Real-time monitoring and management system and method for tobacco leaf transportation

Abstract

The application provides a tobacco leaf transportation real-time monitoring and management system and method, the system comprising an environmental sensor array, a tobacco leaf transportation state analysis model and a state analysis module. The environmental sensor collects and pre-processes transportation environment data to generate standardized environmental parameters. The tobacco leaf transportation state analysis model is composed of three sequentially connected data processing modules, each module containing a data fusion unit and multiple multi-dimensional parameter analysis units. The data fusion unit generates fusion features from the input data and passes them to the multi-dimensional parameter analysis units. The multi-dimensional parameter analysis units generate weighted feature data through an environmental parameter attention module and then output enhanced feature data through a data enhancement module. The state analysis module integrates and maps the model output data to generate transportation state evaluation results. The application can achieve high-precision dynamic monitoring and state evaluation of environmental abnormalities during tobacco leaf transportation, improving the intelligent level of transportation quality control.

Description

Technical Field

[0001] This invention relates to the field of intelligent logistics monitoring technology, and more specifically, to a real-time monitoring and management system and method for tobacco leaf transportation. Background Technology

[0002] In the tobacco transportation sector, existing monitoring methods largely rely on simple sensor data collection, providing only basic environmental parameters such as temperature and humidity, lacking in-depth analysis and comprehensive processing of this data. While these traditional methods can reflect the basic conditions of the transportation environment to some extent, their inability to effectively integrate multi-dimensional data and capture spatiotemporal correlation characteristics leads to inaccurate overall assessments of the transportation status. Furthermore, existing monitoring systems often neglect the dynamic changes in data during processing, making it difficult to adapt to the complex and ever-changing environmental conditions during transportation, thus failing to identify potential environmental anomalies in a timely and accurate manner.

[0003] In the process of implementing the embodiments of the present invention, the inventors have discovered that the prior art has at least the following problems or defects: it is impossible to fully perceive the spatiotemporal correlation characteristics of the transportation environment, the data processing accuracy is insufficient, it is difficult to achieve real-time and accurate assessment of the transportation status, and there is a lack of effective model optimization mechanism, making it unable to adapt to complex and ever-changing transportation environment conditions. These problems have seriously affected the quality assurance and safety management in the process of tobacco leaf transportation. Summary of the Invention

[0004] This invention provides a real-time monitoring and management system and method for tobacco leaf transportation.

[0005] In a first aspect of the present invention, a real-time monitoring and management system for tobacco leaf transportation is provided, comprising:

[0006] An environmental sensor array configured to collect transportation environment data and output standardized environmental parameters;

[0007] The tobacco leaf transportation status analysis model consists of three sequentially connected data processing modules. Each data processing module contains a sequentially connected data fusion unit and at least two multi-dimensional parameter analysis units, wherein the data fusion unit of the first data processing module receives the standardized environmental parameters.

[0008] The data fusion unit is configured to perform feature fusion processing on the input data and output it to the corresponding multidimensional parameter analysis unit;

[0009] The multidimensional parameter analysis unit includes an environmental parameter attention module and a data augmentation module. The environmental parameter attention module is configured to generate feature vectors through multidimensional filtering, generate superimposed features after performing dynamic standardization and spatiotemporal correlation analysis on the feature vectors, generate aggregate parameters based on extreme value sampling and mean sampling, and generate weighted feature data through weight allocation.

[0010] The data augmentation module is configured to perform feature dimensionality reduction, multidimensional filtering, nonlinear activation, noise suppression, and feature dimensionality increase operations sequentially on the input data obtained after adding the weighted feature data and the input data of the multidimensional parameter analysis unit and standardizing the input data to generate augmented feature data; the output of the multidimensional parameter analysis unit is jointly determined by the weighted feature data and the augmented feature data, and serves as the input of the next processing unit;

[0011] The status analysis module is connected to the end-of-line data processing module and is configured to perform feature integration and decision mapping on the output data of the tobacco transportation status analysis model to generate transportation status assessment results.

[0012] Furthermore, the data fusion unit in the first data processing module is a basic data fusion unit, while the data fusion units in other data processing modules are multi-dimensional data fusion units;

[0013] The basic data fusion unit includes processing standardized environmental parameters using a first filtering network to obtain primary features, and performing data standardization processing on the primary features to obtain the output data of the basic data fusion unit.

[0014] The multidimensional data fusion unit includes performing a second filtering network and feature dimensionality reduction processing on the output data of the previous data processing module to obtain a first processing result and a second processing result; performing channel calibration processing on the first processing result to obtain a third processing result; and inputting the third processing result and the second processing result into the feature splicing module to obtain the output data of the multidimensional data fusion unit.

[0015] Furthermore, the multidimensional parameter analysis unit also includes a first data standardization layer, a first feedback network, a second data standardization layer, a second feedback network, and a feature normalization layer;

[0016] The first input data is processed by the first data standardization layer and then processed to obtain the second input data from the multidimensional parameter analysis unit.

[0017] The environmental parameter attention module processes the second input data to obtain weighted feature data;

[0018] The weighted feature data is added to the first input data through the first feedback network, and the result is added to the second data standardization layer.

[0019] The input data of this layer is processed using the second data normalization layer to obtain the third input data;

[0020] The data augmentation module processes the third input data to obtain augmented feature data;

[0021] The enhanced feature data is added to the summation result through the second feedback network and then added to the feature normalization layer to obtain the output data of the multidimensional parameter analysis unit.

[0022] Furthermore, the tobacco leaf transportation status analysis model includes three data processing modules connected in sequence. The second data processing module has three multi-dimensional parameter analysis units connected in sequence, and the remaining data processing modules have two multi-dimensional parameter analysis units connected in sequence.

