the water displacement method (shelled kernels were immersed in a graduated cylinder and the amount of water displaced was measured), and the mass was measured using an electronic balance (accuracy 0.01g). The kernel density was then calculated (e.g., the density of walnut kernels is approximately 0.65g / cm³, and that of hazelnut kernels is approximately 0.72g / cm³). Average thickness, thickness range, and kernel density are uniformly labeled as quantitative characteristic parameters of agricultural products and then summarized. Simultaneously, a parameter combination test signal is generated and sent to the parameter combination test unit; After receiving the parameter combination test signal, the parameter combination test unit sets the X-ray DR imaging parameters according to the quantitative characteristic parameters of agricultural products; Voltage and current gradients are set and uniformly labeled as electrical parameters. Based on the actual set values ​​of voltage and current gradients, the number of parameter combinations is obtained, namely: voltage gradients: 40kV, 45kV, 50kV, 55kV, 60kV (covering the low-radiation, high-penetration range of soft X-rays); current gradients: 3mA, 4mA, 5mA, 6mA, 7mA (controlling radiation dose to avoid damage to agricultural products); thus forming 25 parameter combinations. Batch imaging tests were conducted on the collected agricultural products. The imaging time of the agricultural products during the imaging process was obtained based on the imaging test, and the radiation dose required for imaging was also obtained. Based on the voltage and current settings for imaging agricultural products, the imaging time and radiation dose are analyzed: If the imaging time exceeds the time threshold that the current processing production line can meet, or if the radiation dose exceeds the dose threshold set by the imaging criteria, it is inferred that the quantitative characteristic parameters of the current sample agricultural product do not match the corresponding electrical parameters. The corresponding parameter values ​​are then set as mismatched combinations and sent to the quality control and testing management center. If the imaging time does not exceed the time threshold of the current processing line and the radiation dose does not exceed the dose threshold set by the imaging criteria, then it is inferred that the quantitative characteristic parameters of the current sample agricultural product match the corresponding electrical parameters, the corresponding parameter values ​​are set as the matching combination, and sent to the quality control and testing management center. Simultaneously, a parameter filtering signal is generated and sent to the parameter filtering unit; After receiving the parameter filtering signal, the parameter filtering unit performs imaging electrical parameter filtering on agricultural products; Imaging analysis is performed on the matching combinations. Data is collected based on the agricultural product images to obtain the lowest specification of defects identified in the collected agricultural product images, and the defect specification is marked as a sharpness parameter. The grayscale difference between the outer shell and the kernel of the agricultural product in the collected images is obtained, and the grayscale difference is marked as the grayscale discrimination. Parameters are then filtered based on sharpness and grayscale differentiation. If the clarity parameter is lower than the set specification threshold and the grayscale distinction exceeds the set grayscale threshold under the current matched quantitative characteristic parameters of agricultural products and corresponding electrical parameters, then it is inferred that the electrical parameter matched by the current quantitative characteristic parameters of agricultural products is marked as the optimal electrical parameter. If the clarity parameter is not lower than the set specification threshold or the grayscale discrimination does not exceed the set grayscale threshold under the current matched quantitative characteristic parameters of agricultural products and corresponding electrical parameters, then it is inferred that the electrical parameters matched by the current quantitative characteristic parameters of agricultural products are marked as non-optimal electrical parameters. Gradient division is performed based on the quantitative characteristic parameters of agricultural products, and the optimal electrical parameter range is constructed based on the optimal and non-optimal electrical parameters of the corresponding parameters. The construction method is to determine the range of the optimal electrical parameters and shorten the range based on the non-optimal electrical parameters. The quantitative characteristic parameters of each type of agricultural product, along with the corresponding optimal electrical parameter range, are sent to the quality control monitoring and management center. After receiving the signal, the Quality Control Monitoring and Management Center generates a database construction and update signal and sends it to the Database Construction and Update Unit. After receiving the database build and update signal, the database build and update unit performs database build; Based on the collected agricultural product types, a database is built with the corresponding quantitative characteristic parameters and optimal electrical parameter ranges. Agricultural product types, quantitative characteristic parameters, and optimal electrical parameter ranges are all set as searchable tags. That is, when any data is used as a search tag for retrieval, the two types of data other than the search tag can be displayed as search results. After the database construction is completed, the quality control and testing management center generates a feature extraction signal and sends it to the feature extraction unit; After receiving the feature extraction signal, the feature extraction unit extracts features from the agricultural products on the real-time production line according to the parameter settings of the current database. The characteristic parameters of agricultural products are obtained by imaging them, including porosity, grayscale standard deviation, roundness, area ratio, and local minimum grayscale difference. The porosity is calculated as follows: number of pixels in the defect / cavity area ÷ number of pixels in the overall area of ​​the agricultural product × 100%. Gray-scale standard deviation calculation method: Statistical analysis of gray-scale values ​​in the defective area, and calculation of the standard deviation; Roundness calculation method: Area percentage: Calculation method: Defective area area ÷ Cross-sectional area of ​​agricultural product × 100%; Local minimum grayscale difference: Calculation method: local minimum grayscale value of defective area - grayscale value of healthy kernel area; The real-time collected feature parameters are sent to the feature processing and analysis unit; After receiving the feature parameters, the feature processing and analysis unit analyzes and processes the feature parameters. Based on the characteristic parameter analysis of the agricultural products corresponding to the current production line, if the porosity is within the set porosity range, it is inferred that the porosity of the current agricultural product is qualified; if the porosity is not within the set porosity range, it is inferred that the agricultural product has pests, that is, the more holes there are, the higher the proportion, or if it is lower than the minimum range value, it indicates that there is a deviation in the collection, and a pest signal is generated and sent to the quality control and testing management center along with the current agricultural product number. If the grayscale standard deviation is lower than the set standard deviation threshold, it indicates that the grayscale of the current agricultural product is uniform; conversely, if the grayscale standard deviation is not lower than the set standard deviation threshold, it indicates that the current agricultural product is moldy, causing uneven grayscale, generating an early moldy signal, which is sent to the quality control and testing management center along with the current agricultural product number. If the roundness exceeds the set roundness threshold, it is inferred that the cavity of the agricultural product is close to a regular circle, that is, the roundness is qualified; if the roundness does not exceed the set roundness threshold, it is inferred that the cavity of the agricultural product is a borehole caused by insect damage, that is, an irregular shape, and a borehole deformation signal is generated and sent to the quality control and testing management center along with the current agricultural product number. If the area ratio is lower than the area ratio threshold, it is inferred that the area ratio of the agricultural product is qualified; if the area ratio is not lower than the area ratio threshold, it is inferred that the agricultural product has a defect, and a defect signal is generated and sent to the quality control and testing management center along with the current agricultural product number. If the local minimum grayscale difference is lower than the set minimum grayscale difference threshold, it is inferred that the grayscale difference test of the agricultural product is qualified; if the local minimum grayscale difference is not lower than the set minimum grayscale difference threshold, it is inferred that the grayscale difference of the moldy area of ​​the agricultural product is abnormal, and a mold aggravation signal is generated and sent to the quality control and testing management center along with the current agricultural product number. The quality control and testing management center screens agricultural products based on the agricultural products and corresponding type signals, and conducts traceability control based on the rejected agricultural products to reduce the number of rejected agricultural products in the production line, and packages and stores the remaining agricultural products.

