Agricultural product tracing method

Abstract

The invention provides an agricultural product traceability method, and relates to the technical field of agricultural product traceability, and the method comprises the steps: obtaining an image sequence corresponding to a target agricultural product in a preset period based on an agricultural product traceability platform; the agricultural product tracing platform identifies whether a target human body exists in the image sequence based on a pre-constructed human body identification model; under the condition that the target human body exists in all images of the image sequence, the agricultural product tracing platform inputs the image sequence into a pre-constructed behavior recognition model and outputs a sub-image sequence carrying behavior identifiers, and one behavior identifier corresponds to one behavior category; the sub-image sequence carrying the behavior identifier does not need to be repeatedly added to the traceability information corresponding to the target agricultural product; the tracing information of agricultural products can be automatically generated, the workload is reduced, and the accuracy of the tracing information is ensured.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of agricultural product traceability, and in particular to an agricultural product traceability method. BACKGROUND

[0002] Agricultural product safety is increasingly concerned by people, and traceability of agricultural products can control agricultural product safety to a certain extent. Existing agricultural planting data usually needs to be actively collected by users and then actively uploaded to an agricultural product traceability platform to generate traceability information. There are problems of large workload and low accuracy. SUMMARY

[0003] To solve one of the above technical problems, the present application provides an agricultural product traceability method, which can automatically generate traceability information of agricultural products, reduce workload, and ensure the accuracy of traceability information.

[0004] An agricultural product traceability method is provided in an embodiment of the present application and applied to an agricultural product traceability platform, comprising:

[0005] Based on the agricultural product traceability platform, an image sequence corresponding to a target agricultural product is acquired at a preset period;

[0006] The agricultural product traceability platform identifies whether a target human body exists in the image sequence based on a pre-constructed human body recognition model;

[0007] In a case where all images of the image sequence exist the target human body, the agricultural product traceability platform inputs the image sequence into a pre-constructed behavior recognition model to output a sub-image sequence carrying a behavior identifier, wherein one behavior identifier corresponds to one behavior category;

[0008] Based on the sub-image sequence, traceability information is generated.

[0009] In some embodiments, the behavior category includes planting, pesticide spraying, fertilization, picking, and weeding.

[0010] In some embodiments, comprising:

[0011] The behavior category in the traceability information is counted at a preset time node;

[0012] In a case where the behavior category does not match a preset behavior category, the agricultural product traceability platform feeds back first prompt information to a corresponding user terminal.

[0013] In some embodiments, comprising:

[0014] The behavior frequency corresponding to the behavior category in the traceability information is counted at a preset time node;

[0015] When the number of behaviors corresponding to a behavior category is less than the preset number, the agricultural product traceability platform sends a second prompt message to the corresponding user terminal.

[0016] In some embodiments, the human body recognition model includes:

[0017] The first input layer is used to receive single-frame images from the image sequence;

[0018] A convolutional layer, consisting of multiple convolutional kernels, is used to extract features from a single frame of an image and output a feature map.

[0019] Pooling layers are used to reduce the spatial dimensionality of feature maps;

[0020] The first fully connected layer is used to convert the feature map with reduced spatial dimensions into a one-dimensional vector.

[0021] The first output layer is used to output a binary classification result based on a one-dimensional vector to determine whether a target human body exists.

[0022] In some embodiments, the behavior recognition model includes:

[0023] The second input layer receives the image sequence;

[0024] Spatiotemporal convolutional layers are used to extract spatiotemporal features from image sequences;

[0025] Long Short-Term Memory (LSTM) network layers are used to handle the temporal dependencies of image sequences;

[0026] The second fully connected layer is used to transform the feature vector into a probability distribution of the behavior category;

[0027] The second output layer outputs a sequence of sub-images carrying behavioral identifiers.

[0028] In some embodiments, including:

[0029] If the behavior identifier carried by the sub-image sequence corresponding to the image sequence acquired in the current preset period is the same as the behavior identifier carried by the sub-image sequence corresponding to the image sequence acquired in the previous preset period, there is no need to generate new tracing information based on the sub-image sequence carrying the behavior identifier corresponding to the current preset period.