[0023] Furthermore, the state analysis module includes a parameter integration layer and a decision mapping layer. The parameter integration layer performs 1×1 parameter aggregation on the input data of the state analysis module along the feature dimension. The decision mapping layer performs nonlinear mapping on the output of the parameter integration layer to obtain the transportation state assessment result.

[0024] Furthermore, the constructed environmental anomaly discrimination loss function is used to optimize the tobacco leaf transportation status analysis model. The expression of the environmental anomaly discrimination loss function is as follows:

[0025]

[0026]

[0027] in, Indicates the first Environmental anomalies in each sample The model represents the samples Correctly identified as the actual state Confidence level, The model represents the samples Predicted as The characteristic score of the state Indicates the number of categories of transportation environment status. Indicates the category to which sample i belongs. frequency factor, , Distributed parameters representing environmental parameters This indicates that the model predicts the value of sample i as follows: The feature score, where e is the natural constant.

[0028] Furthermore, the window size of the multi-dimensional filtering in the environmental parameter attention module is 5×5×3, a 7×7×5 three-dimensional convolution kernel is used in the spatiotemporal correlation analysis, the dilation interval is 2, a fully connected layer is used for the linear transformation of the aggregation parameters, the window size of the multi-dimensional filtering in the data augmentation module is 3×3×2, and the ReLU function is applied to the nonlinear activation layer.

[0029] Furthermore, the feature vectors are subjected to dynamic standardization and spatiotemporal correlation analysis, and the results of the two processes are superimposed to obtain the superimposed feature:

[0030]

[0031] in, Indicates superposition features, This represents the multi-dimensional filtering results of the environmental parameter attention module. This indicates dynamic standardization processing. This indicates dilated convolution processing. This indicates an overlay operation.

[0032] Furthermore, the kernel size of the first filtering network is 5×5×2, the kernel size of the second filtering network is 5×5×1, and the sliding step size is 1; the feature dimensionality reduction process uses a 2×2×1 convolution kernel with a step size of 1.

[0033] In a second aspect of the present invention, a method for real-time monitoring and management of tobacco leaf transportation is provided, comprising:

[0034] Step 1: Collect transportation environment data and preprocess it to obtain standardized environmental parameters;

[0035] Step 2: Use the basic data fusion unit to perform feature fusion on the standardized environmental parameters to obtain the first output parameter;

[0036] Step 3: Input the first output parameter into multiple sequentially connected multidimensional parameter analysis units to obtain the second output parameter, which serves as the input data for the multidimensional data fusion unit; capture the spatiotemporal correlation features of the transportation environment through the multidimensional perception capability of the multidimensional parameter analysis unit;

[0037] Step 4: Use the multi-dimensional data fusion unit to perform multi-scale feature fusion processing on the input data of this unit to obtain the third output parameter;

[0038] Step 5: Input the third output parameter into multiple sequentially connected multidimensional parameter analysis units to obtain the fourth output parameter, and input it into the multidimensional data fusion unit;

[0039] Step Six: Repeat Steps Four and Five until the preset number of iterations is reached, then output the fifth output parameter;

[0040] Step 7: Input the fifth output parameter into the status analysis module to obtain the transportation environment status assessment result.

[0041] The embodiments of the present invention have at least the following beneficial effects: the real-time monitoring and management system and method for tobacco leaf transportation can improve the accuracy of environmental monitoring during tobacco leaf transportation. By collecting transportation environment data through an environmental sensor array and preprocessing it to obtain standardized environmental parameters, and then performing in-depth analysis through multiple data processing modules in the tobacco leaf transportation status analysis model, the spatiotemporal correlation characteristics and multi-scale characteristics of the transportation environment can be accurately captured. This multi-dimensional and multi-level data processing method ensures a comprehensive perception of the transportation environment status, providing accurate and reliable data support for subsequent status assessment, thereby better ensuring that the quality of tobacco leaves is not adversely affected by environmental factors during transportation.

[0042] Furthermore, the system enables real-time assessment and optimization of tobacco leaf transportation status. The status analysis module can quickly derive transportation status assessment results based on the output data of the tobacco leaf transportation status analysis model. It optimizes the model using a constructed environmental anomaly discrimination loss function, further improving the model's ability to detect abnormal environmental conditions. This real-time monitoring and assessment mechanism can promptly identify potential environmental anomalies during transportation, allowing relevant personnel to quickly take appropriate measures for adjustment and handling, effectively reducing transportation risks and improving the overall efficiency and safety of tobacco leaf transportation. Attached Figure Description

[0043] The above and other objects, features, and advantages of exemplary embodiments of the present invention will become readily apparent from the following detailed description taken in conjunction with the accompanying drawings. Several embodiments of the invention are illustrated in the drawings by way of example and not limitation, wherein:

[0044] Figure 1 This is a schematic diagram of the structure of a real-time monitoring and management system for tobacco leaf transportation provided in an embodiment of the present invention;

[0045] Figure 2 This is a flowchart illustrating a real-time monitoring and management method for tobacco leaf transportation provided in an embodiment of the present invention.

[0046] Figure 3 A schematic diagram of the structure of an electronic device according to an embodiment of the present invention is shown. Detailed Implementation

[0047] The principles and spirit of the invention will now be described with reference to several exemplary embodiments. It should be understood that these embodiments are provided merely to enable those skilled in the art to better understand and implement the invention, and are not intended to limit the scope of the invention in any way. Rather, these embodiments are provided to make the invention more thorough and complete, and to fully convey the scope of the invention to those skilled in the art.

