A dynamic management and batch risk early warning method based on agricultural product qualification certificates

Abstract

This invention relates to the field of agricultural product safety technology, and in particular to a method for dynamic management and batch risk early warning of agricultural product certificates. The invention determines the batch information of the agricultural product to be tested and obtains the batch physicochemical risk value describing the chemical test results of that batch; it performs visual inspection on the agricultural product to be tested and calculates the individual product appearance risk value based on the abnormal area identification results; it determines the electronic certificate status corresponding to the agricultural product to be tested based on the individual product appearance risk value and the batch physicochemical risk value; and it extends from individual product anomaly identification to batch-level risk early warning and dynamic updating of certificates. This invention overcomes the shortcomings of static electronic certificate status, difficulty in timely linkage and early warning of quality anomalies, and lagging batch-level traceability response. It combines image segmentation detection results with batch traceability relationships based on the dynamic management and batch risk early warning method of agricultural product certificates, realizing batch-level quality early warning and dynamic updating of electronic certificates triggered by individual product anomalies. This improves the efficiency of agricultural product quality anomaly identification, batch traceability capabilities, and the timeliness of quality and safety supervision.

Description

Technical Field

[0001] This invention relates to the field of agricultural product safety technology, and in particular to a method for dynamic management and batch risk early warning of agricultural product qualification certificates. Background Technology

[0002] Agricultural products play a crucial role in ensuring food supply and quality safety. In the processes of agricultural product circulation, storage, sorting, and sales, electronic certificates of conformity have gradually become an important vehicle for traceability and circulation supervision. Currently, most agricultural product certificates of conformity are generated only once during production or packaging, and their status is usually a static record, making it difficult to dynamically adjust them based on actual quality changes during subsequent testing, circulation, or sampling. Meanwhile, some agricultural products may exhibit appearance quality problems such as rot, insect infestation, disease spots, mold, mechanical damage, or abnormal discoloration during post-harvest storage and transportation. Traditional methods mainly rely on manual sampling and registration, which is not only inefficient but also makes it difficult to promptly detect problems and simultaneously link them to products in the same batch in large-scale circulation scenarios. This allows problematic agricultural products to continue entering the circulation chain, increasing food safety risks and the pressure on quality supervision.

[0003] With the development of computer vision technology, image segmentation technology can perform pixel-level recognition of abnormal areas on the surface of agricultural products, providing a new technical means for the automatic detection of agricultural product appearance quality. However, existing visual inspection solutions mostly focus on the defect identification of individual agricultural products, and rarely organically combine the results of single-product appearance inspection with physicochemical indicators, batch-level traceability, electronic certificate status management, and abnormal risk linkage update mechanisms. In addition to appearance abnormalities, agricultural products may also encounter physicochemical quality problems such as excessive pesticide residues and abnormal heavy metal content during the circulation or inspection process. Existing electronic certificate management methods are usually based on static registration, making it difficult to respond promptly to quality risk information generated in subsequent inspection stages, especially when some products from the same batch have already been circulated or managed separately, often failing to achieve rapid linkage updates of abnormal information. If a batch of a single product is determined to have physicochemical abnormalities at any inspection stage, this abnormal risk may actually affect the quality judgment and certificate validity of other agricultural products in the same batch. Therefore, it is necessary to propose an agricultural product quality and safety management method that can combine image segmentation and recognition results, batch physicochemical indicators, traceability relationship between individual products and batches, and dynamic electronic certificate status control, so as to realize automatic identification of agricultural product quality anomalies, batch-level linkage early warning, and dynamic updating of certificate status. Summary of the Invention

[0004] To overcome the shortcomings of existing technologies, such as static status of electronic certificates of conformity for agricultural products, difficulty in timely linkage and early warning of quality anomalies, and lag in batch-level traceability response, this invention proposes a method based on dynamic management of agricultural product certificates of conformity and batch risk early warning that combines image segmentation detection results with batch traceability relationships. This enables batch-level quality early warning and dynamic updates of electronic certificates of conformity triggered by anomalies in individual products, thereby improving the efficiency of agricultural product quality anomaly identification, batch traceability capabilities, and the timeliness of quality and safety supervision.

