A logistics service digital management system

Abstract

This invention relates to the field of logistics monitoring technology, and more particularly to a digital management system for logistics services, comprising: environmental sensors, an electronic nose, and a vision sensor, all installed within a refrigerated logistics transport device used for transporting fruits and vegetables. These sensors are respectively used to acquire environmental data of the refrigerated fruits and vegetables at an environmental acquisition frequency, to acquire odor data of the fruits and vegetables at an electronic nose acquisition frequency, and to acquire surface images of the fruits and vegetables at a vision acquisition frequency; a status judgment module to identify the internal and external rot status, pesticide residue status, and abnormal appearance status of the fruits and vegetables; a frequency adjustment module to adjust the environmental acquisition frequency, electronic nose acquisition frequency, and vision acquisition frequency; and a blockchain storage module to store the repeatedly acquired environmental data, odor data, and surface images in a blockchain ledger. This invention controls the sensor acquisition frequency and the amount of data stored in the blockchain ledger based on multiple fruit and vegetable status judgment results, achieving high processing speed of the blockchain.

Description

Technical Field

[0001] This invention relates to the field of logistics monitoring technology, and in particular to a digital management system for logistics services. Background Technology

[0002] Cold chain logistics generally refers to a systematic project that ensures refrigerated goods are kept in a specified low-temperature environment throughout all stages from production, storage, transportation, and sales to consumption, in order to guarantee the quality of the goods and reduce losses. It was established with the advancement of science and technology and the development of refrigeration technology. It is a low-temperature logistics process based on refrigeration technology and using refrigeration technology as a means.

[0003] To continuously monitor the condition of fruits and vegetables within logistics and transportation facilities, and to determine the relationship between environmental factors such as refrigeration temperature, humidity, and ventilation of the facilities and the causes of quality problems such as rotting, souring, surface pitting, and insect bites, thus accurately assigning responsibility for the decline in fruit and vegetable quality, blockchain technology has been adopted for storing sensor-collected data. The distributed storage and peer-to-peer transmission of blockchain technology ensure the immutability of the sensor-collected data, guaranteeing data security and transparency.

[0004] However, because the ledger of blockchain technology is stored in a distributed manner, when storing data collected by sensors from multiple locations over a long period of time, the amount of data in the ledger can easily become too large, leading to problems such as increased storage costs, slower synchronization speed, increased latency for new nodes joining the network, slower transaction speed, and increased network congestion.

[0005] Therefore, how to enable sensor data collected during logistics transportation to accurately reflect the status of fruits and vegetables while reducing the amount of data stored in the blockchain ledger, thereby improving the performance of the blockchain, is a technical problem that needs to be solved. Summary of the Invention

[0006] To address this, the present invention provides a digital management system for logistics services. By collecting images from visual sensors and data from an electronic nose, it can accurately determine the internal and external rot, pesticide residue, and abnormal appearance of fruits and vegetables. The system then controls the sensor's acquisition frequency and the amount of data stored in the blockchain ledger based on the judgment results. It also provides accurate alarms through blockchain smart contracts, achieving high processing speed and low storage requirements of the blockchain, thus reducing the burden on users to view the ledger.

[0007] To achieve the above objectives, the present invention proposes a digital management system for logistics services, comprising:

[0008] Environmental sensors, electronic noses, and vision sensors are all installed in refrigerated logistics transport equipment used to transport fruits and vegetables. They are used to acquire environmental data of refrigerated fruits and vegetables at an environmental acquisition frequency, acquire odor data of fruits and vegetables at an electronic nose acquisition frequency, and acquire surface images of fruits and vegetables at a vision acquisition frequency, respectively.

[0009] The status judgment module is used to identify the internal and external rot status, pesticide residue status, and abnormal appearance status of fruits and vegetables based on the odor data and the surface image.

[0010] The frequency adjustment module is used to adjust the environmental acquisition frequency, the electronic nose acquisition frequency, and the visual acquisition frequency according to the internal and external decay state, the pesticide residue state, and / or the abnormal appearance state.

