Agricultural product supply chain transportation management system and method

By evaluating freshness in real time and adjusting transportation strategies dynamically in the agricultural product supply chain transportation management system, the problems that are difficult to monitor and adjust in the existing technology are solved, and accurate control and transportation optimization of agricultural product freshness are achieved, reducing loss risk and improving Economic benefits and market competitiveness.

CN118941185AInactive Publication Date: 2024-11-12SHENGMAO DATA TECH DEV CO LTD

Patent Information

Application Number
CN202411420572.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-12
Publication Date
2024-11-12
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing technology is difficult to monitor changes in freshness of agricultural products in real time, and it is impossible to dynamically adjust transportation strategies based on freshness changes, resulting in excessive loss of agricultural products during transportation, affecting quality and economic benefits.

Method used

A agricultural product supply chain transportation management system was designed, including image acquisition module, parameter control module and data processing module. Through image acquisition and preprocessing, the freshness is evaluated in real time using the Transformer neural network model, and the arrival time is predicted according to the fitting function. Freshness, dynamically adjust the refrigeration power and transportation path, and change the destination in time to maintain high freshness.

Benefits of technology

It has achieved accurate monitoring and regulation of the freshness of agricultural products, optimized transportation routes and time, reduced losses, and improved economic benefits and market competitiveness.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118941185A_ABST
    Figure CN118941185A_ABST
Patent Text Reader

Abstract

The invention discloses an agricultural product supply chain transportation management system and method, and belongs to the technical field of agricultural product transportation, and the method specifically comprises the steps: collecting the images of agricultural products in an agricultural product transport vehicle at intervals, and outputting the freshness of the agricultural products of each destination at the current moment through a preset freshness recognition model; the refrigerating power of a compartment refrigerating device is adjusted according to the freshness change speed in unit time and the compartment temperature; generating a time-varying fitting function of the freshness of each destination agricultural product; predicting the expected freshness of the agricultural products at any destination when the agricultural products arrive, if the expected freshness is lower than the standard, obtaining the expected moment when the freshness of the agricultural products is equal to the freshness standard, selecting the destinations before the expected moment, marking the destinations as the preferred destinations, and replacing the original destinations of the batch of agricultural products with the preferred destinations; according to the invention, monitoring and prediction of the freshness of the agricultural products are realized, and the transportation plan is adjusted based on the monitoring and prediction.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of agricultural product transportation, and in particular to an agricultural product supply chain transportation management system and method. Background Art

[0002] In the supply chain management of agricultural products, it is crucial to maintain the freshness of agricultural products. However, due to environmental factors such as temperature, humidity, light, and physiological and biochemical reactions of agricultural products during transportation, the freshness of agricultural products will gradually decrease over time. Therefore, how to effectively monitor and manage the freshness of agricultural products has become an important issue in the supply chain management of agricultural products.

[0003] Traditional methods of managing the freshness of agricultural products mainly rely on manual observation and empirical judgment, which is not only inefficient but also difficult to guarantee accuracy. In recent years, although some studies have attempted to evaluate the freshness of agricultural products through image processing technology, these methods can often only provide static freshness evaluation and cannot monitor the changes in the freshness of agricultural products in real time.

[0004] In addition, the existing agricultural product supply chain management system is unable to dynamically adjust the transportation strategy according to the changes in the freshness of agricultural products to optimize the transportation path and time of agricultural products. This may lead to excessive loss of agricultural products during transportation, affecting the quality and economic benefits of agricultural products. Summary of the invention

[0005] The purpose of the present invention is to provide an agricultural product supply chain transportation management system and method to solve the following technical problems: Existing methods make it difficult to monitor changes in the freshness of agricultural products in real time, and cannot effectively adjust transportation strategies based on changes in freshness.

