Logistics traffic control method based on Internet of Vehicles
By deploying image acquisition equipment and freshness recognition models in the logistics car, monitoring and predicting fruit freshness in real time, and dynamically adjusting the transportation route, the problem of fruit aging due to special circumstances during transportation is solved, and the fruit is achieved at the goal of reaching its goal Maintain optimal freshness in the ground, reducing loss and waste.
Patent Information
- Application Number
- CN202411262031.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-10
- Publication Date
- 2025-02-11
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
During transportation, the freshness of the fruit is aging due to special circumstances. The temperature control alone cannot ensure that the fruit reaches the freshness required by the customer when it arrives at the destination, and there is a risk of product rot and waste.
Using the logistics traffic control method based on the Internet of Vehicles, by deploying image acquisition equipment in the logistics car, collecting fruit images and preprocessing, inputting the freshness recognition model to identify freshness, collecting data at fixed times and drawing a line chart and fitting Curve predicts freshness when reaching the destination and dynamically calculates the backup destination when the expected freshness is below the threshold.
Real-time monitoring and prediction of fruit freshness is achieved, potential problems are discovered in advance, and transportation routes are dynamically adjusted to ensure that fruits remain in the best fresh state when they arrive at their destination, reducing losses and waste, improving consumer satisfaction and product market competitiveness .
Smart Images
Figure CN119417338A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of logistics management, and in particular to a logistics traffic control method based on vehicle networking. Background Art
[0002] With the development of the logistics industry, higher requirements are placed on the freshness management of goods during the logistics process. Especially for perishable foods such as fruits, their freshness is directly related to the value of the goods and consumer satisfaction. Traditional logistics management methods often ignore the real-time monitoring and management of the freshness of goods during transportation, resulting in a decrease in the freshness of goods when they arrive at the destination, affecting sales and use.
[0003] At present, although attempts are made to maintain the freshness of goods during transportation through temperature control and other methods, due to certain special circumstances, the aging of fruits is accelerated, and temperature control alone cannot ensure that the fruits are delivered to the destination with the freshness required by customers. There is a risk of goods rotting and being wasted. Therefore, it is necessary to adjust the logistics route or handling method according to the actual status of the goods, mainly to avoid risks. Summary of the invention
[0004] The purpose of the present invention is to provide a logistics traffic control method based on vehicle networking to solve the following technical problems: Due to certain special circumstances that cause the aging of fruits to accelerate, temperature control alone cannot ensure that the fruits are delivered to the destination with the freshness required by the customer, and there is a risk of the goods rotting and being wasted.
[0005] The purpose of the present invention can be achieved through the following technical solutions: A logistics traffic control method based on vehicle networking includes the following steps: The fruits in the logistics vehicle compartment are divided into several areas according to their varieties, and the fruit images of the fruits in the different variety areas are collected and pre-processed respectively. The pre-processed fruit images are input into the pre-built freshness recognition model to identify the freshness of different varieties of fruits in the current logistics vehicle compartment; Collect freshness data of different varieties of fruits at fixed intervals, draw a line graph of the change of fruit freshness over time, fit the line graph according to the existing freshness data, and generate a fitting curve of the change of freshness of different varieties of fruits over time in real time; Predicting the expected freshness of each variety of fruit when it arrives at the corresponding destination according to the fitting curve, if the expected freshness of the fruit in any area is lower than the corresponding set threshold, marking the fruit in the area as fruit to be processed; According to the fitting curve, the estimated time when the freshness of the fruit to be processed is equal to the set threshold is obtained. According to the time difference between the estimated time and the current time, the estimated position of the logistics vehicle at the estimated time is calculated, and the other destinations closest to the estimated position are traced back. The destination is used as a backup destination and the backup destination is updated in real time according to the fitting curve.