[0025] Thresholds, preset values, or preset ranges are set for result comparison and analysis to determine whether they are good or bad. The value of these thresholds is determined by a combination of large-scale model analysis of sample data and human experience. They can also be adjusted appropriately based on seasonal or rational factors.

[0026] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to any specific implementation. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. An agricultural product quality control and inspection management system based on X-ray DR imaging technology, characterized in that, Includes the Quality Control and Testing Management Center, and its communication connections: The quantitative data acquisition unit collects data based on the physical characteristics of agricultural products, thereby obtaining quantitative characteristic parameters of the agricultural products. The parameter combination test unit, after receiving the parameter combination test signal, sets the X-ray DR imaging parameters according to the quantitative characteristic parameters of agricultural products; Get the matching combination; The parameter filtering unit filters the imaging electrical parameters of agricultural products corresponding to matching combinations. The database build and update unit performs database construction. The feature extraction unit extracts features from agricultural products on the real-time production line based on the parameter settings of the current database. The feature processing and analysis unit analyzes and processes the feature parameters.

2. The agricultural product quality control and inspection management system based on X-ray DR imaging technology according to claim 1, characterized in that, The process of quantization acquisition unit is as follows: Different varieties of agricultural products were selected, and the number of samples was set. The average thickness and thickness range of the shells were measured using high-precision vernier calipers. The data were recorded and classified into thin-shelled products and thick-shelled products. The volume of the kernels was measured using the water displacement method, and the mass was measured using an electronic balance to calculate the kernel density. Average thickness, thickness range, and kernel density are uniformly labeled as quantitative characteristic parameters of agricultural products and then summarized.

3. The agricultural product quality control and inspection management system based on X-ray DR imaging technology according to claim 1, characterized in that, The process of parameter combination testing unit is as follows: Voltage and current gradients are set and uniformly labeled as electrical parameters; the number of parameter combinations is obtained based on the actual set number of voltage and current gradients; batch imaging tests are performed on the collected sample agricultural products, and the imaging time of agricultural products during the imaging process is obtained based on the imaging test, while the radiation dose required for imaging during the imaging process is also obtained.

4. The agricultural product quality control and inspection management system based on X-ray DR imaging technology according to claim 3, characterized in that, Based on the voltage and current settings for imaging agricultural products, the imaging time and radiation dose are analyzed: If the imaging time exceeds the time threshold satisfied by the current processing line, or the radiation dose exceeds the dose threshold set by the imaging criteria, it is inferred that the quantitative characteristic parameters of the current sample agricultural product do not match the corresponding electrical parameters. The corresponding parameter values ​​are then set as a mismatch combination and sent to the quality control and testing management center. If the imaging time does not exceed the time threshold satisfied by the current processing line, and the radiation dose does not exceed the dose threshold set by the imaging criteria, it is inferred that the quantitative characteristic parameters of the current sample agricultural product match the corresponding electrical parameters. The corresponding parameter values ​​are then set as a match combination and sent to the quality control and testing management center.