[0006] The purpose of the present invention can be achieved through the following technical solutions: An agricultural product supply chain transportation management system, comprising: An image acquisition module is used to collect images of agricultural products in the agricultural product transport vehicle at intervals, pre-process the agricultural product images, and obtain agricultural product images corresponding to each destination, respectively input the agricultural product images into a preset freshness recognition model, and output the freshness of the agricultural products at each destination at the current moment; A parameter control module is used to calculate the freshness change rate per unit time. When the freshness change rate is greater than a preset threshold, the cabin temperature change value per unit time is obtained. If the cabin temperature change value is greater than the preset threshold, the cooling power of the cabin refrigeration device is increased; The data processing module is used to fit the collected agricultural product freshness data in time series to obtain a fitting function of the change of agricultural product freshness over time at each destination; The risk judgment module is used to predict the expected freshness of agricultural products when they arrive at any destination based on the fitting function. If the expected freshness is lower than the freshness standard of the current variety of agricultural products, the expected moment when the freshness of the agricultural products at the destination is equal to the freshness standard is obtained based on the fitting function, and the previous destination before the expected moment is selected and marked as the preferred destination, and the original destination of the batch of agricultural products is replaced with the preferred destination.

[0007] As a further solution of the present invention: in the image acquisition module, the preprocessing process is: The edge detection based on low-pass filtering is used to filter out the image edge details of the agricultural product image, and the image is gray-processed, and the gray-scale curve of the image is adjusted to a preset dynamic range. Based on the fusion of BP neural network classification algorithm and decision tree splitting algorithm, the agricultural product image is divided into sub-images representing different appearance features, and the appearance features include gray-scale range, point density and texture.

[0008] As a further solution of the present invention: in the image acquisition module, the process of outputting freshness is: Establish an image library, in which standard agricultural product images with freshness annotated are stored, the standard agricultural product images are preprocessed and token prompts are added, the processed standard sub-images are converted into vectors, and the vectors are input into the Transformer neural network model after position embedding encoding, and the appearance features in different standard sub-images are respectively extracted, and the Transformer neural network model is updated based on the existing freshness annotations by back propagation, and the existing freshness annotations are repeatedly matched with the appearance features until the loss function of the model converges; The agricultural product images collected in real time are also preprocessed and then token prompts are added. The processed real-time sub-images are encoded by position embedding and input into the matched Transformer neural network model. The appearance features in different real-time sub-images are extracted respectively, and the initial freshness labels are generated by clustering. The Transformer neural network model is updated by backpropagation based on the initial freshness labels, and the freshness labels are repeatedly updated until the model loss function converges to obtain the final freshness.

[0009] As a further solution of the present invention: in the data processing module, the fitting function is based on the power exponential function, and the formula is: Fresh = Cexp(a(T+T') b ); Among them, Fresh represents freshness, C is a constant, exp represents a function with base e, T represents the transportation time of the transport vehicle, T' represents the storage time of the agricultural products before the transport vehicle departs, a and b represent function parameters, and the freshness and time data are substituted into the function to obtain the values ​​of the function parameters.

[0010] As a further solution of the present invention: the process of generating the standard agricultural product image is to select agricultural product samples, place the agricultural product samples in an environment with the same lighting, humidity and temperature as the transport vehicle, collect images of the agricultural product samples every unit time, which are standard agricultural product images, and annotate the standard agricultural product images with the length of time the products have been placed when they were taken, which is the freshness.

[0011] As a further solution of the present invention: in the parameter control module, the adjustment process of the refrigeration power is: The heat increase rate P in the compartment is calculated based on the compartment temperature change value d, and the formula is: ; Where β is the preset correction coefficient, β is greater than 1, c is the specific heat capacity of air, ρ is the density of air, V is the volume of the car, and t represents unit time.

[0012] As a further solution of the present invention: in the risk judgment module, when the transport vehicle passes through the replaced original destination, it passes directly without stopping and prompts the staff to shorten the unloading time at each destination.

[0013] As a further solution of the present invention: all destinations before the expected time are selected and marked as pending destinations, and the unloading quantity of each pending destination is obtained, and the destination with the lowest unloading quantity is marked as the preferred destination.