[0048] Those skilled in the art will recognize that embodiments of the present invention can be implemented as a system, apparatus, device, method, or computer program product. Therefore, the present invention can be specifically implemented in the following forms: entirely hardware, entirely software (including firmware, resident software, microcode, etc.), or a combination of hardware and software.

[0049] It should be noted that the number of any elements in the accompanying drawings is for illustrative purposes only and not as a limitation, and any naming is for distinction only and has no limiting meaning.

[0050] The following is for reference. Figure 1 , Figure 1 This is a schematic diagram of the structure of a real-time monitoring and management system for tobacco leaf transportation provided in an embodiment of the present invention. Figure 1 As shown, a real-time monitoring and management system for tobacco leaf transportation includes:

[0051] Environmental sensor array 101 is configured to collect transportation environment data and output standardized environmental parameters;

[0052] The tobacco leaf transportation status analysis model 102 consists of three sequentially connected data processing modules. Each data processing module includes a sequentially connected data fusion unit and at least two multi-dimensional parameter analysis units, wherein the data fusion unit of the first data processing module receives the standardized environmental parameters.

[0053] The data fusion unit is configured to perform feature fusion processing on the input data and output it to the corresponding multidimensional parameter analysis unit;

[0054] The multidimensional parameter analysis unit includes an environmental parameter attention module and a data augmentation module. The environmental parameter attention module is configured to generate feature vectors through multidimensional filtering, generate superimposed features after performing dynamic standardization and spatiotemporal correlation analysis on the feature vectors, generate aggregate parameters based on extreme value sampling and mean sampling, and generate weighted feature data through weight allocation.

[0055] The data augmentation module is configured to perform feature dimensionality reduction, multidimensional filtering, nonlinear activation, noise suppression, and feature dimensionality increase operations sequentially on the input data obtained after adding the weighted feature data and the input data of the multidimensional parameter analysis unit and standardizing the input data to generate augmented feature data; the output of the multidimensional parameter analysis unit is jointly determined by the weighted feature data and the augmented feature data, and serves as the input of the next processing unit;

[0056] The status analysis module 103 is connected to the end data processing module and is configured to perform feature integration and decision mapping on the output data of the tobacco transportation status analysis model to generate transportation status assessment results.

[0057] Specifically, the environmental sensor array 101 collects transportation environment data and preprocesses it to obtain standardized environmental parameters.

[0058] The tobacco leaf transportation status analysis model 102 includes three data processing modules connected in sequence. The standardized environmental parameters are input to the front end of the first data processing module of the tobacco leaf transportation status analysis model. The output of any data processing module will be used as the input of the next data processing module connected to it. Each data processing module includes a data fusion unit and multiple multi-dimensional parameter analysis units connected in sequence.

[0059] The data fusion unit generates fusion feature data from the received data and inputs the processed data into the adjacent multidimensional parameter analysis unit;

[0060] The multidimensional parameter analysis unit includes an environmental parameter attention module and a data augmentation module. The environmental parameter attention module performs a first multidimensional filtering process on the input data to obtain a feature vector; performs dynamic standardization and spatiotemporal correlation analysis on the feature vector, and superimposes the two processing results to obtain a superimposed feature; performs extreme value sampling and mean sampling on the superimposed feature, and merges the two sampling results to obtain an aggregate parameter; performs linear transformation and activation function processing on the aggregate parameter to obtain an attention parameter, and evenly distributes the attention parameter in the dimensional space to obtain a first weight coefficient and a second weight coefficient; performs dimensional segmentation on the superimposed feature to obtain a first superimposed feature and a second superimposed feature; and processes and combines the first superimposed feature and the second superimposed feature based on the first weight coefficient and the second weight coefficient to obtain weighted feature data.

[0061] The data augmentation module includes sequentially processing the input data of the module through feature dimensionality reduction, second multi-dimensional filtering, nonlinear activation layer, noise suppression layer, and feature dimensionality increase to obtain augmented feature data;

[0062] The output data of the multidimensional parameter analysis unit is obtained based on the weighted feature data and the enhanced feature data, and is used as the input data of the adjacent connected multidimensional parameter analysis units;

[0063] The status analysis module 103 is connected to the end of the tobacco transportation status analysis model. The output data of the tobacco transportation status analysis model is input into the status analysis module to obtain the transportation status assessment result.

[0064] The real-time monitoring and management system for tobacco leaf transportation of this invention includes an environmental sensor array, a tobacco leaf transportation status analysis model, and a status analysis module. The environmental sensor array collects and preprocesses transportation environment data to obtain standardized environmental parameters. These parameters are processed values ​​that can be directly used for subsequent analysis, reflecting various characteristics of the transportation environment, such as temperature, humidity, and light intensity. The tobacco leaf transportation status analysis model consists of multiple data processing modules, each containing a data fusion unit and a multi-dimensional parameter analysis unit for in-depth processing and feature extraction of the input data. The data fusion unit integrates data from different sources to generate fused feature data, better reflecting the overall status of the transportation environment. The multi-dimensional parameter analysis unit further analyzes the data to extract more valuable feature information. The status analysis module, based on the output data of the tobacco leaf transportation status analysis model, obtains a transportation status assessment result to determine whether the transportation process is normal and whether there are potential risks.

