Intelligent agricultural product full-life-cycle traceability system based on block chain and Internet of Things

Abstract

The invention discloses an intelligent agricultural product full-life-cycle traceability system based on a block chain and the Internet of Things, and belongs to the technical field of agricultural product traceability. According to the invention, the entity fence, the camera and the environment sensor are deployed in the agricultural product planting area, each key link is monitored in real time, and the multi-modal data is collected. And a multi-party participated collaborative traceability system is established, and full-process collaborative supervision is realized. And selecting an alliance chain architecture according to the circulation area, encrypting the data, storing the data in a distributed storage system, and writing key information. And adopting a swarm intelligent consensus algorithm to automatically evaluate multiple dimensions of the traceability data, and achieving a swarm consensus through iterative optimization to generate a consensus result. And generating a unique traceability code, and scanning the code to display the traceability information of the whole life cycle in a grading manner. According to the invention, the authenticity, security and traceability of traceability data can be significantly improved, the problems of low efficiency, easy tampering, insufficient credibility and the like of a traditional traceability system are solved, and the system has great innovativeness and popularization value.

Description

Technical Field

[0001] This invention belongs to the field of agricultural product traceability technology, and in particular relates to an intelligent agricultural product full life cycle traceability system based on blockchain and the Internet of Things. Background Technology

[0002] Agricultural product traceability systems are quality management systems that achieve full-process tracking of agricultural products by recording information from production, processing, and distribution. They utilize technologies such as the Internet of Things (IoT), blockchain, and QR codes to construct digital product archives. This system generates a unique traceability code for each batch of products via barcodes or QR codes, allowing consumers to scan the code to access data such as the place of origin, production date, and testing reports. Traditional centralized traceability systems suffer from single-point-of-failure risks, and data reliability is difficult to guarantee. Although blockchain technology has been introduced into the traceability field due to its decentralized and tamper-proof characteristics, ensuring the authenticity and reliability of source data before it is uploaded to the blockchain, processing and evaluating complex multimodal traceability data (such as environmental data and video footage), and forming authoritative and credible consensus conclusions remain challenges that current technologies have not yet adequately addressed.

[0003] Therefore, there is an urgent need for a new technical solution to address this problem. Summary of the Invention

[0004] The technical problem to be solved by this invention is to provide an intelligent agricultural product full life cycle traceability system based on blockchain and the Internet of Things, which solves the problem that existing traceability systems are unable to achieve full-process data security, high-precision authentication of product identity, and multi-party collaborative supervision.

[0005] The technical solution adopted in this invention is to provide an intelligent agricultural product full life cycle traceability system based on blockchain and the Internet of Things, comprising:

[0006] The electronic fence and IoT device management module is used to deploy physical fences, distributed camera arrays, and various types of environmental sensors in agricultural product planting areas to monitor and collect multimodal data in real time at key stages of agricultural products from sowing, growth, harvesting, processing, packaging to transportation.

[0007] A multi-party traceability and collaboration system is used to integrate growers, third-party testing institutions and government regulatory departments as nodes in the consortium blockchain to achieve multi-entity collaborative supervision and data verification throughout the entire process;

[0008] The consortium blockchain data storage module is used to select the consortium blockchain architecture based on the commodity circulation area, encrypt the data collected by IoT devices and store it in a distributed storage system, and write the corresponding data summary, content identifier, signature and timestamp on the consortium blockchain.

[0009] The intelligent consensus module is used to automatically evaluate multiple dimensions of the source data using a group intelligent consensus algorithm based on a virtual expert group, and achieve group consensus through iterative optimization to generate consensus results;

[0010] The traceability code generation and query module is used to generate a unique traceability code for each batch or individual agricultural product, and to display traceability information throughout the entire life cycle through this code in a hierarchical manner.

[0011] The electronic fence and the distributed camera array in the IoT device management module are mutually visible to each other, which is used to realize real-time alarm and recording of people and animals in the area; the environmental sensors include at least temperature and humidity sensors, CO2 concentration sensors, light intensity sensors, soil moisture sensors and soil nutrient sensors.