[0014] A method for managing agricultural product supply chain transportation includes the following steps: Collect images of agricultural products in various areas of the agricultural product transport vehicle at intervals, pre-process the agricultural product images, and obtain agricultural product images corresponding to each destination, input the agricultural product images into a preset freshness recognition model, and output the freshness of the agricultural products at each destination at the current moment; Calculate the freshness change rate per unit time. When the freshness change rate is greater than a preset threshold, obtain the cabin temperature change value per unit time. If the cabin temperature change value is greater than the preset threshold, increase the cooling power of the cabin refrigeration device. Fit the collected agricultural product freshness data in time series to obtain a fitting function of the change of agricultural product freshness over time at each destination; The expected freshness of agricultural products upon arrival at any destination is predicted based on the fitting function. If the expected freshness is lower than the freshness standard of the current variety of agricultural products, the expected moment when the freshness of the agricultural products at the destination is equal to the freshness standard is obtained based on the fitting function, and the previous destination before the expected moment is selected and marked as the preferred destination, and the original destination of the batch of agricultural products is replaced with the preferred destination.

[0015] Beneficial effects of the present invention: The present invention provides an agricultural product supply chain transportation management system, which collects agricultural product images in various areas of an agricultural product transportation vehicle at intervals through an image acquisition module, and inputs the images into a preset freshness recognition model after preprocessing, so as to realize accurate evaluation of the freshness of agricultural products at each destination at the current moment; the parameter control module can automatically adjust the refrigeration power according to the freshness change speed and the change value of the compartment temperature, so as to effectively maintain the freshness of agricultural products; the data processing module predicts the expected freshness of agricultural products when arriving at any destination through a fitting function, and adjusts the transportation plan accordingly, so as to ensure that the agricultural products still maintain high freshness when arriving at the destination; the risk judgment module can adjust the destination in time when the expected freshness is lower than the standard, so as to avoid the reduction of freshness of agricultural products due to long transportation time, thereby reducing economic losses and improving customer satisfaction; in addition, the present invention improves the accuracy and robustness of the model by performing feature-based training on agricultural product images; in short, the present invention can monitor and regulate the freshness of agricultural products in real time, optimize transportation routes and time, reduce loss risks, and improve economic benefits and market competitiveness. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The present invention will be further described below in conjunction with the accompanying drawings.

[0017] Figure 1 It is a structural schematic diagram of an agricultural product supply chain transportation management system of the present invention; Figure 2 It is a flow chart of a method for agricultural product supply chain transportation management of the present invention. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] See also Figure 1 As shown, the present invention is an agricultural product supply chain transportation management system, comprising: An image acquisition module is used to collect images of agricultural products in the agricultural product transport vehicle at intervals, pre-process the agricultural product images, and obtain agricultural product images corresponding to each destination, respectively input the agricultural product images into a preset freshness recognition model, and output the freshness of the agricultural products at each destination at the current moment; A parameter control module is used to calculate the freshness change rate per unit time. When the freshness change rate is greater than a preset threshold, the cabin temperature change value per unit time is obtained. If the cabin temperature change value is greater than the preset threshold, the cooling power of the cabin refrigeration device is increased; The data processing module is used to fit the collected agricultural product freshness data in time series to obtain a fitting function of the change of agricultural product freshness over time at each destination; The risk judgment module is used to predict the expected freshness of agricultural products when they arrive at any destination based on the fitting function. If the expected freshness is lower than the freshness standard of the current variety of agricultural products, the expected moment when the freshness of the agricultural products at the destination is equal to the freshness standard is obtained based on the fitting function, and the previous destination before the expected moment is selected and marked as the preferred destination, and the original destination of the batch of agricultural products is replaced with the preferred destination.

[0020] The present invention can collect and analyze agricultural product images in real time, accurately assessing the freshness of agricultural products at each destination at the current moment. This real-time monitoring capability makes the quality control of agricultural products during transportation more accurate, and problems can be discovered and measures taken in a timely manner.