[0065] Specifically, the standardized environmental parameters obtained after preprocessing the transportation environment data collected by the environmental sensor array are numerical values ​​converted from the raw data into a unified format and dimension. For example, temperature data is standardized to values ​​within a Celsius range, and humidity data is standardized to percentage form. The first data processing module in the tobacco transportation status analysis model receives these standardized environmental parameters and inputs them into the data fusion unit. The basic data fusion unit in the data fusion unit processes the standardized environmental parameters using a first filtering network. This filtering network is a mathematical model used to remove noise from the data and extract useful primary features. Primary features refer to the set of data that can reflect the basic characteristics of the data after preliminary processing. Subsequently, the primary features are standardized to meet the requirements of subsequent processing modules. The environmental parameter attention module in the multidimensional parameter analysis unit performs multidimensional filtering on the input data. This is a method of analyzing data through multiple dimensions, which can extract the features of the data in different dimensions. Dynamic standardization processing refers to dynamically adjusting the standardization parameters of the data according to real-time changes in the data to better adapt to data changes. Spatiotemporal correlation analysis processing considers the correlation between data in time and space, and analyzes the changing patterns of data in different time and spatial locations. Parameter aggregation combines multiple feature parameters to obtain a comprehensive parameter representation. The data augmentation module enhances the feature representation capability of data through a series of operations such as feature dimensionality reduction, multi-dimensional filtering, nonlinear activation layers, noise suppression layers, and feature dimensionality enhancement, making it more suitable for subsequent analysis and processing.

[0066] Preferably, the construction process of the tobacco leaf transportation status analysis model is as follows: First, based on the types of environmental parameters that need to be monitored during tobacco leaf transportation, appropriate sensors are selected to form an environmental sensor array, and these arrays are calibrated and standardized to ensure the accuracy and reliability of the collected data. Then, the specific structure and parameter settings of the data fusion unit and multidimensional parameter analysis unit in the data processing module are designed. For example, the first filtering network can adopt a convolutional neural network structure, and its kernel size can be selected according to the characteristics of the actual data, such as 5×5×2, to extract local features of the data. The kernel size of the second filtering network can be set to 5×5×1 to further refine the features. Feature dimensionality reduction can be achieved using methods such as principal component analysis to reduce the dimensionality of the data while retaining the main feature information. In the multidimensional parameter analysis unit, the multidimensional filtering processing of the environmental parameter attention module can extract features of the data in different dimensions by setting an appropriate window size, such as 5×5×3. Dynamic standardization can be adjusted in real time according to the mean and standard deviation of the data to ensure data stability. Spatiotemporal correlation analysis can use a three-dimensional convolutional kernel, such as 7×7×5, with a dilation interval of 2, to capture the temporal and spatial correlation of the data. Linear transformation of aggregated parameters can be achieved through fully connected layers, integrating multiple feature parameters into a comprehensive parameter representation. The nonlinear activation layer in the data augmentation module can employ the ReLU function to enhance the nonlinear characteristics of the data. Finally, the parameter integration layer and decision mapping layer in the state analysis module can be designed according to actual needs. For example, the parameter integration layer can perform 1×1 parameter aggregation on the feature dimension of the input data, and the decision mapping layer can use nonlinear mapping methods, such as the softmax function, to map the aggregated parameters to the transportation state assessment results. During model optimization, the constructed environmental anomaly discrimination loss function can be used to adjust the model's parameters, making the model more capable of discriminating against environmental anomalies.

[0067] In some embodiments, the data fusion unit in the first data processing module is a basic data fusion unit, and the data fusion units in other data processing modules are multi-dimensional data fusion units;

[0068] The basic data fusion unit includes processing standardized environmental parameters using a first filtering network to obtain primary features, and performing data standardization processing on the primary features to obtain the output data of the basic data fusion unit.

[0069] The multidimensional data fusion unit includes performing a second filtering network and feature dimensionality reduction processing on the output data of the previous data processing module to obtain a first processing result and a second processing result; performing channel calibration processing on the first processing result to obtain a third processing result; and inputting the third processing result and the second processing result into the feature splicing module to obtain the output data of the multidimensional data fusion unit.

[0070] It should be noted that the real-time monitoring and management system for tobacco leaf transportation mentioned in this invention has a clear hierarchical division in its data processing module design. The data fusion unit in the first data processing module is called the basic data fusion unit, while the data fusion units in other data processing modules are multi-dimensional data fusion units. The main function of the basic data fusion unit is to perform preliminary processing on the standardized environmental parameters of the initial input, generating primary feature data that can be used for subsequent analysis. The multi-dimensional data fusion unit, based on the basic data fusion unit, further integrates and optimizes the data from multiple dimensions to extract more representative and accurate feature information, providing richer data support for subsequent transportation status assessment. This hierarchical data fusion approach can better adapt to the data processing needs at different stages, improving the overall data processing efficiency and accuracy of the system.

[0071] Specifically, the first filtering network in the basic data fusion unit is a mathematical model used for preliminary processing of standardized environmental parameters. Its function is to remove noise from the data and extract primary features. Primary features refer to the data set that reflects the basic characteristics of the data after preliminary processing, such as the temperature change trend extracted from temperature data. Data standardization converts the primary feature data into a unified format and dimension so that subsequent processing modules can better process the data. The second filtering network and feature dimensionality reduction in the multidimensional data fusion unit are key steps for further data processing. The second filtering network is used to filter the output data of the previous data processing module again to extract more refined features. Feature dimensionality reduction removes redundant information by reducing the dimensionality of the data while retaining the main feature information, for example, by using methods such as principal component analysis. Channel calibration calibrates different channels of the data to ensure data consistency and accuracy. The feature stitching module stitches the data after channel calibration and feature dimensionality reduction to generate the output data of the multidimensional data fusion unit, providing richer input data for the subsequent multidimensional parameter analysis unit.