[0011] A blockchain storage module is used to store the environmental data, odor data and surface images collected multiple times in the blockchain ledger based on the internal and external decay status, the pesticide residue status and / or the abnormal appearance status.

[0012] An alarm module is used to determine whether to generate alarm information including environmental data, odor data, and surface image based on the internal and external decay status, pesticide residue status, and abnormal appearance status, and to send the alarm information through a blockchain smart contract.

[0013] Furthermore, the state determination module includes a support vector machine model, a neural network model, and a comprehensive state determination unit;

[0014] The support vector machine model is used to identify the odor decay characteristics and the pesticide residue status based on the odor data.

[0015] The neural network model is used to identify surface decay features and abnormal appearance based on the surface image.

[0016] The comprehensive state determination unit is used to determine the internal and external decay state based on the odor decay characteristics and the surface decay characteristics.

[0017] Furthermore, the internal and external decay states include simultaneous internal and external decay states, internal decay states, external decay states, and causes of decay, wherein the causes of decay include physical decay, microbial decay, and oxidative decay.

[0018] The frequency adjustment module is used to increase the environmental acquisition frequency to a first environmental dynamic frequency when the comprehensive state determination unit determines that the internal and external decay state is a state of simultaneous internal and external decay or an external decay state, and the cause of decay is microbial decay or oxidative decay.

[0019] The frequency adjustment module is also used to increase the electronic nose collection frequency by a first odor dynamic frequency when the comprehensive state determination unit determines that the internal and external decay state is an internal decay state and the cause of decay is microbial decay or oxidative decay.

[0020] The frequency adjustment module is further configured to increase the visual acquisition frequency by a first visual dynamic frequency when the comprehensive state determination unit determines that the cause of decay is physical decay.

[0021] Furthermore, the internal and external decay state also includes decay level;

[0022] The frequency adjustment module is used to adjust the first environmental dynamic frequency, the first odor dynamic frequency, or the first visual dynamic frequency according to the duration of the change in the decay level.

[0023] Furthermore, the pesticide residue status includes pesticide residue level, and the abnormal appearance status includes abnormal appearance level;

[0024] The frequency adjustment module is also used to increase the electronic nose acquisition frequency by a second odor dynamic frequency and the environmental acquisition frequency by a second environmental dynamic frequency when the support vector machine model identifies the pesticide residue status. The second odor dynamic frequency and the second environmental dynamic frequency are respectively related to the duration of the change in the pesticide residue level.

[0025] The frequency adjustment module is further configured to increase the visual acquisition frequency by a second visual dynamic frequency and the environmental acquisition frequency by a third environmental dynamic frequency when the neural network model identifies the abnormal appearance state. The second visual dynamic frequency and the third environmental dynamic frequency are respectively related to the duration of the change in the abnormal appearance level.

[0026] The above solution enables dynamic adjustment of the sensor's acquisition frequency based on various fruit and vegetable conditions, reducing the energy consumption of the sensor and the amount of data stored in the blockchain ledger.

[0027] Furthermore, the alarm module is used to send alarm information via a blockchain smart contract when the environmental data exceeds an environmental threshold, wherein the environmental threshold is determined based on the internal and external decay status, the pesticide residue status, and / or the abnormal appearance status.

[0028] Furthermore, the blockchain storage module is used to merge and store multiple environmental data, odor data, and surface images in the blockchain ledger when the state judgment module identifies the internal and external rot state, the pesticide residue state, or the abnormal appearance state.

[0029] Furthermore, the blockchain storage module is used to store the standard values ​​of multiple environmental data, the standard values ​​of multiple odor data, and the fused image generated by the multiple surface images through an image fusion algorithm in the ledger of the blockchain.

[0030] The above solution achieves the merging and processing of blockchain ledger storage data, ensuring data integrity and reducing useless data.

[0031] Furthermore, the alarm information includes the internal and external rot status, the pesticide residue status, and the abnormal appearance status, as well as the correlation determination with the environmental data.