[0006] As a further solution of the present invention: the pretreatment process is: The fruit image is subjected to noise reduction processing, and the grayscale image is subjected to edge detection based on the Sobel edge detection operator, the grayscale image after edge detection is converted into a binary image, and the binary image is framed based on a preset rectangular bounding box, the proportion of the rectangular bounding box is set according to a fruit packaging box, a pixel area that meets the rectangular bounding box is selected, the fruit image is divided, and a fruit image containing only the fruit area is obtained.
[0007] As a further solution of the present invention: the construction process of the freshness recognition model is: An image library is established, in which fruit image samples with marked placement time and unmarked fruit image materials are stored, wherein the placement time represents the time after picking, each fruit image sample is divided into a plurality of image regions, and each image region is mapped into an image vector, a freshness recognition model is constructed based on a Transformer network, the image vector is encoded and input into the freshness recognition model, and freshness features in the fruit image samples are extracted; and the model is updated by back propagation based on the time length annotation until the loss function converges; the fruit image materials are also mapped into image vectors and input into the freshness recognition model, freshness features in the fruit image samples are extracted, the placement time is identified and clustered according to the freshness features, and the model is updated by back propagation based on the clustering results until the loss function converges.
[0008] As a further solution of the present invention: the appearance of the fruit every hour under a preset standard environment after picking is defined as freshness, and the unit of the freshness is standard hours.
[0009] As a further solution of the present invention: the formula for fitting the change of fruit freshness over time is: F=Cexp(BT A ); Where F represents freshness, T represents time, A and B are parameters, B is a preset constant, and exp() represents the natural exponential function. The values of A and B are calculated based on the freshness data.
[0010] As a further solution of the present invention: a corresponding freshness recognition model and a fruit freshness fitting formula are constructed for each variety of fruit.
[0011] As a further solution of the present invention: the specific process of calculating the backup destination is: According to the real-time fitting curve formula, the estimated time when the freshness of the fruit to be processed is equal to the set threshold is calculated. According to the time difference between the estimated time and the current time, and the current driving speed of the logistics vehicle, the driving distance and the corresponding estimated position of the logistics vehicle when arriving at the estimated time are calculated. The driving distance is traced back from the estimated position, and the destination closest to the estimated position is updated as the backup destination. When there is no destination in the driving distance, it is prompted to place the fruit to be processed separately or discard it.
[0012] Beneficial effects of the present invention: The present invention monitors the freshness of fruits at fixed intervals and draws a freshness change curve, thereby realizing real-time monitoring and prediction of the freshness of fruits, so as to discover potential freshness problems in advance, and dynamically calculate alternative destinations according to the freshness prediction results, so that logistics scheduling can respond to changes in the freshness of fruits in a timely manner, optimize transportation routes, and ensure that the fruits are still in the best fresh state when they arrive at the destination; through accurate monitoring and management of freshness, fruits with freshness below a threshold can be handled in a timely manner, and effective measures such as changing the destination or adjusting the storage environment can be taken to reduce loss and waste; ensuring that fruits and other commodities maintain high quality and freshness during transportation can improve consumer satisfaction when receiving the commodities, thereby enhancing the market competitiveness and brand image of the commodities. BRIEF DESCRIPTION OF THE DRAWINGS
[0013] The present invention will be further described below in conjunction with the accompanying drawings.
[0014] Figure 1 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION
[0015] 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.
[0016] See also Figure 1 As shown, the present invention is a logistics traffic control method based on vehicle networking, comprising the following steps: The fruits in the logistics vehicle compartment are divided into several areas according to their varieties, and the fruit images of the fruits in the different variety areas are collected and pre-processed respectively. The pre-processed fruit images are input into the pre-built freshness recognition model to identify the freshness of different varieties of fruits in the current logistics vehicle compartment; Collect freshness data of different varieties of fruits at fixed intervals, draw a line graph of the change of fruit freshness over time, fit the line graph according to the existing freshness data, and generate a fitting curve of the change of freshness of different varieties of fruits over time in real time; Predicting the expected freshness of each variety of fruit when it arrives at the corresponding destination according to the fitting curve, if the expected freshness of the fruit in any area is lower than the corresponding set threshold, marking the fruit in the area as fruit to be processed; According to the fitting curve, the estimated time when the freshness of the fruit to be processed is equal to the set threshold is obtained. According to the time difference between the estimated time and the current time, the estimated position of the logistics vehicle at the estimated time is calculated, and the other destinations closest to the estimated position are traced back. The destination is used as a backup destination and the backup destination is updated in real time according to the fitting curve.