[0072] Preferably, the first filtering network in the basic data fusion unit can adopt a convolutional neural network structure, and its kernel size can be selected according to the characteristics of the actual data, for example, set to 5×5×2, to extract local features of the data. Data standardization can be performed using the Z-score standardization method, which converts the data into standardized data with a mean of 0 and a standard deviation of 1 by calculating the mean and standard deviation of the data. The kernel size of the second filtering network in the multidimensional data fusion unit can be set to 5×5×1 to further refine the features. Feature dimensionality reduction can be performed using principal component analysis, which extracts the main components of the data by calculating the covariance matrix and eigenvalues ​​of the data, thereby reducing the dimensionality of the data. Channel calibration can be performed by calculating the differences between different channels and adjusting them to ensure data consistency. The feature stitching module can stitch the data after channel calibration and feature dimensionality reduction according to certain rules, such as stitching according to the time order or spatial location of the data, to generate the output data of the multidimensional data fusion unit. This hierarchical data fusion method can better adapt to the data processing needs of different stages, improve the data processing efficiency and accuracy of the entire system, and provide more reliable data support for real-time monitoring and evaluation of tobacco transportation status.

[0073] In some embodiments, the multidimensional parameter analysis unit further includes a first data standardization layer, a first feedback network, a second data standardization layer, a second feedback network, and a feature normalization layer;

[0074] The first input data is processed by the first data standardization layer and then processed to obtain the second input data from the multidimensional parameter analysis unit.

[0075] The environmental parameter attention module processes the second input data to obtain weighted feature data;

[0076] The weighted feature data is added to the first input data through the first feedback network, and the result is added to the second data standardization layer.

[0077] The input data of this layer is processed using the second data normalization layer to obtain the third input data;

[0078] The data augmentation module processes the third input data to obtain augmented feature data;

[0079] The enhanced feature data is added to the summation result through the second feedback network and then added to the feature normalization layer to obtain the output data of the multidimensional parameter analysis unit.

[0080] It should be noted that the multidimensional parameter analysis unit mentioned in this invention introduces several key modules in the data processing process, including a first data standardization layer, a first feedback network, a second data standardization layer, a second feedback network, and a feature normalization layer. These modules are designed to further optimize the data processing flow and improve the accuracy and stability of feature extraction. The first data standardization layer performs preliminary standardization on the input data so that subsequent modules can better process the data. The environmental parameter attention module uses weighted feature data to highlight important features, enhancing the model's ability to perceive key information. The first and second feedback networks further enhance the feature representation capability of the data by combining the processed data with the original data. The feature normalization layer normalizes the final output data, ensuring the consistency and comparability of the output results. This multi-layered feedback and normalization approach effectively improves the robustness and accuracy of the system.

[0081] Specifically, the first data standardization layer is a module that standardizes the input data, converting it into a uniform format and unit to facilitate better processing by subsequent modules. For example, Z-score standardization can be used to adjust the mean to 0 and the standard deviation to 1. The environmental parameter attention module highlights important features by weighting the input data. Weighted feature data assigns different weights based on the importance of the data, thereby enhancing the model's ability to perceive key information. The first and second feedback networks combine the processed data with the original data, further enhancing the data's feature representation capabilities through a feedback mechanism. Feedback networks can be implemented using methods such as weighted summation to fuse the processed and original data. The feature normalization layer normalizes the final output data, ensuring consistency and comparability of the output results. Normalization can use min-max normalization to adjust the data range to between 0 and 1. These modules enable the multidimensional parameter analysis unit to process data more effectively and extract more valuable feature information.

[0082] Preferably, the construction process of the multidimensional parameter analysis unit can be further refined. First, the first data standardization layer can use the Z-score standardization method to convert the data into standardized data with a mean of 0 and a standard deviation of 1 by calculating the mean and standard deviation of the data. When processing data, the environmental parameter attention module can dynamically adjust the weights according to the importance and relevance of the data, for example, by calculating the confidence level and correlation coefficient of the data to determine the weights. The first and second feedback networks can be implemented by weighted summation to fuse the processed data with the original data, enhancing the feature representation capability of the data. The weights of the feedback networks can be dynamically adjusted according to the feature importance of the data to ensure the accuracy and stability of the data. The feature normalization layer can use the min-max normalization method to adjust the data range to between 0 and 1, ensuring the consistency and comparability of the output results. Through these optimization steps, the multidimensional parameter analysis unit can process data more effectively, extract more valuable feature information, and provide more reliable data support for the real-time monitoring and evaluation of tobacco transportation status.

[0083] In some embodiments, the tobacco transport status analysis model includes three data processing modules connected in sequence. The second data processing module has three multi-dimensional parameter analysis units connected in sequence, and the remaining data processing modules have two multi-dimensional parameter analysis units connected in sequence.

[0084] It should be noted that the tobacco leaf transportation status analysis model mentioned in this invention consists of three sequentially connected data processing modules. The second data processing module contains three multi-dimensional parameter analysis units, while the remaining data processing modules each contain two sequentially connected multi-dimensional parameter analysis units. This structural design is intended for more detailed data processing. By increasing the number of analysis units in the second module, it is possible to more deeply mine the feature information in the data, thereby improving the accuracy and reliability of the tobacco leaf transportation status assessment. The data processing module is the core part of the system, used for multi-level processing and feature extraction of the input standardized environmental parameters. The multi-dimensional parameter analysis units are modules within each module used for specific data analysis and processing. Through the collaborative work of multiple units, complex relationships and patterns in the data can be better captured.