[0012] When the intelligent consensus module executes consensus:

[0013] The system randomly generates N virtual experts to evaluate the M dimensions of the traceability data, forming an initial trust matrix T;

[0014] Based on the preset initial opinion matrix O, calculate the consensus degree between any two experts i and j. ;

[0015] Based on the consensus Calculate the group consensus level GCL using the initial trust matrix T;

[0016] The DBSCAN clustering algorithm is used to cluster expert opinions and identify outlier experts.

[0017] By iteratively adjusting the opinion matrix O to O', the group consensus level GCL is brought to or exceeds the preset threshold τ while minimizing the adjustment cost.

[0018] Based on the adjusted opinion matrix O', the final consensus score for each dimension is calculated.

[0019] The consensus The calculation formula is:

[0020]

[0021] In the formula, O is the initial opinion matrix with a size of (N×M), O' is the adjusted opinion matrix with a size of (N×M), M is the number of dimensions of the source data to be evaluated, and N is the number of virtual experts randomly generated by the system. Let i be the adjusted opinion value of the i-th expert in the k-th dimension. Let j be the adjusted opinion value of the j-th expert in the k-th dimension. The maximum value among all expert opinions across all dimensions. It is the minimum value among all expert opinions across all dimensions.

[0022] The formula for calculating the Group Consensus Level (GCL) is as follows:

[0023]

[0024] In the formula, O' is the adjusted opinion matrix, and N is the number of virtual experts randomly generated by the system. Let be the trust level of the i-th expert over the j-th expert in the initial trust matrix T. The initial trust matrix T has a size of (N×N). Let represent the consensus degree between the i-th expert and the j-th expert.

[0025] The traceability code generation and query module implements a hierarchical display that shows different levels of traceability information based on the value of the product or user permissions. The levels include at least: basic origin information, environmental data and planting process, detailed traceability data throughout the entire process, and in-depth information including DNA identification reports and expert evaluation reports.

[0026] Another technical solution adopted by this invention is to provide a smart agricultural product full life cycle traceability method based on blockchain and the Internet of Things, including the following steps:

[0027] By deploying electronic fences, distributed camera arrays, and various types of environmental sensors in agricultural product planting areas, real-time monitoring and multimodal data collection are carried out on key links of agricultural products from sowing, growth, harvesting, processing, packaging to transportation.

[0028] Establish a multi-party collaborative traceability system covering growers, third-party testing institutions and government regulatory departments, with each party participating as a node in the consortium blockchain to collaboratively supervise and verify data throughout the entire process;

[0029] The consortium blockchain architecture is selected based on the commodity circulation area. The collected data is encrypted and stored in a distributed storage system. The corresponding data digest, content identifier, signature and timestamp are written on the consortium blockchain.

[0030] A swarm intelligence consensus algorithm based on virtual expert groups is adopted to automatically evaluate multiple dimensions of the source data and reach a consensus through iterative optimization to generate consensus results.

[0031] A unique traceability code is generated for each batch or individual agricultural product, and the traceability information throughout the entire life cycle is displayed through this code in a hierarchical manner.

[0032] The steps of evaluating the source tracing data and reaching a consensus using a swarm intelligence consensus algorithm include:

[0033] The system randomly generates N virtual experts to evaluate the M dimensions of the traceability data, forming an initial trust matrix T;

[0034] Based on the preset initial opinion matrix O, calculate the consensus degree between any two experts i and j. ;

[0035] Based on the consensus Calculate the group consensus level GCL using the initial trust matrix T;

[0036] The DBSCAN clustering algorithm is used to cluster expert opinions and identify outlier experts.

[0037] By iteratively adjusting the opinion matrix O to O', the group consensus level GCL is brought to or exceeds the preset threshold τ while minimizing the adjustment cost.

[0038] Based on the adjusted opinion matrix O', the final consensus score for each dimension is calculated.

[0039] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of a smart agricultural product lifecycle traceability method based on blockchain and the Internet of Things.

[0040] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of a smart agricultural product lifecycle traceability method based on blockchain and the Internet of Things.