[0017] By deploying image acquisition equipment in logistics carriages and implementing fruit freshness recognition models, the freshness of different varieties of fruits can be automatically and quickly evaluated, reducing the need for manual inspection and improving the accuracy and efficiency of freshness management.
[0018] It allows monitoring the freshness of fruits at fixed intervals and drawing freshness change curves, thereby achieving real-time monitoring and prediction of fruit freshness to detect potential freshness problems in advance. Dynamically calculating alternative destinations based on freshness prediction results allows logistics scheduling to respond to changes in fruit freshness in a timely manner, optimize transportation routes, and ensure that fruits remain in the best fresh state when they arrive at their destinations. Through accurate monitoring and management of freshness, fruits with freshness below the threshold can be handled in a timely manner, and effective measures such as changing destinations or adjusting storage environments can be taken to reduce losses and waste.
[0019] In another preferred embodiment of the present invention, the pretreatment process is: The fruit image is subjected to noise reduction processing, and the grayscale image is subjected to edge detection based on the Sobel edge detection operator, the grayscale image after edge detection is converted into a binary image, and the binary image is framed based on a preset rectangular bounding box, the proportion of the rectangular bounding box is set according to a fruit packaging box, a pixel area that meets the rectangular bounding box is selected, the fruit image is divided, and a fruit image containing only the fruit area is obtained.
[0020] In another preferred embodiment of the present invention, the construction process of the freshness recognition model is: An image library is established, in which fruit image samples with marked placement time and unmarked fruit image materials are stored, wherein the placement time represents the time after picking, each fruit image sample is divided into a plurality of image regions, and each image region is mapped into an image vector, a freshness recognition model is constructed based on a Transformer network, the image vector is encoded and input into the freshness recognition model, and freshness features in the fruit image samples are extracted; and the model is updated by back propagation based on the time length annotation until the loss function converges; the fruit image materials are also mapped into image vectors and input into the freshness recognition model, freshness features in the fruit image samples are extracted, the placement time is identified and clustered according to the freshness features, and the model is updated by back propagation based on the clustering results until the loss function converges.
[0021] In another preferred embodiment of the present invention, the appearance of the fruit every hour under a preset standard environment after picking is defined as freshness, and the unit of the freshness is standard hours.
[0022] The storage time corresponding to different changes in the appearance of fruits under standard conditions is defined as freshness, and then this standard is used to measure the standard time corresponding to the appearance of fruits in different environments, so that the freshness of fruits can be judged through a unified standard and errors can be prevented.
[0023] In another preferred embodiment of the present invention, the formula for fitting the change of fruit freshness over time is: F=Cexp(BT A ); Where F represents freshness, T represents time, A and B are parameters, B is a preset constant, and exp() represents the natural exponential function. The values of A and B are calculated based on the freshness data.
[0024] In another preferred embodiment of the present invention, a corresponding freshness recognition model and a fruit freshness fitting formula are constructed for each variety of fruit.
[0025] In another preferred embodiment of the present invention, the specific process of calculating the backup destination is: According to the real-time fitting curve formula, the estimated time when the freshness of the fruit to be processed is equal to the set threshold is calculated. According to the time difference between the estimated time and the current time, and the current driving speed of the logistics vehicle, the driving distance and the corresponding estimated position of the logistics vehicle when arriving at the estimated time are calculated. The driving distance is traced back from the estimated position, and the destination closest to the estimated position is updated as the backup destination. When there is no destination in the driving distance, it is prompted to place or discard the fruit to be processed separately, so as to avoid the damage of the fruit to affect other fresh fruits and reduce losses.