[0005] This invention proposes a method for dynamic management of agricultural product qualification certificates and batch risk early warning, which determines the batch information of agricultural products to be tested and obtains the batch physicochemical risk value describing the chemical test results of the batch of agricultural products; performs visual inspection on the agricultural products to be tested and calculates the individual product appearance risk value based on the abnormal area identification results; The status of the electronic certificate of conformity for the agricultural product to be tested is determined based on the individual product appearance risk value and the batch physicochemical risk value. Determine whether the electronic certificate status meets the preset linkage conditions; if it does, obtain other agricultural products in the batch to which the agricultural product to be tested belongs, and update the electronic certificate status of other agricultural products in the same batch synchronously according to the preset batch linkage rules.

[0006] Preferably, it includes the following steps: S1. Obtain image information of the agricultural product to be detected and batch information corresponding to the agricultural product to be detected; preprocess the image information to obtain a standardized input image; S2. Identify abnormal regions in the standardized input image; S3. Extract abnormal features based on the abnormal region segmentation results, and generate individual product appearance risk values ​​based on the abnormal features; obtain the batch physicochemical risk value of the batch of agricultural products to be tested, and determine the electronic certificate status of the agricultural products to be tested based on the individual product appearance risk value and the batch physicochemical risk value. S4. Determine whether the electronic certificate status meets the preset linkage conditions; if it does, obtain other agricultural products in the batch to which the agricultural product to be tested belongs, and update the electronic certificate status of other agricultural products in the same batch synchronously according to the preset batch linkage rules.

[0007] Preferably, the abnormal characteristics include: the percentage of abnormal area, the number of abnormal areas, the category of abnormal areas, and the severity of the abnormality; the method for calculating the individual appearance risk value of the agricultural product to be tested is as follows:

[0008] in, Indicates the first The individual appearance risk value of agricultural products. This indicates the percentage of the area in the abnormal region. The score indicates the severity of the abnormality. Indicates the weight of the anomaly category. The weighting coefficients are set.

[0009] The preferred formula for calculating the batch physicochemical risk value is:

[0010] in, Indicates the first The comprehensive quality risk value of an individual agricultural product. This indicates the risk value of an individual item's appearance. Indicates the batch to which the agricultural product belongs. The physicochemical risk value, For fusion weighting coefficients.

[0011] Preferably, the electronic certificate status includes a healthy status, a warning status, and an abnormal status, each of which has a corresponding range of comprehensive quality risk values ​​for a single product. In step S3, the comprehensive quality risk value of a single product is obtained by weighted summation of the appearance risk value of a single product and the physicochemical risk value of the batch, and then the electronic certificate status is determined according to the range of the comprehensive quality risk value of the single product.

[0012] Preferably, the linkage conditions include: (1) The overall quality risk value of a single agricultural product is greater than the first preset threshold; (2) The proportion of abnormal agricultural products in the same batch is greater than the second preset threshold; abnormal agricultural products are agricultural products with corresponding abnormal status; (3) A preset high-risk anomaly category was detected; (4) Abnormalities were detected multiple times in a row.

[0013] Preferably, the batch linkage rules include at least one of the following: (1) Strict linkage mode: single point triggers the batch status update rule. When any single product is abnormal, the electronic certificate of conformity of the entire batch of agricultural products is uniformly updated to abnormal. (2) Proportional linkage mode: The abnormal proportion triggers the batch status update rule. When the proportion of abnormal agricultural products in the same batch exceeds the threshold, the electronic certificate of conformity of the entire batch of agricultural products is uniformly updated to abnormal. (3) Severity linkage mode: The severity of the anomaly triggers the batch status update rule. When high-risk anomaly categories such as excessive heavy metals are detected, the entire batch of anomalies is directly triggered.

[0014] (4) Progressive linkage mode: hierarchical progressive status update rules, the first abnormality triggers an early warning, and if the re-inspection is still abnormal, it is updated to abnormal again.

[0015] Preferably, abnormal areas include: rotten areas, insect-infested areas, moldy areas, lesion areas, abrasion areas, cracked areas, and abnormally discolored areas.