[0041] Through the above design scheme, the present invention can bring the following beneficial effects:

[0042] 1. By combining IoT devices with electronic fences, the real and comprehensive collection of physical world data is ensured, guaranteeing data reliability from the source.

[0043] 2. The introduction of an intelligent consensus mechanism based on a virtual expert group enables automated and intelligent evaluation of complex multimodal data, solving the problem of determining the authenticity of data in traditional traceability and enhancing the authority and credibility of the conclusions.

[0044] 3. By utilizing consortium blockchain and distributed storage technology, the immutability, traceability, and efficient storage of traceability data are ensured, achieving data transparency and security.

[0045] 4. By using a unique traceability code and a tiered information display mechanism, it satisfies consumers' right to know about product information while protecting core business data, thus achieving a balance between information transparency and data security.

[0046] 5. The technical framework of this system can also be extended to the traceability and quality supervision of other high-value Chinese medicinal materials and food, and has broad application prospects. Attached Figure Description

[0047] Figure 1 This is a schematic diagram of the traceability process of an intelligent agricultural product full life cycle traceability system based on blockchain and the Internet of Things according to the present invention;

[0048] Figure 2 This is a schematic diagram of the consensus mechanism of a smart agricultural product lifecycle traceability system based on blockchain and the Internet of Things according to the present invention. Detailed Implementation

[0049] The present invention will be further described below with reference to the accompanying drawings and specific embodiments:

[0050] This invention provides a smart agricultural product lifecycle traceability system based on blockchain and the Internet of Things, taking ginseng as an example, such as... Figure 1 As shown, its workflow includes the following steps:

[0051] I. Electronic Fence and IoT Device Management: This involves real-time monitoring and data collection of key aspects of ginseng growth, harvesting, and processing within a designated area via IoT terminals. Electronic fence information and environmental sensor data are then uploaded to the blockchain. This includes environmental parameters, fence status, and video footage, providing fundamental data support for subsequent traceability data processing and storage.

[0052] Specifically, the system deployed a complete IoT monitoring network at the planting base. Environmental sensors were arranged at a density of n acres per node, with each node integrating soil and air temperature and humidity meters, carbon dioxide concentration sensors, light meters, and pH level detection equipment. The electronic fence employed a three-layer protection system consisting of high-definition cameras, infrared sensors, and acoustic alarms. Harvesting personnel wore safety helmets equipped with cameras to record the entire operation process, and the data was encrypted and uploaded in real time. Environmental sensor data for ginseng cultivation is shown in Table 1.

[0053] Table 1. Ginseng Cultivation Environment Sensor Data Recording Table

[0054]

[0055] Second, a multi-party product traceability system covering growers, third-party testing agencies, and government regulatory departments.

[0056] Specifically, a multi-party participation model is adopted to construct the traceability system, with each participant undertaking different responsibilities and holding different permissions within the system. Growers are responsible for providing original data such as planting environment and harvesting details, and receiving training to standardize operating procedures; third-party testing institutions are responsible for quality testing of ginseng samples, supervising the harvesting and processing process, and issuing authoritative supervisory reports; government regulatory departments are responsible for policy formulation, data auditing, and quality supervision. The system integrates these three roles into different nodes in a consortium blockchain, achieving multi-party verification of data through a distributed consensus mechanism to ensure the authenticity and reliability of information.

[0057] Third, select a consortium blockchain architecture based on the commodity circulation area, encrypt the large amount of growth, operation and testing data and store it in IPFS (distributed storage), and write data digests, CID (content identifier), signatures and timestamps on the consortium blockchain to achieve verifiable on-chain evidence storage and multi-party consensus.

[0058] Specifically: The system constructs a consortium blockchain network, deploying dedicated channels for different sales regions. The traceability data in the system includes, but is not limited to: planting environment data (soil, climate parameters), growth process records (fertilization, irrigation records), harvesting process videos, processing records (washing, drying, etc.), inspection and testing reports, and logistics information. All data is strictly formatted and stored on the blockchain, and cannot be modified by any single party. Each piece of data is written to the consortium blockchain with a summary, CID, uploader and testing agency signatures, and a timestamp to achieve verifiable on-chain evidence storage and multi-party consensus.