[0030] In some embodiments, including:

[0031] The sub-image sequence carrying the behavior identifier is hashed to generate a hash value;

[0032] The hash value and the product information of the target agricultural product are packaged into a block;

[0033] Broadcast the block to the blockchain network;

[0034] Nodes in a blockchain network verify blocks;

[0035] Once verified, the block is added to the blockchain.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] By using an agricultural product traceability platform, image sequences corresponding to target agricultural products are acquired at preset intervals. The agricultural product traceability platform identifies whether a target human body exists in the image sequence based on a pre-built human body recognition model. If a target human body is present in all images of the image sequence, the agricultural product traceability platform inputs the image sequence into a pre-built behavior recognition model and outputs a sub-image sequence carrying behavior identifiers, where each behavior identifier corresponds to a behavior category. Traceability information is generated based on the sub-image sequence. This method can automatically generate traceability information for agricultural products, reducing workload and ensuring the accuracy of traceability information. Attached Figure Description

[0038] The embodiments of the present invention will be further described below with reference to the accompanying drawings:

[0039] Figure 1 This is a schematic diagram illustrating the implementation process of an agricultural product traceability method provided in an embodiment of the present invention;

[0040] Figure 2 This is a schematic diagram of the composition structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limitations on 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.

[0042] In the following description, references are made to “some embodiments,” which describe a subset of all possible embodiments. However, it is understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0043] If the application documents contain similar descriptions such as "first, second, third", the following explanation shall be added: In the following description, the terms "first, second, third" are used only to distinguish similar objects and do not represent a specific order of objects. It is understood that "first, second, third" may be interchanged in a specific order or sequence where permitted, so that the embodiments of the present invention described herein can be implemented in an order other than that illustrated or described herein.

[0044] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein is for the purpose of describing embodiments of the invention only and is not intended to limit the invention.

[0045] To address the problems existing in related technologies, this invention provides a method for tracing the origin of agricultural products. The subject of this method can be an electronic device. The electronic device can be various types of terminals such as laptops, tablets, desktop computers, set-top boxes, and mobile devices (e.g., mobile phones, portable music players, personal digital assistants, dedicated messaging devices, portable gaming devices), or it can be implemented as a server. The server can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms.

[0046] In some embodiments, the functions implemented by the tracing method provided in this invention can be achieved by the processor of an electronic device calling program code, wherein the program code can be stored in a computer storage medium.

[0047] This invention provides a method for tracing the origin of agricultural products. Figure 1 This is a schematic diagram illustrating the implementation process of an agricultural product traceability method provided in an embodiment of the present invention, as shown below. Figure 1 As shown, the application in the agricultural product traceability platform includes:

[0048] Step S1: Based on the agricultural product traceability platform, obtain the image sequence corresponding to the target agricultural product at a preset cycle;

[0049] In this embodiment of the invention, an image acquisition device is used to acquire image sequences corresponding to the target agricultural products. The image acquisition device is connected to the agricultural product traceability platform. The agricultural product traceability platform acquires image sequences corresponding to the target agricultural products at a preset period. The user can set the preset period according to the type of the target agricultural product to adapt to different types of agricultural products.

[0050] Step S2: The agricultural product traceability platform identifies whether a target human body exists in the image sequence based on a pre-built human body recognition model;

[0051] In some embodiments, the human body recognition model includes:

[0052] The first input layer is used to receive single-frame images from the image sequence;

[0053] A convolutional layer, consisting of multiple convolutional kernels, is used to extract features from a single frame of an image and output a feature map.

[0054] Pooling layers are used to reduce the spatial dimensionality of feature maps;

[0055] The first fully connected layer is used to convert the feature map with reduced spatial dimensions into a one-dimensional vector.

[0056] The first output layer is used to output a binary classification result based on a one-dimensional vector to determine whether a target human body exists.

[0057] In this embodiment of the invention, a single-frame image from an image sequence is received through a first input layer. A convolutional layer slides a convolution kernel across the image, performing convolution operations on each local region to extract local features. The parameters of the convolution kernel are learned through training and can automatically capture important features in the image (such as edges and textures). This effectively extracts local features, reduces the number of parameters, and improves the model's computational efficiency and generalization ability. A pooling layer reduces the spatial dimension of the feature map through max pooling or average pooling operations while retaining important feature information. This reduces the size of the feature map, lowers the computational load, enhances the robustness of the features, and makes the model more robust to changes in image translation and scaling. A second fully connected layer converts the two-dimensional feature map into a one-dimensional feature vector and integrates global feature information through matrix operations. This integrates local features into global features, providing a comprehensive feature representation for classification tasks and improving classification accuracy. A second output layer performs binary classification based on the feature vector and outputs the result of whether a target human body exists. This filters the image sequence and reduces the workload of subsequent processing.