[0032] Furthermore, the blockchain ledger is simultaneously set up within the systems of regulatory agencies and enterprises;

[0033] A blockchain storage module is used to synchronously send the alarm information to the systems of the regulatory agencies and enterprises.

[0034] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0035] 1. By collecting images from visual sensors and data from electronic noses, the system can accurately determine the internal and external rot, pesticide residue, and abnormal appearance of fruits and vegetables. The system can then control the sensor's acquisition frequency and the amount of data stored in the blockchain ledger based on the judgment results. Accurate alarms can be triggered through blockchain smart contracts, achieving high processing speed and low storage requirements of the blockchain, and reducing the burden on users to view the ledger.

[0036] 2. It enables dynamic adjustment of the sensor's acquisition frequency based on various fruit and vegetable conditions, reducing the energy consumption of the sensor and the amount of data stored in the blockchain ledger.

[0037] 3. It realizes the merging and processing of blockchain ledger storage data, ensuring data integrity and reducing useless data. Attached Figure Description

[0038] Figure 1 This is a schematic diagram of the general structure of the digital logistics service management system according to an embodiment of the present invention;

[0039] Figure 2 This is a schematic diagram of the general process of the digital logistics service management system according to an embodiment of the present invention;

[0040] Figure 3 This is a detailed schematic diagram of the sensor acquisition frequency adjustment process of the logistics service digital management system according to an embodiment of the present invention;

[0041] Figure 4This is a schematic diagram illustrating the detailed blockchain ledger and smart contract sending judgment of the logistics service digital management system according to an embodiment of the present invention. Detailed Implementation

[0042] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0043] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0044] It should be noted that in the description of this invention, the terms "upper", "lower", "left", "right", "inner", "outer", etc., which indicate directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings. This is only for the convenience of description and is not intended to indicate or imply that the device or element must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of this invention.

[0045] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0046] like Figures 1 to 4 As shown, this invention provides a digital management system for logistics services. By collecting images from visual sensors and data from an electronic nose, it can accurately determine the internal and external rot, pesticide residue, and abnormal appearance of fruits and vegetables. The system then controls the sensor's acquisition frequency and the amount of data stored in the blockchain ledger based on the judgment results. Accurate alarms are triggered through blockchain smart contracts, achieving high processing speed and low storage requirements of the blockchain, thus reducing the burden on users to view the ledger.

[0047] in, Figure 1 This is a schematic diagram of the general structure of the digital logistics service management system according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the general process of the digital logistics service management system according to an embodiment of the present invention; Figure 3 This is a detailed schematic diagram of the sensor acquisition frequency adjustment process of the logistics service digital management system according to an embodiment of the present invention; Figure 4This is a schematic diagram illustrating the detailed blockchain ledger and smart contract sending judgment of the logistics service digital management system according to an embodiment of the present invention.

[0048] Example 1:

[0049] like Figures 1 to 4 As shown, this embodiment proposes a digital management system for logistics services, including: an environmental sensor, an electronic nose, and a vision sensor, all of which are installed in a refrigerated logistics transportation device for transporting fruits and vegetables. They are respectively used to acquire environmental data of refrigerated fruits and vegetables at an environmental acquisition frequency, acquire odor data of fruits and vegetables at an electronic nose acquisition frequency, and acquire surface images of fruits and vegetables at a vision acquisition frequency.

[0050] The status judgment module is used to identify the internal and external rot status, pesticide residue status, and abnormal appearance status of fruits and vegetables based on the odor data and the surface image; the frequency adjustment module is used to adjust the environmental acquisition frequency, the electronic nose acquisition frequency, and the visual acquisition frequency based on the internal and external rot status, the pesticide residue status, and / or the abnormal appearance status.

[0051] A blockchain storage module is used to store the environmental data, odor data, and surface images collected multiple times in the blockchain ledger based on the internal and external decay status, pesticide residue status, and / or abnormal appearance status; an alarm module is used to determine whether to generate alarm information including the environmental data, odor data, and surface image based on the internal and external decay status, pesticide residue status, and abnormal appearance status, and to send the alarm information through a blockchain smart contract.