[0059] IV. Data encryption storage and consensus mechanism for key links: The testing agency terminal is responsible for data verification and supervision report generation to ensure data authenticity and immutability.

[0060] Specifically, the system employs multi-layered encryption and a smart consensus mechanism based on a GPT-4.1 virtual expert ensemble to ensure data security and reliability. The consensus mechanism is based on the following mathematical model: the system randomly generates N virtual experts who evaluate the source data across M dimensions. Initial trust matrix. The initial opinion matrix represents the degree of trust that expert i has in expert j. This represents the score (1-5 points) given by expert i to dimension k.

[0061] Consensus between two experts i and j Defined as:

[0062]

[0063] In the formula, O is the initial opinion matrix with a size of (N×M), O' is the adjusted opinion matrix with a size of (N×M), M is the number of dimensions of the source data to be evaluated, and N is the number of virtual experts randomly generated by the system. Let i be the adjusted opinion value of the i-th expert in the k-th dimension. Let j be the adjusted opinion value of the j-th expert in the k-th dimension. The maximum value among all expert opinions across all dimensions. It is the minimum value among all expert opinions across all dimensions.

[0064] The formula for calculating the Group Consensus Level (GCL) is:

[0065]

[0066] In the formula, O' is the adjusted opinion matrix, and N is the number of virtual experts randomly generated by the system. Let be the trust level of the i-th expert over the j-th expert in the initial trust matrix T. The initial trust matrix T has a size of (N×N). Let represent the consensus degree between the i-th expert and the j-th expert.

[0067] The system clusters expert opinions using the DBSCAN algorithm and iteratively adjusts the opinion matrix from O to O' to minimize adjustment costs while simultaneously achieving a preset threshold for group consensus. .

[0068] After the consensus assessment results are confirmed by the testing agency, they are encrypted and uploaded to the blockchain along with the original testing report to form the final authoritative traceability report, thus completing the entire traceability loop.

[0069] V. Traceability Code Generation and Scanning Inquiry: A unique traceability code is generated for each batch or single ginseng plant. After scanning the code, real traceability data including growth environment, operation process, brand certification, etc. are displayed. Traceability information is displayed in a hierarchical manner, showing multi-dimensional and visual traceability content such as sowing, environmental parameters, disease detection, harvesting video, processing and packaging.

[0070] Specifically, the system generates a unique traceability code for each batch or individual ginseng. The traceability code generation process includes: collecting batch information and obtaining the blockchain data address. After the consumer scans the code, the system verifies the signature validity and data integrity, displaying different levels of information based on user type, such as origin information, environmental data, harvesting videos, and quality inspection reports. The system implements a five-level hierarchical display architecture based on product value and user permissions. For product grading: ordinary products provide basic traceability information; intermediate products add environmental data and planting process displays; advanced products provide detailed traceability data throughout the entire process; and premium products additionally provide DNA identification reports and expert evaluation reports.

[0071] Example:

[0072] Settings: N = 4 experts, M = 3 evaluation dimensions.

[0073] Dimensions: k=1 Environmental compliance, k=2 Process consistency, k=3 Quality indication.

[0074] Dimension weights: w = (0.3, 0.4, 0.3).

[0075] Initial opinion matrix (unit: points):

[0076]

[0077] Initial trust matrix (row i indicates "who i trusts"):

[0078]

[0079] 1. Calculate pairwise consensus.

[0080] by For example:

[0081] Weighted difference:

[0082]

[0083] Similarly, we can obtain (rounded to four decimal places):

[0084]

[0085] 2. Mutual Trust Pairing Rights Group consensus level (GCL)

[0086] Mutual trust pairing weights are taken as a symmetric average:

[0087]

[0088] Group consensus level is weighted Dual similarity average:

[0089]

[0090] By multiplying each term and summing them, we can obtain:

[0091]

[0092] Less than the preset threshold The adjustment process needs to be initiated.

[0093] 3. The system applies DBSCAN clustering to the expert opinion matrix O to identify opinion factions and outlier experts.