[0058] Step S3: When the target human body is present in all images of the image sequence, the agricultural product traceability platform inputs the image sequence into a pre-built behavior recognition model and outputs a sub-image sequence carrying a behavior identifier, wherein one behavior identifier corresponds to one behavior category;

[0059] In some embodiments, the behavior recognition model includes:

[0060] The second input layer receives the image sequence;

[0061] Spatiotemporal convolutional layers are used to extract spatiotemporal features from image sequences;

[0062] Long Short-Term Memory (LSTM) network layers are used to handle the temporal dependencies of image sequences;

[0063] The second fully connected layer is used to transform the feature vector into a probability distribution of the behavior category;

[0064] The second output layer outputs a sequence of sub-images carrying behavioral identifiers.

[0065] In this embodiment of the invention, an image sequence is received through a second input layer. The spatiotemporal convolutional layer processes both spatial (e.g., object shape, position) and temporal (e.g., object trajectory) information of the image through convolution operations. It uses convolutional kernels to slide across the image sequence, extracting local spatiotemporal features. This effectively captures the dynamic changes of human actions while preserving the spatial information of the image, improving the accuracy of behavior recognition. The Long Short-Term Memory (LSTM) network layer controls the flow of information through gating mechanisms (input gate, forget gate, output gate), enabling it to remember long-term dependencies and short-term changes. It can handle complex action sequences; for example, an action may consist of multiple steps (e.g., reaching out, grabbing, and finally moving). LSTM can remember the relationships between these steps, thus more accurately recognizing the entire action. The second fully connected layer converts the extracted feature vectors into probability distributions of behavior categories. It maps the feature vectors to the category space using a weight matrix and calculates the probability of each category using an activation function (e.g., softmax). This transforms complex feature vectors into intuitive probability distributions, facilitating the model's output of the final behavior category. The second output layer selects the most probable behavior category based on the probability distribution and appends the behavior identifier to the corresponding sub-image sequence, giving the output clear semantic information for easier subsequent analysis and application. The behavior categories include planting, spraying pesticides, fertilizing, harvesting, and weeding.

[0066] Step S4: Generate source information based on the sub-image sequence.

[0067] In some embodiments, including:

[0068] Step 41: If the behavior identifier carried by the sub-image sequence corresponding to the image sequence obtained in the current preset period is the same as the behavior identifier carried by the sub-image sequence corresponding to the image sequence obtained in the previous preset period, there is no need to generate new traceability information based on the sub-image sequence carrying the behavior identifier corresponding to the current preset period.

[0069] In this embodiment of the invention, the purpose of traceability information is to record key behaviors in the agricultural product production process, ensuring that consumers can trace the product's origin and production process. Repeatedly adding the same behavioral identifiers leads to information redundancy, increases data storage costs, and reduces the readability and usability of the traceability information. By avoiding the repeated addition of the same behavioral identifiers, the traceability information becomes more concise and clear, reduces unnecessary data redundancy, and improves the efficiency and reliability of the traceability system.

[0070] In some embodiments, including:

[0071] Step S101: At preset time points, statistically analyze the behavioral categories in the traceability information;

[0072] Step S102: If the behavior category does not match the preset behavior category, the agricultural product traceability platform will send a first prompt message to the corresponding user terminal.

[0073] In this embodiment of the invention, by setting time-node statistical behavior categories, the distribution of agricultural product behavior at different stages can be understood in a timely manner, providing data support for subsequent analysis and decision-making. By setting preset behavior categories, unexpected behaviors can be detected promptly, avoiding interference from erroneous or abnormal behaviors with traceability information. By sending an initial notification to the user terminal, relevant personnel can be quickly notified to take measures, improving the efficiency of problem-solving. This ensures the accuracy and completeness of traceability information, enhancing the reliability of the traceability system.