[0016] This invention proposes a dynamic management system for agricultural product qualification certificates based on image segmentation, comprising: Image acquisition module, image preprocessing module, abnormal region segmentation module, quality assessment module, batch association module, status linkage module, and electronic certificate output module; The image acquisition module is used to acquire images and batch information of the agricultural products to be inspected; The image preprocessing module is used to standardize the images acquired by the image acquisition module. The abnormal region segmentation module is used to detect and identify abnormal regions in agricultural products; The quality assessment module is used to generate individual product quality risk values ​​and determine the status of electronic certificates of conformity based on the segmentation results and batch physicochemical indicators. The batch association module is used to establish the mapping relationship between individual agricultural products, batches, and electronic certificates of conformity; The status linkage module is used to update the status of agricultural products in the same batch according to preset linkage conditions and batch linkage rules; The electronic certificate output module is used to output the updated status information of the electronic certificate.

[0017] Preferably, it also includes an exception handling module, which triggers at least one of the following after the batch linkage rule is triggered: a sales suspension prompt, a re-inspection prompt, an inventory lock prompt, or a recall notice for the corresponding batch.

[0018] The advantages of this invention are: (1) This invention integrates the risk information of individual product appearance obtained from image segmentation with the batch physicochemical test results to form a model, and establishes a dynamic linkage mechanism between individual product, batch, and electronic certificate of conformity through batch association. When any link detects an abnormality in the appearance of an individual product or its batch, it can quickly trigger the synchronous adjustment of the status of the electronic certificate of conformity for the same batch of agricultural products, realizing the extension from individual product abnormality identification to batch-level risk warning and dynamic updating of certificates of conformity. In this way, by combining the individual product appearance test results with the physicochemical test results and associating them with the batch traceability relationship, the batch to which the product belongs can be quickly located after an abnormality is found, improving the efficiency of quality abnormality location and traceability.

[0019] (2) The present invention uses image segmentation technology to perform pixel-level recognition of abnormal areas on the surface of agricultural products, which can improve the automation and detection accuracy of abnormal appearance quality recognition of agricultural products.

[0020] (3) By setting batch linkage rules, this invention can realize the dynamic update of the electronic certificate status of agricultural products in the same batch triggered by the abnormality of a single product, which can reduce the risk of problematic agricultural products continuing to circulate.

[0021] (4) This invention uses electronic statuses such as health, early warning and abnormality to classify and express the quality risks of agricultural products, which helps to improve the intuitiveness of agricultural product quality and safety management and the efficiency of regulatory response.

[0022] (5) This invention supports the integration of abnormal area visualization results, physicochemical indicators and batch linkage update records into electronic certificates of conformity, which can enhance the interpretability of test results and the visualization of traceability information. Attached Figure Description

[0023] Figure 1 This is a flowchart of a method for dynamic management and batch risk early warning of agricultural product qualification certificates. Detailed Implementation

[0024] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. 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 of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0025] like Figure 1 As shown, this invention provides a method for dynamic management and batch risk early warning of agricultural product qualification certificates, including the following steps: S1. Obtain image information of the agricultural product to be detected and the batch information corresponding to the agricultural product to be detected; preprocess the image information to obtain a standardized input image.

[0026] Image information is acquired through one or more of the following: industrial cameras, mobile terminal cameras, sorting line cameras, or warehouse monitoring terminals. Batch information may include: batch number, unique identifier for each item, electronic certificate number, harvesting time, packaging time, and warehousing details.

[0027]

[0028] in, Represents the original acquired image. Indicates the first An example of an agricultural product, Indicates the first Single agricultural product With its batch and electronic certificate of conformity Establish relationships.

[0029] This step is used to acquire image information of the agricultural products to be inspected and their corresponding traceability information, and to establish a mapping relationship between individual agricultural products, batches and electronic certificates of conformity, so as to provide basic data support for subsequent anomaly identification, status determination and batch linkage updates.

[0030] The preprocessing includes at least one of the following: image denoising, color correction, illumination compensation, and image size normalization.