[0052] From a structural perspective, blockchain exists as a distributed ledger, with each block containing the hash value of the previous block, forming an immutable chain structure. In this system, all participants, including regulators, farmers, sellers, and transporters, can access complete on-chain information. The distributed storage design ensures that even a single node failure will not result in information loss, thus improving system stability and reliability. The blockchain's smart contract function guarantees automatic contract execution when specific conditions are met. In this embodiment, it automatically generates and issues an alarm when internal / external rot, pesticide residue, or abnormal appearance are detected for a duration exceeding a set time. The set time is 30 seconds to avoid misjudgments due to inaccurate sensor readings. The set time can also be modified according to the specific fruit and vegetable type. Modifications to the initial sensor judgment time and the set value are recorded in the blockchain ledger to prevent malicious modification by the transporter.

[0053] Furthermore, the state determination module includes a support vector machine model, a neural network model, and a comprehensive state determination unit; the support vector machine model is used to identify odor decay characteristics and pesticide residue status based on the odor data; the neural network model is used to identify surface decay characteristics and abnormal appearance status based on the surface image; and the comprehensive state determination unit is used to determine the internal and external decay status based on the odor decay characteristics and the surface decay characteristics.

[0054] Specifically, the electronic nose can capture volatile organic compounds (VOCs) in the odors produced by a large number of fruits and vegetables within a refrigerated logistics transportation device, and use machine learning algorithms for data analysis and classification. Specifically, the odor classification task is performed using a Support Vector Machine (SVM) model. The SVM model achieves a classification accuracy of up to 96% for pesticide odors; therefore, pesticide residue detection is performed solely using the SVM model and odor data. However, its accuracy in identifying the odors of fruit and vegetable decay caused by mold and microorganisms is only around 80%. Therefore, it is necessary to combine the output odor classification results with image-based judgment results for comprehensive assessment to improve the accuracy of fruit and vegetable decay detection.

[0055] More specifically, the Support Vector Machine (SVM) model comprises two processes: model training and odor recognition. The model training process involves: odor data with category labels undergoing preprocessing and feature extraction to obtain a 45-dimensional basic pattern recognition library; feature normalization is performed; a kernel function for the SVM is selected; a radial basis function (RBF) SVM is chosen; and the optimal parameters of the kernel function are found using a grid method. The category labels include pesticide categories such as imidacloprid, chlorpyrifos, benzyl dichlorvos, soybean seed dressing agents, and organophosphorus pesticides, as well as acidic and foul odors indicating decay. The SVM model is trained using the normalized feature set, and the parameters of the SVM model are output. The odor recognition process involves: odor data to be tested undergoing preprocessing and feature extraction to obtain a 45-dimensional test set; the normalized test set is input into the trained SVM model, and the category of the odor data is output. By calculating the time required to determine the odor of pesticide residues and decay, the pesticide residue level and odor decay level of fruits and vegetables are determined. The pesticide residue level and the odor and decay level are level one to level five, with the odor judgment time increasing sequentially. The odor judgment time is 40 seconds, 50 seconds and up to 80 seconds respectively.

[0056] Appearance quality is the most intuitive quality characteristic of fruits and vegetables. This paper utilizes convolutional neural networks to achieve automatic defect detection of fruits and vegetables, accurately identifying various fruits and vegetables in a given image through an efficient segmentation process. It generates a comprehensive image rot assessment and rot level for fruits and vegetables based on indentation features, rot features, discoloration features, and deformation features. Specifically, when rot features are detected, the fruit and vegetable are determined to be in a rotten state. If no rot features are present, but at least two of the indentation, discoloration, and deformation features are present, the fruit and vegetable are determined to be in an abnormal appearance state. The image rot level and appearance abnormality level are determined based on the proportion of the area containing rot / indentation, discoloration, and deformation features to the total image detection window area. The image rot level and appearance abnormality level are levels one through five, with proportions increasing sequentially from over 40%, over 50%, up to over 80%.