[0026] 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. A logistics traffic control method based on vehicle networking, characterized in that: The following steps are involved: The fruits in the logistics vehicle compartment are divided into several areas according to their varieties, and the fruit images of the fruits in the different variety areas are collected and pre-processed respectively. The pre-processed fruit images are input into the pre-built freshness recognition model to identify the freshness of different varieties of fruits in the current logistics vehicle compartment; Collect freshness data of different varieties of fruits at fixed intervals, draw a line graph of the change of fruit freshness over time, fit the line graph according to the existing freshness data, and generate a fitting curve of the change of freshness of different varieties of fruits over time in real time; Predicting the expected freshness of each variety of fruit when it arrives at the corresponding destination according to the fitting curve, if the expected freshness of the fruit in any area is lower than the corresponding set threshold, marking the fruit in the area as fruit to be processed; According to the fitting curve, the estimated time when the freshness of the fruit to be processed is equal to the set threshold is obtained. According to the time difference between the estimated time and the current time, the estimated position of the logistics vehicle at the estimated time is calculated, and the other destinations closest to the estimated position are traced back. The destination is used as a backup destination and the backup destination is updated in real time according to the fitting curve.
2. A method for controlling logistics and traffic based on Internet of Vehicles according to claim 1, characterized in that: The pre-processing process is as follows: The fruit image is subjected to noise reduction processing, and the grayscale image is subjected to edge detection based on the Sobel edge detection operator, the grayscale image after edge detection is converted into a binary image, and the binary image is framed based on a preset rectangular bounding box, the proportion of the rectangular bounding box is set according to a fruit packaging box, a pixel area that meets the rectangular bounding box is selected, the fruit image is divided, and a fruit image containing only the fruit area is obtained.
3. The method for controlling logistics and traffic based on Internet of Vehicles according to claim 1, characterized in that: The construction process of the freshness recognition model is as follows: An image library is established, in which fruit image samples with marked placement time and unmarked fruit image materials are stored, wherein the placement time represents the time after picking, each fruit image sample is divided into a plurality of image regions, and each image region is mapped into an image vector, a freshness recognition model is constructed based on a Transformer network, the image vector is encoded and input into the freshness recognition model, and freshness features in the fruit image samples are extracted; and the model is updated by back propagation based on the time length annotation until the loss function converges; the fruit image materials are also mapped into image vectors and input into the freshness recognition model, freshness features in the fruit image samples are extracted, the placement time is identified and clustered according to the freshness features, and the model is updated by back propagation based on the clustering results until the loss function converges.
4. The method for controlling logistics and traffic based on Internet of Vehicles according to claim 1, characterized in that: The appearance of the fruit every hour under a preset standard environment after picking is defined as freshness, and the unit of the freshness is standard hour.
5. The method for controlling logistics and traffic based on Internet of Vehicles according to claim 1, characterized in that: The formula for fitting the change of fruit freshness over time is: F=Cexp(BT A ); Where F represents freshness, T represents time, A and B are parameters, B is a preset constant, and exp() represents the natural exponential function. The values of A and B are calculated based on the freshness data.
6. A method for controlling logistics and traffic based on Internet of Vehicles according to claim 3 or 5, characterized in that: A corresponding freshness recognition model and fruit freshness fitting formula are constructed for each variety of fruit.
7. The method for controlling logistics and traffic based on Internet of Vehicles according to claim 1, characterized in that: The specific process of calculating the alternative destination is: According to the real-time fitting curve formula, the estimated time when the freshness of the fruit to be processed is equal to the set threshold is calculated. According to the time difference between the estimated time and the current time, and the current driving speed of the logistics vehicle, the driving distance and the corresponding estimated position of the logistics vehicle when arriving at the estimated time are calculated. The driving distance is traced back from the estimated position, and the destination closest to the estimated position is updated as the backup destination. When there is no destination in the driving distance, it is prompted to place the fruit to be processed separately or discard it.