[0031]

[0032] in, This represents the i-th original acquired image. This represents the standardized input image after preprocessing. This represents the preprocessing function.

[0033] S2. Perform abnormal region identification on the standardized input image. Specifically, an image segmentation model can be used to segment the abnormal region to obtain the abnormal region segmentation result. The image segmentation model is specifically an instance segmentation model. Let the image segmentation model be... ,but:

[0034] in, Indicates the first The results of segmenting abnormal regions for individual agricultural products. The standardized image after preprocessing, if containing multiple anomaly category regions, can be represented as follows:

[0035] in Indicates the first The first agricultural product Anomaly category regions.

[0036] Abnormal areas include at least one of the following: rotten areas, insect-infested areas, moldy areas, lesion areas, abrasion areas, cracked areas, and abnormally discolored areas.

[0037] S3. Extract abnormal features based on the abnormal region segmentation results, and generate individual product appearance risk values ​​based on the abnormal features; obtain the batch physicochemical risk value of the agricultural product to be tested, and determine the electronic certificate status of the agricultural product to be tested based on the individual product appearance risk value and the batch physicochemical risk value.

[0038] The abnormal characteristics include: the percentage of abnormal area, the number of abnormal areas, the category of abnormal areas, and the severity of the abnormality.

[0039] The risk value for the appearance of a single item is calculated based on the following relationship:

[0040] in, Indicates the first The individual appearance risk value of agricultural products. This indicates the percentage of the area in the abnormal region. The score indicates the severity of the abnormality. This represents the weight of the anomaly category. When multiple anomaly regions exist, then... Take the maximum weight. These are the weighting coefficients.

[0041] Specifically, the severity of anomalies can be scored using a trained model, resulting in an anomaly severity score. The weights for each anomaly category are set values ​​and correspond one-to-one with the anomaly category.

[0042] The batch physicochemical risk value is calculated based on the following relationship:

[0043] in, Indicates batch Batch physicochemical risk values, Indicates batch The pesticide residue test value, This indicates the pesticide residue limit threshold. Indicates batch The heavy metal detection value, This indicates the threshold for heavy metal limits. These are the weighting coefficients. This is a function to enhance the risk of exceeding the limit. This is the growth coefficient.

[0044] The status of the electronic certificate of conformity for the agricultural product to be tested can be determined based on the comprehensive quality risk value of the individual product.

[0045] The overall quality risk value of a single item is calculated based on the following relationship:

[0046] in, Indicates the first The comprehensive quality risk value of an individual agricultural product. This indicates the risk value of an individual item's appearance. Indicates the batch to which the agricultural product belongs. The physicochemical risk value, For fusion weighting coefficients.

[0047] The electronic certificate of conformity status includes a healthy status, a warning status, and an abnormal status, each with a corresponding range of comprehensive quality risk values ​​for a single product. The healthy status corresponds to a low-risk status, the warning status corresponds to a pending review status, and the abnormal status corresponds to an abnormal warning status.

[0048] The unique traceability identifier for agricultural products is established through coding. It can be generated based on at least one of the following: time information, batch number, single product number, place of origin information, or random coding, and expressed in the form of QR code, barcode, or electronic code. After the unique traceability identifier is generated, it is bound to the image data of the corresponding agricultural product, the batch identifier, and the electronic certificate of conformity, and stored in the traceability management database. This allows the sales end to search for the corresponding electronic certificate of conformity in the traceability management database through the unique traceability identifier, thereby checking the health status of the agricultural product.

[0049] The agricultural products to be tested are linked to their batch identifier and electronic certificate of conformity through a unique traceability identifier. The status information of the electronic certificate of conformity may also include visualization results of abnormal areas, batch physicochemical indicators, and batch linkage records.

[0050] S4. Determine whether the electronic certificate status meets the preset linkage conditions; if it does, obtain other agricultural products in the batch to which the agricultural product to be tested belongs, and update the electronic certificate status of other agricultural products in the same batch synchronously according to the preset batch linkage rules.