[0057] Furthermore, the internal and external decay states include simultaneous internal and external decay states, internal decay states, external decay states, and causes of decay, wherein the causes of decay include physical decay, microbial decay, and oxidative decay.

[0058] The frequency adjustment module is used to increase the environmental acquisition frequency by a first environmental dynamic frequency when the comprehensive state determination unit determines that the internal and external decay state is a simultaneous internal and external decay state or an external decay state, and the cause of decay is microbial decay or oxidative decay; the frequency adjustment module is also used to increase the electronic nose acquisition frequency by a first odor dynamic frequency when the comprehensive state determination unit determines that the internal and external decay state is an internal decay state, and the cause of decay is microbial decay or oxidative decay; the frequency adjustment module is also used to increase the visual acquisition frequency by a first visual dynamic frequency when the comprehensive state determination unit determines that the cause of decay is physical decay.

[0059] Specifically, if the odor decay level is less than or equal to level two and the image decay level is greater than or equal to level three, then the internal and external decay state is determined to be external decay state, the cause of decay is physical decay, and the decay level is medium.

[0060] If the odor decay level is less than or equal to level two, and the image decay level is less than or equal to level two, then the internal and external decay state is determined to be a state of simultaneous internal and external decay. The cause of decay is determined to be microbial decay or oxidative decay based on the odor data, and the decay level is low.

[0061] If the odor decay level is greater than or equal to level three and the image decay level is less than or equal to level two, then the internal and external decay state is determined to be internal decay state, and the cause of decay is microbial decay or oxidative decay determined based on the odor data, and the decay level is medium.

[0062] If the odor decay level is greater than or equal to level three, and the image decay level is greater than or equal to level three, then the internal and external decay state is determined to be a state of simultaneous internal and external decay. The cause of decay is a comprehensive decay including physical decay, microbial decay, or oxidative decay, and the decay level is high.

[0063] Furthermore, the internal and external decay state also includes decay level; the frequency adjustment module is used to adjust the first environmental dynamic frequency, the first odor dynamic frequency, or the first visual dynamic frequency according to the duration of change of the decay level.

[0064] Specifically, the initial values ​​for both the environmental sampling frequency and the electronic nose sampling frequency are 10 seconds / time. The range of both frequencies is constrained to 4 seconds / time to 15 seconds / time. The first environmental dynamic frequency / first odor dynamic frequency is calculated by dividing the standard value corresponding to the decay level by the logarithm of the decay level change duration (base 10), and then subtracting the standard value corresponding to the decay level. Therefore, within 10 seconds of a decay level change, if the logarithm of the change duration (base 10) is less than 1, the standard value divided by a number less than 1 increases, and the first environmental dynamic frequency / first odor dynamic frequency obtained by subtracting the original standard value from the increased standard value is positive. Similarly, after 10 seconds of the change, the first environmental dynamic frequency / first odor dynamic frequency becomes negative. The standard values ​​for low, medium, and high levels are 2 seconds / time, 3 seconds / time, and 5 seconds / time, respectively. After a change duration exceeding 40 seconds, the environmental sampling frequency and the electronic nose sampling frequency are restored to their initial value of 10 seconds / time.

[0065] Specifically, the initial visual acquisition frequency is 15 seconds / time, with a range constrained from 10 seconds / time to 25 seconds / time. The first visual dynamic frequency is calculated by dividing the standard value corresponding to the decay level by the logarithm of the decay level change duration (base 15), and then subtracting the standard value corresponding to the decay level. Therefore, within 15 seconds of a decay level change, if the logarithm of the change duration (base 15) is less than 1, the standard value divided by a number less than 1 increases, and the first visual dynamic frequency obtained by subtracting the original standard value from the increased standard value is positive. Similarly, after 15 seconds of the change, the first visual dynamic frequency becomes negative, and the visual acquisition frequency decreases. The standard values ​​for low, medium, and high levels are 5 seconds / time, 6 seconds / time, and 8 seconds / time, respectively.