[0021] In another preferred embodiment of the present invention, in the image acquisition module, the preprocessing process is: The edge detection based on low-pass filtering is used to filter out the image edge details of the agricultural product image, and the image is gray-processed, and the gray-scale curve of the image is adjusted to a preset dynamic range. Based on the fusion of BP neural network classification algorithm and decision tree splitting algorithm, the agricultural product image is divided into sub-images representing different appearance features, and the appearance features include gray-scale range, point density and texture.

[0022] In another preferred embodiment of the present invention, in the image acquisition module, the process of outputting freshness is: Establish an image library, in which standard agricultural product images with freshness annotated are stored, the standard agricultural product images are preprocessed and token prompts are added, the processed standard sub-images are converted into vectors, and the vectors are input into the Transformer neural network model after position embedding encoding, and the appearance features in different standard sub-images are respectively extracted, and the Transformer neural network model is updated based on the existing freshness annotations by back propagation, and the existing freshness annotations are repeatedly matched with the appearance features until the loss function of the model converges; The agricultural product images collected in real time are also preprocessed and then token prompts are added. The processed real-time sub-images are encoded by position embedding and input into the matched Transformer neural network model. The appearance features in different real-time sub-images are extracted respectively, and the initial freshness labels are generated by clustering. The Transformer neural network model is updated by backpropagation based on the initial freshness labels, and the freshness labels are repeatedly updated until the model loss function converges to obtain the final freshness.

[0023] In another preferred embodiment of the present invention, in the data processing module, the fitting function is based on a power exponential function, and the formula is: Fresh = Cexp(a(T+T') b ); Among them, Fresh represents freshness, C is a constant, exp represents a function with base e, T represents the transportation time of the transport vehicle, T' represents the storage time of the agricultural products before the transport vehicle departs, a and b represent function parameters, and the freshness and time data are substituted into the function to obtain the values ​​of the function parameters.

[0024] The data processing module and the risk judgment module work together to predict the expected freshness of agricultural products when they arrive at the destination through fitting functions, and adjust the destination if necessary. This can effectively avoid losses caused by substandard freshness and improve the overall sales quality and economic benefits of agricultural products.

[0025] In another preferred embodiment of the present invention, the generation process of the standard agricultural product image is to select agricultural product samples, place the agricultural product samples in an environment with the same lighting, humidity and temperature as the transport vehicle, collect images of the agricultural product samples every unit time, which are standard agricultural product images, and annotate the standard agricultural product images with the length of time the products have been placed when they were taken, which is the freshness.

[0026] By generating standard images of produce under the same environmental conditions as the trucks that transport them, the accuracy and consistency of the images can be ensured. This approach simulates the actual environment in which produce is transported, providing a more accurate benchmark for freshness assessment.

[0027] By collecting images at intervals, we can capture details of how the freshness of agricultural products changes over time. This time series data helps analyze the speed and pattern of the decline in freshness of agricultural products, providing a scientific basis for freshness management.

[0028] Labeling each standard image with the shelf life of the product as a freshness indicator allows the image features to be associated with quantitative data on the freshness of the produce. This labeling method provides training data for freshness identification using machine learning and artificial intelligence techniques.

[0029] Simulating the same light, humidity and temperature conditions in the transport vehicle helps to study the specific effects of these environmental factors on the freshness of produce. This controlled environment simulation can be used to test and optimize preservation methods and transportation conditions.

[0030] In another preferred embodiment of the present invention, in the parameter control module, the adjustment process of the refrigeration power is: The heat increase rate P in the compartment is calculated based on the compartment temperature change value d, and the formula is: ; Where β is the preset correction coefficient, β is greater than 1, c is the specific heat capacity of air, ρ is the density of air, V is the volume of the car, and t represents unit time.

[0031] The parameter control module can dynamically adjust the refrigeration power according to the speed of freshness change and the change value of the compartment temperature to ensure that the agricultural products are kept in the best state. This dynamic adjustment capability reduces the quality decline of agricultural products caused by environmental discomfort.