[0094] The results showed that Expert 3's score differed significantly from the other experts and was identified as an "outlier".

[0095] 4. Adjust the opinion matrix O→O′

[0096] To improve group consensus, the system makes gentle adjustments to the opinions of outlier experts, resulting in O′.

[0097] The optimization objective is to minimize the adjustment cost while ensuring that the level of group consensus does not fall below the threshold.

[0098] Mathematical form:

[0099]

[0100] Assuming an increase of 0.5 points in each of the three dimensions of Expert 3, we get:

[0101]

[0102] Recalculate now:

[0103]

[0104] The threshold has been reached.

[0105] 5. Reaching a final consensus.

[0106] Based on the adjusted opinion matrix O′, calculate the final scores for each dimension:

[0107]

[0108] The result is:

[0109]

[0110] 6. Testing agency confirmation and blockchain integration

[0111] The testing agency confirms the consensus results and uploads them, along with the original testing report, to the blockchain in an encrypted manner to generate the final authoritative traceability report, thus completing the traceability loop.

[0112] This invention proposes a smart agricultural product lifecycle traceability system based on blockchain and the Internet of Things. By deploying electronic fences, distributed cameras, and various environmental sensors in the planting area, real-time monitoring and multimodal data collection of the entire process, including sowing, growth, harvesting, processing, packaging, and transportation, are achieved, solving the problems of fragmented and difficult-to-manage traditional traceability information. The system introduces a virtual expert group intelligent consensus algorithm to automatically evaluate multimodal data and combines DBSCAN clustering optimization and graph neural network modeling to effectively improve the fairness and consistency of the consensus results. All consensus results and test reports are encrypted and stored on the consortium blockchain to ensure data authenticity, integrity, and immutability. This system can also be extended to the intelligent traceability and quality testing of other Chinese medicinal materials and agricultural products, providing a replicable technological paradigm for agricultural digitalization and the intelligent upgrading of the traditional Chinese medicine industry, possessing significant technological innovation value and broad industrial application prospects.

[0113] The implementation of the present invention is not limited to the above embodiments. Any changes, modifications, substitutions, combinations, or simplifications made without departing from the spirit and principle of the present invention shall be considered equivalent substitutions and are included within the protection scope of the present invention.

Claims

1. A smart agricultural product lifecycle traceability system based on blockchain and the Internet of Things, characterized by: include: The electronic fence and IoT device management module is used to deploy physical fences, distributed camera arrays, and various types of environmental sensors in agricultural product planting areas to monitor and collect multimodal data in real time at key stages of agricultural products from sowing, growth, harvesting, processing, packaging to transportation. A multi-party traceability and collaboration system is used to integrate growers, third-party testing institutions and government regulatory departments as nodes in the consortium blockchain to achieve multi-entity collaborative supervision and data verification throughout the entire process; The consortium blockchain data storage module is used to select the consortium blockchain architecture based on the commodity circulation area, encrypt the data collected by IoT devices and store it in a distributed storage system, and write the corresponding data summary, content identifier, signature and timestamp on the consortium blockchain. The intelligent consensus module is used to automatically evaluate multiple dimensions of the source data using a group intelligent consensus algorithm based on a virtual expert group, and achieve group consensus through iterative optimization to generate consensus results; The traceability code generation and query module is used to generate a unique traceability code for each batch or individual agricultural product, and to display traceability information throughout the entire life cycle through this code in a hierarchical manner.

2. The intelligent agricultural product full life cycle traceability system based on blockchain and the Internet of Things as described in claim 1, characterized in that: The electronic fence and the distributed camera array in the IoT device management module are mutually visible to each other, which is used to realize real-time alarm and recording of people and animals in the area; the environmental sensors include at least temperature and humidity sensors, CO2 concentration sensors, light intensity sensors, soil moisture sensors and soil nutrient sensors.