[0066] Furthermore, the pesticide residue status includes pesticide residue level, and the appearance abnormality status includes appearance abnormality level; the frequency adjustment module is also used to increase the electronic nose acquisition frequency by a second odor dynamic frequency and the environmental acquisition frequency by a second environmental dynamic frequency when the support vector machine model identifies the pesticide residue status, wherein the second odor dynamic frequency and the second environmental dynamic frequency are respectively related to the change duration of the pesticide residue level; the frequency adjustment module is also used to increase the visual acquisition frequency by a second visual dynamic frequency and the environmental acquisition frequency by a third environmental dynamic frequency when the neural network model identifies the appearance abnormality status, wherein the second visual dynamic frequency and the third environmental dynamic frequency are respectively related to the change duration of the appearance abnormality level.

[0067] Specifically, the second odor dynamic frequency / second environmental dynamic frequency is calculated by dividing the standard value corresponding to the pesticide residue level by the logarithm of the pesticide residue level change duration (base 16), and then subtracting the standard value corresponding to the pesticide residue level. Therefore, within 16 seconds of the change in the decay level, if the logarithm of the change duration (base 16) is less than 1, the standard value divided by a number less than 1 increases, and the second odor dynamic frequency / second environmental dynamic frequency obtained by subtracting the original standard value from the increased standard value is positive. Similarly, if the second odor dynamic frequency / second environmental dynamic frequency is positive, it becomes negative after 16 seconds of change. The standard values ​​for the pesticide residue levels, from lowest to highest, are 2 seconds / time, 3 seconds / time, and 5 seconds / time.

[0068] Specifically, the second visual dynamic frequency / third environmental dynamic frequency is calculated by dividing the standard value corresponding to the appearance anomaly level by the logarithm of the change duration of the appearance anomaly level (base 16), and then subtracting the standard value corresponding to the appearance anomaly level. Therefore, if the logarithm of the change duration (base 16) is less than 1, the standard value divided by a number less than 1 increases, and the second visual dynamic frequency / third environmental dynamic frequency obtained by subtracting the original standard value from the increased standard value is positive. Similarly, 16 seconds after the change occurs, the second odor dynamic frequency / second environmental dynamic frequency becomes negative. The standard values ​​for the pesticide residue levels, from low to high, are 2 seconds / time, 3 seconds / time, and 5 seconds / time.

[0069] The above solution enables dynamic adjustment of the sensor's acquisition frequency based on various fruit and vegetable conditions, reducing the energy consumption of the sensor and the amount of data stored in the blockchain ledger.

[0070] Furthermore, the alarm module is used to send alarm information via a blockchain smart contract when the environmental data exceeds an environmental threshold, wherein the environmental threshold is determined based on the internal and external decay status, the pesticide residue status, and / or the abnormal appearance status.

[0071] Specifically, the environmental sensors include a temperature sensor, a humidity sensor, and a ventilation volume sensor. The lower and upper limits of the environmental thresholds corresponding to the temperature sensor are -8℃ and 12℃, respectively; the lower and upper limits of the environmental thresholds corresponding to the humidity sensor are 45% and 75%, respectively; and the lower and upper limits of the environmental thresholds corresponding to the ventilation volume sensor are 6000m³ / h, respectively. 3 / h、7400m 3 / h. If any of the following conditions are detected: internal or external rot, pesticide residue, or abnormal appearance, then the lower and upper limits of the environmental threshold corresponding to the temperature sensor will be increased by 1°C and decreased by 1°C, respectively; the lower and upper limits of the environmental threshold corresponding to the humidity sensor will be increased by 3% and decreased by 3%, respectively; and the lower and upper limits of the environmental threshold corresponding to the ventilation volume sensor will be increased by 50m. 3 / h, reduce by 50m 3 / h. Complies with national standard GB29753 "Safety Requirements and Test Methods for Refrigerated Vehicles for Road Transport, Food and Biological Products".