[0032] In another preferred embodiment of the present invention, in the risk judgment module, when the transport vehicle passes through the replaced original destination, it passes directly without stopping and prompts the staff to shorten the unloading time at each destination.

[0033] The system reduces unnecessary transportation stops and unloading time and improves transportation efficiency through precise freshness management and transportation strategy optimization. At the same time, it reduces the loss of agricultural products during transportation and maintains a high commodity value.

[0034] In another preferred embodiment of the present invention, all destinations before the expected time are selected and marked as pending destinations, and the unloading quantity of each pending destination is obtained, and the destination with the lowest unloading quantity is marked as the preferred destination.

[0035] See also Figure 2 As shown, the present invention also provides a method for managing agricultural product supply chain transportation, comprising the following steps: Collect images of agricultural products in various areas of the agricultural product transport vehicle at intervals, pre-process the agricultural product images, and obtain agricultural product images corresponding to each destination, input the agricultural product images into a preset freshness recognition model, and output the freshness of the agricultural products at each destination at the current moment; Calculate the freshness change rate per unit time. When the freshness change rate is greater than a preset threshold, obtain the cabin temperature change value per unit time. If the cabin temperature change value is greater than the preset threshold, increase the cooling power of the cabin refrigeration device. Fit the collected agricultural product freshness data in time series to obtain a fitting function of the change of agricultural product freshness over time at each destination; The expected freshness of agricultural products upon arrival at any destination is predicted based on the fitting function. If the expected freshness is lower than the freshness standard of the current variety of agricultural products, the expected moment when the freshness of the agricultural products at the destination is equal to the freshness standard is obtained based on the fitting function, and the previous destination before the expected moment is selected and marked as the preferred destination, and the original destination of the batch of agricultural products is replaced with the preferred destination.

[0036] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. An agricultural product supply chain transportation management system, characterized in that: include: An image acquisition module is used to collect images of agricultural products in the agricultural product transport vehicle at intervals, pre-process the agricultural product images, and obtain agricultural product images corresponding to each destination, respectively input the agricultural product images into a preset freshness recognition model, and output the freshness of the agricultural products at each destination at the current moment; A parameter control module is used to calculate the freshness change rate per unit time. When the freshness change rate is greater than a preset threshold, the cabin temperature change value per unit time is obtained. If the cabin temperature change value is greater than the preset threshold, the cooling power of the cabin refrigeration device is increased; The data processing module is used to fit the collected agricultural product freshness data in time series to obtain a fitting function of the change of agricultural product freshness over time at each destination; The risk judgment module is used to predict the expected freshness of agricultural products when they arrive at any destination based on the fitting function. If the expected freshness is lower than the freshness standard of the current variety of agricultural products, the expected time when the freshness of the agricultural products at the destination is equal to the freshness standard is obtained based on the fitting function, and the destination before the expected time is selected and marked as the preferred destination, and the original destination of the batch of agricultural products is replaced with the preferred destination.

2. The agricultural product supply chain transportation management system according to claim 1, characterized in that: In the image acquisition module, the preprocessing process is: The edge detection based on low-pass filtering is used to filter out the image edge details of the agricultural product image, and the image is gray-processed, and the gray-scale curve of the image is adjusted to a preset dynamic range. Based on the fusion of BP neural network classification algorithm and decision tree splitting algorithm, the agricultural product image is divided into sub-images representing different appearance features, and the appearance features include gray-scale range, point density and texture.

3. The agricultural product supply chain transportation management system according to claim 2, characterized in that: In the image acquisition module, the process of outputting freshness is as follows: Establish an image library, in which standard agricultural product images with freshness annotated are stored, the standard agricultural product images are preprocessed and token prompts are added, the processed standard sub-images are converted into vectors, and the vectors are input into the Transformer neural network model after position embedding encoding, and the appearance features in different standard sub-images are respectively extracted, and the Transformer neural network model is updated based on the existing freshness annotations by back propagation, and the existing freshness annotations are repeatedly matched with the appearance features until the loss function of the model converges; The agricultural product images collected in real time are also preprocessed and then token prompts are added. The processed real-time sub-images are encoded by position embedding and input into the matched Transformer neural network model. The appearance features in different real-time sub-images are extracted respectively, and the initial freshness labels are generated by clustering. The Transformer neural network model is updated by backpropagation based on the initial freshness labels, and the freshness labels are repeatedly updated until the model loss function converges to obtain the final freshness.