[0085] Specifically, the data processing module is the basic building block of the tobacco transport status analysis model. Each module contains a data fusion unit and a multidimensional parameter analysis unit. The data fusion unit in the first data processing module is the basic data fusion unit, primarily responsible for the initial processing of the input standardized environmental parameters to generate primary feature data. The multidimensional parameter analysis unit further analyzes and processes this primary feature data, extracting more valuable feature information through components such as the environmental parameter attention module and the data augmentation module. The second data processing module contains three multidimensional parameter analysis units, meaning that at this stage, the data undergoes three detailed analyses and processing steps, enabling a deeper extraction of feature information from the data. The remaining data processing modules each contain two multidimensional parameter analysis units. This design ensures both the depth of data processing and avoids unnecessary computational complexity. Each multidimensional parameter analysis unit includes multiple processing steps, such as feature dimensionality reduction, multidimensional filtering, and nonlinear activation. These steps work together to ensure that the data is effectively processed and optimized after passing through each unit.

[0086] Preferably, the construction process of the tobacco leaf transportation status analysis model can be further refined. First, based on the types of environmental parameters that need to be monitored during tobacco leaf transportation, appropriate sensors are selected to form an environmental sensor array, which is then calibrated and standardized to ensure the accuracy and reliability of the collected data. Then, the specific structure and parameter settings of the data processing module are designed. For example, the basic data fusion unit of the first data processing module can adopt a convolutional neural network structure, with its kernel size selected according to the characteristics of the actual data, such as 5×5×2, to extract local features of the data. In the second data processing module, the three multi-dimensional parameter analysis units can be set with different parameters to adapt to the data processing needs at different stages. For example, the first multi-dimensional parameter analysis unit can focus on extracting short-term features of the data, the second unit can analyze the medium-term trend of data changes, and the third unit can explore the long-term correlation of the data. The environmental parameter attention module in each multi-dimensional parameter analysis unit can determine the weights by calculating the confidence level and correlation coefficient of the data, highlighting important features. The data augmentation module can enhance the feature representation capability of the data through a series of operations such as feature dimensionality reduction, multi-dimensional filtering, nonlinear activation layers, noise suppression layers, and feature dimensionality enhancement. Through this hierarchical and multi-stage data processing approach, the tobacco transportation status analysis model can process data more effectively, extract more valuable feature information, and provide more reliable data support for real-time monitoring and evaluation of tobacco transportation status.

[0087] In some embodiments, the state analysis module includes a parameter integration layer and a decision mapping layer. The parameter integration layer performs 1×1 parameter aggregation on the input data of the state analysis module along the feature dimension. The decision mapping layer performs nonlinear mapping on the output of the parameter integration layer to obtain the transportation state assessment result.

[0088] It should be noted that the status analysis module is a key component of the real-time monitoring and management system for tobacco transportation. Its main function is to perform a final status assessment on the data processed by the multi-level data processing modules. The status analysis module consists of a parameter integration layer and a decision mapping layer. The parameter integration layer aggregates the input data along the feature dimensions to extract more representative feature information. The decision mapping layer then maps this aggregated feature information to specific transportation status assessment results, thereby achieving real-time monitoring and assessment of the tobacco transportation status. This modular design effectively improves the system's decision-making efficiency and accuracy, ensuring the reliable quality of tobacco during transportation.

[0089] Specifically, the parameter integration layer and the decision mapping layer are the two core components of the state analysis module. The parameter integration layer performs 1×1 parameter aggregation on the input data along the feature dimension. This 1×1 aggregation refers to compressing and integrating multi-dimensional feature data along the feature dimension using a special aggregation method to generate a more representative low-dimensional feature vector. This aggregation method effectively reduces data redundancy while retaining key information, enabling the subsequent decision mapping layer to perform state assessment more efficiently. The decision mapping layer then uses a nonlinear mapping method to map the low-dimensional feature vector output by the parameter integration layer to specific transportation state assessment results. Nonlinear mapping is a process of transforming input data into output results through complex nonlinear functions, which can better capture complex relationships and patterns in the data, thereby improving the accuracy of state assessment. In practical applications, nonlinear mapping can be implemented in various ways, such as using activation functions in neural networks or complex mathematical models.

[0090] Preferably, the construction and optimization process of the state analysis module can be further refined. In the parameter integration layer, 1×1 convolution operations from convolutional neural networks can be used to achieve parameter aggregation. 1×1 convolution is a special type of convolution operation that can perform weighted summation of feature channels without changing the data space dimension, thereby achieving feature aggregation and compression. For example, when the input data is a multi-channel feature map, 1×1 convolution can weightedly merge the features of these channels to generate a new feature representation. In the decision mapping layer, the softmax function can be used to achieve nonlinear mapping. The softmax function is a commonly used activation function that can map the input feature vector to a probability distribution, making the output results more interpretable. For example, for the evaluation of tobacco transportation status, the softmax function can map the input feature vector to a probability distribution of normal, abnormal, and other states, thereby achieving accurate classification of transportation status. Furthermore, to further improve the performance of the state analysis module, regularization techniques, such as L2 regularization, can be introduced to prevent model overfitting and improve the model's generalization ability. Through these optimization measures, the status analysis module can process input data more effectively, achieve accurate assessment of tobacco transportation status, and provide reliable technical support for tobacco transportation management.

[0091] In some embodiments, the constructed environmental anomaly discrimination loss function is used to optimize the tobacco leaf transportation status analysis model. The expression of the environmental anomaly discrimination loss function is as follows:

[0092]

[0093] in, Indicates the first Environmental anomalies in each sample The model represents the samples Correctly identified as the actual state Confidence level, The model represents the samples Predicted as The characteristic score of the state Indicates the number of categories of transportation environment status. Indicates the category to which sample i belongs. frequency factor, , express Number of samples per category express The number of samples in the category to which the sample belongs. , Distributed parameters representing environmental parameters This indicates that the model predicts the value of sample i as follows: The feature score, where e is the natural constant.