[0072] Furthermore, the blockchain storage module is used to merge and store multiple environmental data, odor data, and surface images in the blockchain ledger when the state judgment module identifies the internal and external rot state, the pesticide residue state, or the abnormal appearance state.

[0073] Furthermore, the blockchain storage module is used to store the standard values ​​of multiple environmental data, the standard values ​​of multiple odor data, and the fused image generated by the multiple surface images through an image fusion algorithm in the ledger of the blockchain.

[0074] Specifically, the average of multiple environmental data collected within 20 seconds is used as the standard value of the environmental data, and the average of multiple odor data collected within 20 seconds is used as the standard value of the odor data.

[0075] Specifically, the image fusion algorithm employs an alpha fusion model. Multiple surface images acquired within 10 seconds are cropped around the same labeled fruit / vegetable as the center, and then input into the alpha fusion model for image fusion. The alpha fusion model directly weights the corresponding images at each position in the two images participating in the fusion, controlling the proportion of each frame participating in the fusion process.

[0076] The above solution achieves the merging and processing of blockchain ledger storage data, ensuring data integrity and reducing useless data.

[0077] Furthermore, the alarm information includes the correlation between the internal and external decay status, the pesticide residue status, and the abnormal appearance status, and the environmental data. The correlation is determined by the duration of overlap between the occurrence time of the internal and external decay status, the pesticide residue status, and the abnormal appearance status and the change time of the environmental data; that is, the longer the overlap duration, the greater the correlation.

[0078] Furthermore, the blockchain ledger is synchronously set up within the systems of both the regulatory agency and the enterprise; the blockchain storage module is used to synchronously send the alarm information to the systems of both the regulatory agency and the enterprise. The regulatory agency is the Ministry of Agriculture and Rural Affairs.

[0079] Understandably, by using images captured by visual sensors and data collected by an electronic nose, accurate judgments can be made regarding the internal and external decay status, pesticide residue levels, and abnormal appearance of fruits and vegetables. The judgment results then control the sensor's acquisition frequency and the amount of data stored in the blockchain ledger. Accurate alarms are triggered via blockchain smart contracts, achieving high processing speed and low storage requirements, thus reducing the burden on users viewing the ledger. The system dynamically adjusts the sensor's acquisition frequency based on various fruit and vegetable conditions, reducing energy consumption during sensor operation and the amount of data stored in the blockchain ledger. Furthermore, it achieves merged processing of the blockchain ledger data, ensuring data integrity and reducing useless data.

[0080] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

[0081] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A digital management system for logistics services, characterized in that, include: Environmental sensors, electronic noses, and vision sensors are all installed in refrigerated logistics transport equipment used to transport fruits and vegetables. They are used to acquire environmental data of refrigerated fruits and vegetables at an environmental acquisition frequency, acquire odor data of fruits and vegetables at an electronic nose acquisition frequency, and acquire surface images of fruits and vegetables at a vision acquisition frequency, respectively. The status judgment module is used to identify the internal and external rot status, pesticide residue status, and abnormal appearance status of fruits and vegetables based on the odor data and the surface image. The frequency adjustment module is used to adjust the environmental acquisition frequency, the electronic nose acquisition frequency, and the visual acquisition frequency according to the internal and external decay state, the pesticide residue state, and / or the abnormal appearance state. A blockchain storage module is used to store the environmental data, odor data and surface images collected multiple times in the blockchain ledger based on the internal and external decay status, the pesticide residue status and / or the abnormal appearance status. An alarm module is used to determine whether to generate alarm information including environmental data, odor data and surface image based on the internal and external decay status, pesticide residue status and abnormal appearance status, and to send the alarm information through a blockchain smart contract; The alarm module is used to send alarm information through a blockchain smart contract when the environmental data exceeds an environmental threshold, wherein the environmental threshold is determined based on the internal and external decay status, the pesticide residue status, and / or the abnormal appearance status. The blockchain storage module is used to calculate the average of multiple collected environmental data as the standard value of the environmental data, calculate the average of multiple collected odor data as the standard value of the odor data, generate a fused image from multiple surface images through an alpha fusion model, and store the standard value of the environmental data, the standard value of the odor data, and the fused image in the ledger of the blockchain. The state determination module includes a support vector machine model, a neural network model, and a comprehensive state determination unit; The support vector machine model is used to identify odor decay characteristics based on the odor data; The neural network model is used to identify surface decay features based on the surface image; The comprehensive state determination unit is used to determine the internal and external decay state based on the odor decay characteristics and the surface decay characteristics. The training process of the support vector machine model includes: 45-dimensional basic pattern recognition library obtained by preprocessing and feature extraction of odor data with category labels; feature normalization processing; selection of the kernel function of the support vector machine; selection of radial basis kernel function support vector machine; and finding the optimal parameters of the kernel function using the grid method. The category labels include pesticide categories such as imidacloprid, imidacloprid, benzyl dichlorvos, soybean seed dressing agents, and organophosphorus pesticides, as well as rot categories such as acidic odor and foul odor. Specifically, when the neural network model determines that the fruits and vegetables do not exhibit rotten characteristics, but have at least two of the following characteristics: dented features, discoloration features, and deformation features, the fruits and vegetables are judged to be in an abnormal appearance state.