4. The agricultural product supply chain transportation management system according to claim 1, characterized in that: In the data processing module, the fitting function is based on the power exponential function, and the formula is: Fresh=Cexp(a(T+T’) b ); Among them, Fresh represents freshness, C is a constant, exp represents a function with base e, T represents the transportation time of the transport vehicle, T' represents the storage time of the agricultural products before the transport vehicle departs, a and b represent function parameters, and the freshness and time data are substituted into the function to obtain the values ​​of the function parameters.

5. The agricultural product supply chain transportation management system according to claim 3 is characterized in that: The generation process of the standard agricultural product image is to select agricultural product samples, place the agricultural product samples in an environment with the same lighting, humidity and temperature as the transport vehicle, collect images of the agricultural product samples every unit time, which are standard agricultural product images, and annotate the standard agricultural product images with the length of time the products have been placed when they were taken, which is the freshness.

6. The agricultural product supply chain transportation management system according to claim 1, characterized in that: In the parameter control module, the adjustment process of the cooling power is as follows: The heat increase rate P in the compartment is calculated based on the compartment temperature change value d, and the formula is: ; Where β is the preset correction coefficient, β is greater than 1, c is the specific heat capacity of air, ρ is the density of air, V is the volume of the car, and t represents unit time.

7. The agricultural product supply chain transportation management system according to claim 1, characterized in that: In the risk judgment module, when the transport vehicle passes through the replaced original destination, it passes directly without stopping and prompts the staff to shorten the unloading time at each destination.

8. The agricultural product supply chain transportation management system according to claim 1, characterized in that: The process of selecting the preferred destination is as follows: Select all destinations before the expected time and mark them as pending destinations. Get the unloading quantity of each pending destination and mark the destination with the lowest unloading quantity as the preferred destination.

9. A method for transport management of agricultural product supply chain, characterized in that: The following steps are involved: Collect images of agricultural products in various areas of the agricultural product transport vehicle at intervals, pre-process the agricultural product images, and obtain agricultural product images corresponding to each destination, input the agricultural product images into a preset freshness recognition model, and output the freshness of the agricultural products at each destination at the current moment; Calculate the freshness change rate per unit time. When the freshness change rate is greater than a preset threshold, obtain the cabin temperature change value per unit time. If the cabin temperature change value is greater than the preset threshold, increase the cooling power of the cabin refrigeration device. Fit the collected agricultural product freshness data in time series to obtain a fitting function of the change of agricultural product freshness over time at each destination; The expected freshness of agricultural products upon arrival at any destination is predicted based on the fitting function. If the expected freshness is lower than the freshness standard of the current variety of agricultural products, the expected moment when the freshness of the agricultural products at the destination is equal to the freshness standard is obtained based on the fitting function, and the previous destination before the expected moment is selected and marked as the preferred destination, and the original destination of the batch of agricultural products is replaced with the preferred destination.

Citation Information

Patent Citations

  • Cold chain transportation risk early warning method and system based on Internet of Things data monitoring

    CN116976675A

  • Agricultural product supply chain-based agricultural product collaborative scheduling method and system

    CN117495019A

  • Frozen fresh fruit intelligent distribution method and system

    CN117726258A

  • Cold-chain food epidemic prevention detection system and method

    CN117763465A

  • Information convergence technology system for managing and distributing fresh agricultural products

    KR1020150136159A

Cited By

  • Intelligent cold-chain logistics temperature monitoring system for logistics field

    CN120146427A

  • Full-process management method and system applied to fresh-keeping storage and transportation of agricultural products

    CN120338648A