[0094] It should be noted that the environmental anomaly discrimination loss function mentioned in this invention is a key technical means for optimizing the tobacco leaf transportation status analysis model. This loss function evaluates and adjusts the model's output by comprehensively considering multiple factors, including the environmental anomaly factor, confidence level, feature score, and frequency factor of the state category. Its purpose is to enable the model to better learn the characteristic patterns of environmental anomalies during training, thereby improving its ability to identify abnormal situations and ultimately enhancing the reliability and accuracy of the entire monitoring system. The environmental anomaly factor reflects the probability of anomalies occurring in the transportation environment, the confidence level represents the model's certainty in classifying the sample, the feature score measures the characteristic performance of the sample under different states, and the frequency factor is used to balance the weights of different categories of samples during training.

[0095] Specifically, the environmental anomaly factor in the environmental anomaly discrimination loss function is calculated using parameters such as sample confidence and state category frequency factor. Confidence refers to the model's certainty that a sample is correctly classified as the actual state; it reflects the model's accuracy in classifying samples. Feature score represents the model's feature performance when a sample is predicted to be in a certain state, measuring the feature strength of a sample under different states. The state category frequency factor is related to the number of samples in each state category and the total number of samples; it is used to adjust the weights of samples from different categories during training, avoiding model bias towards the majority class due to class imbalance. Through the comprehensive calculation of these parameters, the loss function can comprehensively evaluate the model's output and guide the model to continuously optimize during training, improving its ability to discriminate environmental anomalies.

[0096] Preferably, the process of using the environmental anomaly discrimination loss function in constructing and optimizing the tobacco transportation status analysis model can be further refined. First, the model's input data includes standardized environmental parameters collected and preprocessed from an environmental sensor array. These parameters are processed by a multi-level data processing module to generate feature vectors for status assessment. During model training, the environmental anomaly factor for each sample is calculated, and its confidence level and feature score are combined with the environmental anomaly discrimination loss function to evaluate the model's output. The calculation result of the loss function reflects the model's classification performance for the current sample, especially its ability to identify environmental anomalies. Based on the value of the loss function, the model adjusts its internal parameters through a backpropagation algorithm, continuously optimizing the model structure and improving the model's accuracy in identifying environmental anomalies. For example, during training, if the model has a low confidence level for a certain anomaly sample, the value of the loss function will increase accordingly, prompting the model to adjust its parameters and improve its ability to identify similar anomalies. Through this optimization process based on the environmental anomaly discrimination loss function, the tobacco transportation status analysis model can better adapt to complex transportation environments, accurately identify and warn of environmental anomalies, and provide strong support for the safety and quality assurance of tobacco transportation.

[0097] In some embodiments, the window size of the first multidimensional filtering process is 5×5×3, the spatiotemporal correlation analysis process uses a 7×7×5 three-dimensional convolution kernel with a dilation interval of 2, the linear transformation of the aggregation parameters uses a fully connected layer, the window size of the second multidimensional filtering process is 3×3×2, and the nonlinear activation layer applies the ReLU function.

[0098] It should be noted that the first multi-dimensional filtering and spatiotemporal correlation analysis processes in the multi-dimensional parameter analysis unit mentioned in this invention are key data processing steps. The first multi-dimensional filtering process is used to extract features of the data in multiple dimensions, while the spatiotemporal correlation analysis process is used to capture the correlation between the data in time and space. These processing steps can effectively enhance the model's ability to perceive data features, thereby improving the accuracy of the assessment of tobacco transportation status. By reasonably setting the window size of the filtering process and the convolution kernel parameters of the spatiotemporal correlation analysis, it is possible to better adapt to complex transportation environment data and ensure that the model can accurately identify potential anomalies.

[0099] Specifically, the first multi-dimensional filtering process uses a 5×5×3 window size, meaning that filtering operations are performed using window sizes of 5, 5, and 3 in the three dimensions, respectively. This window size selection effectively extracts feature information from local regions of the data while balancing computational efficiency and feature representation capability. In the spatiotemporal correlation analysis, a 7×7×5 three-dimensional convolutional kernel with a dilation interval of 2 is used. This setting captures long-distance correlations between data in time and space. The size of the three-dimensional convolutional kernel determines the coverage in the time and space dimensions, while the dilation interval expands the receptive field of the kernel, enabling the model to perceive a wider range of spatiotemporal information. Furthermore, a fully connected layer is used for the linear transformation of the aggregation parameters. This is a common neural network structure used to convert the input feature vector into a higher-level feature representation. By learning the linear relationship between input and output features, the fully connected layer can effectively integrate feature information from different dimensions. The second multi-dimensional filtering process uses a 3×3×2 window size. This smaller window size is suitable for extracting detailed features of the data, further enriching the model's ability to perceive the data. The nonlinear activation layer uses the ReLU function, a commonly used activation function that can introduce nonlinear factors, enabling the model to learn complex feature patterns.

[0100] Preferably, these operational steps can be further refined when constructing the multidimensional parameter analysis unit. For example, in the first multidimensional filtering process, a convolutional layer in a convolutional neural network can be used, with the input data being preprocessed standardized environmental parameters and the output being the filtered feature vector. In the spatiotemporal correlation analysis process, a three-dimensional convolutional operation with dilation interval can be used, with the input being the output feature vector of the first multidimensional filtering process and the output being the feature vector after spatiotemporal correlation analysis. For the linear transformation of the aggregation parameters, a fully connected layer can be constructed, with the input being the feature vector after spatiotemporal correlation analysis and the output being the feature vector after linear transformation. In the second multidimensional filtering process, a convolutional layer can also be used, with the input being the feature vector after linear transformation and the output being the feature vector after the second multidimensional filtering process. Finally, the ReLU function is applied in the nonlinear activation layer, with the input being the feature vector after the second multidimensional filtering process and the output being the feature vector after nonlinear activation. Through these specific construction steps and parameter settings, the multidimensional parameter analysis unit can effectively process the input data, extract spatiotemporally correlated feature information, and provide more accurate data support for subsequent tobacco transportation status assessment.