2. The digital management system for logistics services according to claim 1, characterized in that, The internal and external decay states include simultaneous internal and external decay, internal decay, external decay, and causes of decay. The causes of decay include physical decay, microbial decay, and oxidative decay. The frequency adjustment module is used to increase the environmental acquisition frequency to a first environmental dynamic frequency when the comprehensive state determination unit determines that the internal and external decay state is a state of simultaneous internal and external decay or an external decay state, and the cause of decay is microbial decay or oxidative decay. The frequency adjustment module is also used to increase the electronic nose collection frequency by a first odor dynamic frequency when the comprehensive state determination unit determines that the internal and external decay state is an internal decay state and the cause of decay is microbial decay or oxidative decay. The frequency adjustment module is further configured to increase the visual acquisition frequency by a first visual dynamic frequency when the comprehensive state determination unit determines that the cause of decay is physical decay.

3. The digital management system for logistics services according to claim 2, characterized in that, The internal and external decay conditions also include decay levels; The frequency adjustment module is used to adjust the first environmental dynamic frequency, the first odor dynamic frequency, or the first visual dynamic frequency according to the duration of the change in the decay level.

4. The digital management system for logistics services according to claim 3, characterized in that, The pesticide residue status includes the pesticide residue level, and the abnormal appearance status includes the abnormal appearance level. The frequency adjustment module is also used to increase the electronic nose acquisition frequency by a second odor dynamic frequency and the environmental acquisition frequency by a second environmental dynamic frequency when the support vector machine model identifies the pesticide residue status. The second odor dynamic frequency and the second environmental dynamic frequency are respectively related to the duration of the change in the pesticide residue level. The frequency adjustment module is further configured to increase the visual acquisition frequency by a second visual dynamic frequency and the environmental acquisition frequency by a third environmental dynamic frequency when the neural network model identifies the abnormal appearance state. The second visual dynamic frequency and the third environmental dynamic frequency are respectively related to the duration of the change in the abnormal appearance level.

5. The digital management system for logistics services according to claim 1, characterized in that, The blockchain storage module is used to store the standard values ​​of multiple environmental data, the standard values ​​of multiple odor data, and the fused image generated by the image fusion algorithm of multiple surface images in the ledger of the blockchain.

6. The digital management system for logistics services according to any one of claims 1 to 5, characterized in that, The alarm information includes the internal and external rot status, the pesticide residue status, and the abnormal appearance status, as well as the correlation determination with the environmental data.

7. The digital management system for logistics services according to claim 6, characterized in that, The blockchain ledger is synchronously set up within the systems of regulatory agencies and enterprises; A blockchain storage module is used to synchronously send the alarm information to the systems of the regulatory agencies and enterprises.

Images