[0074] In some embodiments, including:

[0075] Step S201: At a preset time node, count the number of behaviors corresponding to the behavior categories in the traceability information;

[0076] Step S202: When the number of behaviors corresponding to a behavior category is less than the preset number, the agricultural product traceability platform sends a second prompt message to the corresponding user terminal.

[0077] In this embodiment of the invention, by statistically analyzing the number of actions, the frequency of each action can be determined, providing data support for subsequent analysis and decision-making. By setting a preset number of actions, insufficient execution frequency of certain key actions can be detected in a timely manner, avoiding the omission of important steps. By sending a second prompt message to the user terminal, relevant personnel can be quickly notified to take measures, improving the efficiency of problem-solving. This ensures the completeness and accuracy of traceability information, enhancing the credibility of the traceability system.

[0078] In some embodiments, including:

[0079] Step S301: Perform hash processing on the sub-image sequence carrying the behavior identifier to generate a hash value;

[0080] Step S302: Package the hash value and the product information of the target agricultural product into a block;

[0081] Step S303: Broadcast the block to the blockchain network;

[0082] Step S304: Nodes in the blockchain network verify the block;

[0083] Step S305: After successful verification, add the block to the blockchain.

[0084] In this embodiment of the invention, a hash value is generated by hashing the sub-image sequence carrying the behavior identifier. Hash processing is a one-way encryption algorithm that converts data of arbitrary length into a fixed-length hash value. The hash value can be used to verify data integrity. If the sub-image sequence is tampered with, its hash value will change, thus detecting data anomalies. The hash value compresses large amounts of data into a fixed-length value, facilitating storage and transmission. The generated hash value is packaged with the product information of the target agricultural product (such as origin, planting time, harvesting time, etc.) into a block. Packaging the hash value with the product information ensures the correlation between the behavior identifier and the product information, facilitating subsequent traceability. The chain structure of the blockchain ensures that once data is recorded, it cannot be tampered with, ensuring the authenticity and credibility of traceability information. The packaged block is broadcast to all nodes in the blockchain network. The blockchain network is a distributed network where each node holds a copy of the entire blockchain. The broadcast operation sends the new block to all nodes in the network for verification and synchronization. The decentralized nature of the blockchain network means that data storage and verification do not depend on a single central node, improving the reliability and security of the system. Broadcast operations ensure that all nodes can obtain the latest data in a timely manner, maintaining the consistency of the blockchain. Nodes in the blockchain network verify the received blocks, including verifying the correctness of the hash value, the integrity of the data, and whether it conforms to the blockchain rules. The verification process ensures that only legitimate blocks can be added to the blockchain. Nodes check whether the block's hash value is correct, whether the data is complete, and whether it follows the blockchain's consensus mechanism. Through node verification, it is ensured that only legitimate data can be recorded, preventing malicious tampering and forgery. The verification process is based on the blockchain's consensus mechanism, ensuring that all nodes in the network reach a consensus and maintaining the blockchain's credibility. After successful verification, the block is added to the blockchain, forming a new block. Once a block is verified, it will be permanently recorded in the blockchain, forming an immutable record. The immutability of the blockchain ensures that traceability information cannot be modified once recorded, guaranteeing the authenticity and credibility of the traceability information. The chain structure of the blockchain allows traceability information to be traced to every link, facilitating the tracking of the entire life cycle of agricultural products.

[0085] It should be noted that, in the embodiments of the present invention, if the above-described traceability method is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiments of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), magnetic disks, or optical disks. Thus, the embodiments of the present invention are not limited to any specific hardware and software combination.

[0086] Accordingly, embodiments of the present invention provide a storage medium storing a computer program thereon, characterized in that the computer program, when executed by a processor, implements the steps in the tracing method provided in the above embodiments.

[0087] This invention provides an electronic device; Figure 2 This is a schematic diagram of the composition structure of an electronic device provided in an embodiment of the present invention, such as... Figure 2 As shown, the electronic device 400 includes: a processor 401, at least one communication bus 402, a user interface 403, at least one external communication interface 404, and a memory 405. The communication bus 402 is configured to enable communication between these components. The user interface 403 may include a display screen, and the external communication interface 404 may include standard wired and wireless interfaces. The processor 401 is configured to execute a program of a tracing method stored in the memory to implement the steps of the tracing method provided in the above embodiment.