[0051] Specifically, the linkage conditions include: (1) The overall quality risk value of a single agricultural product is greater than the first preset threshold; (2) The proportion of abnormal agricultural products in the same batch is greater than the second preset threshold; abnormal agricultural products are agricultural products with corresponding abnormal status; (3) Pre-defined high-risk anomaly categories were detected, such as rotten areas, moldy areas, etc.; (4) Abnormalities were detected multiple times in a row.

[0052] In practice, to improve efficiency, when the linkage condition (4) occurs, it is not necessary to continue testing the remaining items in this batch.

[0053] Batch linkage rules include at least one of the following: (1) Strict linkage mode: single point triggers the batch status update rule. When any single product is abnormal, the electronic certificate of conformity of the entire batch of agricultural products is uniformly updated to abnormal; specifically, it can be set as: trigger linkage condition (1) or (3); (2) Proportional linkage mode: The abnormal proportion triggers the batch status update rule. When the proportion of abnormal agricultural products in the same batch exceeds the threshold, the electronic certificate of conformity of the entire batch of agricultural products is uniformly updated to abnormal. Specifically, it can be set as: trigger linkage condition (2). (3) Severity linkage mode: The severity of the anomaly triggers the batch status update rule. For example, when a high-risk anomaly category such as heavy metal exceeding the standard is detected, the entire batch of anomalies is directly triggered. (4) Progressive linkage mode: hierarchical progressive status update rules. The first abnormality triggers an early warning. If the re-inspection is still abnormal, it will be updated to an abnormality again. This situation indicates that the abnormality characteristics become more serious over time, and all electronic certificates of conformity in the same batch need to be updated.

[0054] In this implementation, the linkage condition is used to determine whether batch linkage needs to be triggered; the linkage rules clarify how to update the status of the electronic certificate of conformity for the same batch after the linkage condition is triggered.

[0055] In practice, if the batch linkage rule is triggered, after the electronic certificate of conformity status of the same batch of agricultural products is updated to an abnormal status, at least one of the following will be triggered: sales suspension notice, re-inspection notice, inventory lock notice, or recall notice, so as to achieve a rapid response to abnormal results.

[0056] Of course, those skilled in the art will recognize that the present invention is not limited to the details of the exemplary embodiments described above, but also includes the same or similar structures that can be implemented in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered illustrative and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0057] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

[0058] The technologies, shapes, and structures not described in detail in this invention are all known technologies.

Claims

1. A method for dynamic management and batch risk early warning of agricultural product qualification certificates, characterized in that, Determine the batch information of the agricultural products to be tested and obtain the batch physicochemical risk value describing the chemical test results of the batch of agricultural products; conduct visual inspection on the agricultural products to be tested and calculate the individual product appearance risk value based on the abnormal area identification results; The status of the electronic certificate of conformity for the agricultural product to be tested is determined based on the individual product appearance risk value and the batch physicochemical risk value. Determine whether the electronic certificate status meets the preset linkage conditions; if it does, obtain other agricultural products in the batch to which the agricultural product to be tested belongs, and update the electronic certificate status of other agricultural products in the same batch synchronously according to the preset batch linkage rules.

2. The method for dynamic management and batch risk early warning based on agricultural product qualification certificates as described in claim 1, characterized in that, Includes the following steps: S1. Obtain image information of the agricultural product to be detected and the batch information corresponding to the agricultural product to be detected; preprocess the image information to obtain a standardized input image; S2. Identify abnormal regions in the standardized input image; S3. Extract abnormal features based on the abnormal region segmentation results, and generate individual product appearance risk values ​​based on the abnormal features; Obtain the batch physicochemical risk value of the agricultural product to be tested, and determine the electronic certificate status of the agricultural product to be tested based on the individual product appearance risk value and the batch physicochemical risk value. S4. Determine whether the electronic certificate status meets the preset linkage conditions; if it does, obtain other agricultural products in the batch to which the agricultural product to be tested belongs, and update the electronic certificate status of other agricultural products in the same batch synchronously according to the preset batch linkage rules.