3. The intelligent agricultural product lifecycle traceability system based on blockchain and the Internet of Things as described in claim 1, characterized in that: When the intelligent consensus module executes consensus: The system randomly generates N virtual experts to evaluate the M dimensions of the traceability data, forming an initial trust matrix T; Based on the preset initial opinion matrix O, calculate the consensus degree between any two experts i and j. ; Based on the consensus Calculate the group consensus level GCL using the initial trust matrix T; The DBSCAN clustering algorithm is used to cluster expert opinions and identify outlier experts. By iteratively adjusting the opinion matrix O to O', the group consensus level GCL is brought to or exceeds the preset threshold τ while minimizing the adjustment cost. Based on the adjusted opinion matrix O', the final consensus score for each dimension is calculated.

4. The intelligent agricultural product full life cycle traceability system based on blockchain and the Internet of Things as described in claim 3, characterized in that: The consensus The calculation formula is: In the formula, O is the initial opinion matrix with a size of (N×M), O' is the adjusted opinion matrix with a size of (N×M), M is the number of dimensions of the source data to be evaluated, and N is the number of virtual experts randomly generated by the system. Let i be the adjusted opinion value of the i-th expert in the k-th dimension. Let j be the adjusted opinion value of the j-th expert in the k-th dimension. The maximum value among all expert opinions across all dimensions. It is the minimum value among all expert opinions across all dimensions.

5. The intelligent agricultural product lifecycle traceability system based on blockchain and the Internet of Things as described in claim 3, characterized in that: The formula for calculating the Group Consensus Level (GCL) is as follows: In the formula, O' is the adjusted opinion matrix, and N is the number of virtual experts randomly generated by the system. Let be the trust level of the i-th expert over the j-th expert in the initial trust matrix T. The initial trust matrix T has a size of (N×N). Let represent the consensus degree between the i-th expert and the j-th expert.

6. The intelligent agricultural product full life cycle traceability system based on blockchain and the Internet of Things as described in claim 1, characterized in that: The traceability code generation and query module implements a hierarchical display that shows different levels of traceability information based on the value of the product or user permissions. The levels include at least: basic origin information, environmental data and planting process, detailed traceability data throughout the entire process, and in-depth information including DNA identification reports and expert evaluation reports.

7. A smart agricultural product lifecycle traceability method based on blockchain and the Internet of Things, using the system as described in any one of claims 1-6, characterized in that: Includes the following steps: By deploying electronic fences, distributed camera arrays, and various types of environmental sensors in agricultural product planting areas, real-time monitoring and multimodal data collection are carried out on key links of agricultural products from sowing, growth, harvesting, processing, packaging to transportation. Establish a multi-party collaborative traceability system covering growers, third-party testing institutions and government regulatory departments, with each party participating as a node in the consortium blockchain to collaboratively supervise and verify data throughout the entire process; The consortium blockchain architecture is selected based on the commodity circulation area. The collected data is encrypted and stored in a distributed storage system. The corresponding data digest, content identifier, signature and timestamp are written on the consortium blockchain. A swarm intelligence consensus algorithm based on virtual expert groups is adopted to automatically evaluate multiple dimensions of the traceability data and reach a consensus through iterative optimization to generate consensus results. A unique traceability code is generated for each batch or individual agricultural product, and the traceability information throughout the entire life cycle is displayed through this code in a hierarchical manner.

8. The intelligent agricultural product lifecycle traceability method based on blockchain and the Internet of Things as described in claim 7, characterized in that: The steps of evaluating the source tracing data and reaching a consensus using a swarm intelligence consensus algorithm include: The system randomly generates N virtual experts to evaluate the M dimensions of the traceability data, forming an initial trust matrix T; Based on the preset initial opinion matrix O, calculate the consensus degree between any two experts i and j. ; Based on the consensus Calculate the group consensus level GCL using the initial trust matrix T; The DBSCAN clustering algorithm is used to cluster expert opinions and identify outlier experts. By iteratively adjusting the opinion matrix O to O', the group consensus level GCL is brought to or exceeds the preset threshold τ while minimizing the adjustment cost. Based on the adjusted opinion matrix O', the final consensus score for each dimension is calculated.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, it implements the steps of the method as described in any one of claims 7 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by the processor, it implements the steps of the method as described in any one of claims 7 to 8.

Images