[0088] It should be noted that the descriptions of the storage medium and electronic device embodiments above are similar to the descriptions of the method embodiments above, and have similar beneficial effects. For technical details not disclosed in the storage medium and device embodiments of the present invention, please refer to the descriptions of the method embodiments of the present invention for understanding.

[0089] It should be understood that the phrase "one embodiment" or "an embodiment" throughout the specification means that a specific feature, structure, or characteristic related to the embodiment is included in at least one embodiment of the invention. Therefore, "in one embodiment" or "in an embodiment" appearing throughout the specification does not necessarily refer to the same embodiment. Furthermore, these specific features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. It should be understood that in the various embodiments of the invention, the sequence numbers of the above-described processes do not imply a sequential order of execution; the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the invention. The sequence numbers of the above-described embodiments of the invention are merely descriptive and do not represent the superiority or inferiority of the embodiments.

[0090] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, object, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, object, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, object, or apparatus that includes that element.

[0091] In the several embodiments provided by this invention, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0092] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units. They may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.

[0093] In addition, in the various embodiments of the present invention, each functional unit can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0094] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media that can store program code, such as mobile storage devices, read-only memory (ROM), magnetic disks, or optical disks.

[0095] Alternatively, if the integrated units of this invention are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of this invention, or the parts that contribute to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a controller to execute all or part of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROMs, magnetic disks, or optical disks.

[0096] The above description is merely an embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method of tracing an agricultural product, characterized by, The application is applied to an agricultural product traceability platform, comprising: Based on the agricultural product traceability platform, an image sequence corresponding to a target agricultural product is obtained in a preset period; The agricultural product traceability platform identifies whether a target human body exists in the image sequence based on a pre-constructed human body recognition model; In the case that all images of the image sequence exist the target human body, the agricultural product traceability platform inputs the image sequence into a pre-constructed behavior recognition model, and outputs a sub-image sequence carrying a behavior identifier, wherein one behavior identifier corresponds to one behavior category; Based on the sub-image sequence, traceability information is generated.

2. The method of claim 1, wherein, The behavior category includes planting, pesticide spraying, fertilization, picking and weeding.

3. The method of claim 1, wherein, Comprising: Statistical behavior categories in the traceability information at a preset time node; In the case that the behavior category does not match the preset behavior category, the agricultural product traceability platform feeds back first prompt information to the corresponding user terminal.

4. The method of claim 1, wherein, Comprising: Statistical behavior times corresponding to the behavior category in the traceability information at a preset time node; In the case that the behavior times corresponding to the behavior category are less than the preset times, the agricultural product traceability platform feeds back second prompt information to the corresponding user terminal.

5. The method of claim 1, wherein, The human body recognition model comprises: A first input layer for receiving a single frame image in the image sequence; A convolution layer comprising a plurality of convolution kernels for extracting features of the single frame image and outputting a feature map; A pooling layer for reducing the spatial dimension of the feature map; A first full connection layer for converting the feature map with reduced spatial dimension into a one-dimensional vector; A first output layer for outputting a binary classification result of whether a target human body exists based on the one-dimensional vector.

6. The method of claim 1, wherein, The behavior recognition model comprises: A second input layer receiving the image sequence; A spatio-temporal convolution layer for extracting spatio-temporal features in the image sequence; A long short-term memory network layer for processing the time dependence of the image sequence; A second full connection layer for converting the feature vector into a probability distribution of the behavior category; A second output layer for outputting a sub-image sequence carrying a behavior identifier.

7. The method of claim 1, wherein, Comprising: In the case that the behavior identifier carried by the sub-image sequence corresponding to the image sequence obtained in the current preset period is the same as the behavior identifier carried by the sub-image sequence corresponding to the image sequence obtained in the previous preset period, it is not necessary to generate new traceability information based on the sub-image sequence carrying the behavior identifier corresponding to the current preset period.

8. The method of claim 1, wherein, Comprising: Hash processing is performed on the sub-image sequence carrying the behavior identifier to generate a hash value; The hash value and product information of the target agricultural product are packaged into a block; The block is broadcast to a blockchain network; The nodes in the blockchain network verify the block; After verification, the block is added to the blockchain.

Images