5. The agricultural product quality control and inspection management system based on X-ray DR imaging technology according to claim 1, characterized in that, The parameter filtering process is as follows: Imaging analysis is performed on the matching combinations, data is collected based on agricultural product imaging, the lowest specification of defects is identified in the collected agricultural product images, and the defect specification is marked as a sharpness parameter. The grayscale difference between the outer shell and the kernel of the agricultural product is obtained from the collected images of the agricultural product, and the grayscale difference is marked as the grayscale discrimination value; and the parameters are filtered according to the sharpness parameter and the grayscale discrimination value.

6. The agricultural product quality control and inspection management system based on X-ray DR imaging technology according to claim 5, characterized in that, If the clarity parameter is lower than the set specification threshold and the grayscale distinction exceeds the set grayscale threshold under the current matched quantitative characteristic parameters of agricultural products and corresponding electrical parameters, then it is inferred that the electrical parameter matched by the current quantitative characteristic parameters of agricultural products is marked as the optimal electrical parameter. If the clarity parameter is not lower than the set specification threshold or the grayscale discrimination does not exceed the set grayscale threshold under the current matched quantitative characteristic parameters of agricultural products and corresponding electrical parameters, then it is inferred that the electrical parameters matched by the current quantitative characteristic parameters of agricultural products are marked as non-optimal electrical parameters. Gradient division is performed based on the quantitative characteristic parameters of agricultural products. Based on the optimal and non-optimal electrical parameters of the corresponding parameters, the optimal electrical parameter range is constructed for the current quantitative characteristic parameters of agricultural products. The construction method is to determine the range of the optimal electrical parameters and shorten the range based on the non-optimal electrical parameters. The quantitative characteristic parameters of each type of agricultural product and the corresponding optimal electrical parameter range are sent to the quality control monitoring and management center.

7. The agricultural product quality control and inspection management system based on X-ray DR imaging technology according to claim 1, characterized in that, The database construction and update unit builds a database based on the collected agricultural product types, corresponding quantitative characteristic parameters, and optimal electrical parameter ranges. Agricultural product types, quantitative characteristic parameters, and optimal electrical parameter ranges are all set as searchable tags. That is, when any data is used as a search tag for retrieval, the two types of data other than the search tag can be displayed as search results.

8. The agricultural product quality control and inspection management system based on X-ray DR imaging technology according to claim 1, characterized in that, The process of the feature extraction unit is as follows: The characteristic parameters of agricultural products are obtained by imaging them, including porosity, grayscale standard deviation, roundness, area ratio, and local minimum grayscale difference. The porosity is calculated as follows: number of pixels in the defect / cavity area ÷ number of pixels in the overall area of ​​the agricultural product × 100%. Gray-scale standard deviation calculation method: Statistical analysis of gray-scale values ​​in the defective area, and calculation of the standard deviation; Roundness calculation method: ; Area percentage: Calculation method: Defective area area ÷ Cross-sectional area of ​​agricultural product × 100%; Local minimum grayscale difference: Calculation method: local minimum grayscale value of defective area - grayscale value of healthy kernel area; The feature parameters collected in real time are sent to the feature processing and analysis unit.

9. The agricultural product quality control and inspection management system based on X-ray DR imaging technology according to claim 8, characterized in that, The feature processing and analysis unit analyzes the feature parameters of the agricultural product corresponding to the current production line. If the porosity is within the set porosity range, it is inferred that the porosity of the current agricultural product is qualified; if the porosity is not within the set porosity range, a pest signal is generated and sent to the quality control and testing management center along with the current agricultural product number. If the grayscale standard deviation is lower than the set standard deviation threshold, it indicates that the grayscale of the current agricultural product is uniform; conversely, if the grayscale standard deviation is not lower than the set standard deviation threshold, an early moldy signal is generated and sent to the quality control and testing management center along with the current agricultural product number. If the roundness exceeds the set roundness threshold, it is inferred that the cavity of the agricultural product is close to a regular circle, that is, the roundness is qualified; If the roundness does not exceed the set roundness threshold, a hole deformation signal is generated and sent to the quality control and testing management center along with the current agricultural product number; If the area ratio is lower than the area ratio threshold, it is inferred that the area ratio of the agricultural product is qualified; if the area ratio is not lower than the area ratio threshold, a defect signal is generated and sent to the quality control and testing management center along with the current agricultural product number; if the local minimum grayscale difference is lower than the set minimum grayscale difference threshold, it is inferred that the grayscale difference test of the agricultural product is qualified; if the local minimum grayscale difference is not lower than the set minimum grayscale difference threshold, a mold aggravation signal is generated and sent to the quality control and testing management center along with the current agricultural product number.

10. The agricultural product quality control and inspection management system based on X-ray DR imaging technology according to claim 9, characterized in that, The quality control and testing management center screens agricultural products based on the agricultural products and corresponding type signals, and conducts traceability control based on the rejected agricultural products to reduce the number of rejected agricultural products in the production line, and packages and stores the remaining agricultural products.

Images