3. The method for dynamic management and batch risk early warning based on agricultural product qualification certificates as described in claim 2, characterized in that, Abnormal characteristics include: the percentage of abnormal area, the number of abnormal areas, the category of abnormal areas, and the severity of the abnormality; the calculation method for the appearance risk value of the agricultural product to be tested is as follows: in, Indicates the first The individual appearance risk value of agricultural products. This indicates the percentage of the area in the abnormal region. The score indicates the severity of the abnormality. Indicates the weight of the anomaly category. The weighting coefficients are set.

4. The method for dynamic management and batch risk early warning based on agricultural product qualification certificates as described in claim 3, characterized in that, The formula for calculating the batch physicochemical risk value is: in, Indicates the first The comprehensive quality risk value of an individual agricultural product. Indicates the risk value of an individual item's appearance. Indicates the batch to which the agricultural product belongs. The physicochemical risk value, For fusion weighting coefficients.

5. The method for dynamic management and batch risk early warning based on agricultural product qualification certificates as described in claim 4, characterized in that, The electronic certificate status includes health status, warning status, and abnormal status, each with a corresponding range of individual product comprehensive quality risk value. In step S3, the individual product appearance risk value and the batch physicochemical risk value are weighted and summed to obtain the individual product comprehensive quality risk value. Then, the electronic certificate status is determined based on the range of the individual product comprehensive quality risk value.

6. The method for dynamic management and batch risk early warning based on agricultural product qualification certificates as described in claim 5, characterized in that, The linkage conditions include: (1) The overall quality risk value of a single agricultural product is greater than the first preset threshold; (2) The proportion of abnormal agricultural products in the same batch is greater than the second preset threshold; abnormal agricultural products are agricultural products with corresponding abnormal status; (3) A preset high-risk anomaly category was detected; (4) Abnormalities were detected multiple times in a row.

7. The method for dynamic management and batch risk early warning based on agricultural product qualification certificates as described in claim 6, characterized in that, Batch linkage rules include at least one of the following: (1) Strict linkage mode: single point triggers the batch status update rule. When any single product is abnormal, the electronic certificate of conformity of the entire batch of agricultural products is uniformly updated to abnormal. (2) Proportional linkage mode: The abnormal proportion triggers the batch status update rule. When the proportion of abnormal agricultural products in the same batch exceeds the threshold, the electronic certificate of conformity of the entire batch of agricultural products is uniformly updated to abnormal. (3) Severity linkage mode: The severity of the anomaly triggers the batch status update rule. When high-risk anomaly categories such as excessive heavy metals are detected, the entire batch of anomalies is directly triggered. (4) Progressive linkage mode: hierarchical progressive status update rules, the first abnormality triggers an early warning, and if the re-inspection is still abnormal, it is updated to abnormal again.

8. The method for dynamic management and batch risk early warning based on agricultural product qualification certificates as described in claim 1, characterized in that, Abnormal areas include: rotten areas, insect-infested areas, moldy areas, lesion areas, abrasion areas, cracked areas, and abnormally discolored areas.

9. A dynamic management system for agricultural product qualification certificates based on image segmentation, used to implement the dynamic management system for agricultural product qualification certificates based on image segmentation as described in any one of claims 1-8, characterized in that, include: Image acquisition module, image preprocessing module, abnormal region segmentation module, quality assessment module, batch association module, status linkage module, and electronic certificate output module; The image acquisition module is used to acquire images and batch information of the agricultural products to be inspected; The image preprocessing module is used to standardize the images acquired by the image acquisition module. The abnormal region segmentation module is used to detect and identify abnormal regions in agricultural products; The quality assessment module is used to generate individual product quality risk values ​​and determine the status of electronic certificates of conformity based on the segmentation results and batch physicochemical indicators. The batch association module is used to establish the mapping relationship between individual agricultural products, batches, and electronic certificates of conformity; The status linkage module is used to update the status of agricultural products in the same batch according to preset linkage conditions and batch linkage rules; The electronic certificate output module is used to output the updated status information of the electronic certificate.

10. The dynamic management system for agricultural product qualification certificates based on image segmentation as described in claim 9, characterized in that, It also includes an exception handling module. After the batch linkage rule is triggered, the exception handling module will trigger at least one of the following for the corresponding batch: a sales suspension notice, a re-inspection notice, an inventory lock notice, or a recall notice.

Images