[0101] In some embodiments, the feature vectors are subjected to dynamic standardization and spatiotemporal correlation analysis, and the results of the two processes are superimposed to obtain superimposed features:

[0102]

[0103] in, This represents the result of the first multi-dimensional filtering process. This indicates dynamic standardization processing. This indicates dilated convolution processing. This indicates an overlay operation.

[0104] It should be noted that the dynamic standardization and spatiotemporal correlation analysis of feature vectors mentioned in this invention are important steps in the multidimensional parameter analysis unit. Their purpose is to better integrate and optimize data features, thereby improving the model's ability to perceive the state of the transportation environment. Dynamic standardization adjusts the data distribution according to real-time changes, making it more suitable for model processing; while spatiotemporal correlation analysis captures the correlation between data in time and space, further enhancing the expressive power of features. The combination of these two processing methods effectively improves the model's adaptability and accuracy in complex transportation environments.

[0105] Specifically, dynamic standardization is a method that adjusts the data distribution based on real-time changes. It transforms data into standardized data with a mean of zero and a standard deviation of one by calculating the mean and standard deviation. This process removes scale differences, making data from different dimensions comparable and improving model convergence speed. Spatiotemporal correlation analysis considers the temporal and spatial correlations of data. By using three-dimensional convolution kernels, it captures long-distance dependencies in time and space. For example, using a 7×7×5 three-dimensional convolution kernel with a dilation interval of 2 expands the receptive field, allowing the model to perceive broader spatiotemporal information. Feature overlay combines the results of two processing steps to generate richer feature representations. This operation preserves the details of the original data while introducing spatiotemporal correlation information, enhancing the expressive power of the features.

[0106] Preferably, the operational steps can be further refined when implementing dynamic standardization and spatiotemporal correlation analysis. For example, in dynamic standardization, the mean and standard deviation of the input data can be calculated in real time, and the data can be standardized based on these statistics. Specifically, for each input feature vector, its mean and standard deviation are first calculated, and then standardized by subtracting the mean and dividing by the standard deviation. In spatiotemporal correlation analysis, a three-dimensional convolutional layer can be constructed, with the input being the feature vector after dynamic standardization, the kernel size set to 7×7×5, and the dilation interval set to 2. With this setting, the convolutional layer can effectively capture the temporal and spatial correlations of the data. Feature overlay operations can be achieved through simple addition or concatenation operations, merging the feature vector after dynamic standardization with the feature vector after spatiotemporal correlation analysis to generate the final fused feature. This fused feature can better reflect the feature information of the data in different dimensions, providing richer data support for subsequent model processing.

[0107] In some embodiments, the kernel size of the first filtering network is 5×5×2, the kernel size of the second filtering network is 5×5×1, and the sliding step size is 1; the feature dimensionality reduction process uses a 2×2×1 convolution kernel with a step size of 1.

[0108] It should be noted that the first filtering network, the second filtering network, and the feature dimensionality reduction processing mentioned in this invention are key components of the basic data fusion unit and the multidimensional data fusion unit. The first and second filtering networks are used to filter the input data to extract useful features and remove noise. Feature dimensionality reduction processing reduces the dimensionality of the data while retaining the main feature information, thereby improving the efficiency of data processing and the performance of the model. By appropriately setting the convolution kernel size and sliding stride, it is possible to better adapt to the environmental data collected during tobacco leaf transportation, ensuring the accuracy and efficiency of data processing.

[0109] Specifically, the first filtering network uses a 5×5×2 convolutional kernel, meaning that filtering operations are performed using window sizes of 5, 5, and 2 in three dimensions, respectively. This size selection effectively extracts feature information from local regions of the data while balancing computational efficiency and feature representation capability. The second filtering network uses a 5×5×1 convolutional kernel with a stride of 1. This setting is suitable for further refining the data and extracting more detailed features. The feature dimensionality reduction process uses a 2×2×1 convolutional kernel with a stride of 1. This setting effectively reduces the dimensionality of the data, removes redundant information, and retains the main feature information. Through the reasonable setting of these parameters, it can be ensured that the data, after filtering and dimensionality reduction, better reflects the characteristics of the tobacco transportation environment, providing high-quality data support for subsequent data fusion and analysis.

[0110] Preferably, these operational steps can be further refined when constructing the basic data fusion unit and the multidimensional data fusion unit. For example, in the first filtering network, a convolutional layer in a convolutional neural network can be used. The input data is preprocessed standardized environmental parameters, and the output is the primary features after filtering. In the second filtering network, a convolutional layer is also used. The input is the output data of the previous data processing module, and the output is the feature data after further filtering. In the feature dimensionality reduction process, a convolutional layer can be constructed. The input is the feature data processed by the second filtering network, using a 2×2×1 convolutional kernel and a stride of 1. The output is the dimensionality-reduced feature data. Through these specific construction steps and parameter settings, the basic data fusion unit and the multidimensional data fusion unit can effectively process the input data, extract representative feature information, and provide more accurate data support for subsequent tobacco leaf transportation status assessment.

[0111] The above embodiments of the present invention have the following beneficial effects: The present invention can achieve comprehensive dynamic monitoring of the tobacco transportation environment through the collaborative work of an environmental sensor array and a multi-level data processing module. The hierarchical processing architecture of the basic data fusion unit and the multi-dimensional data fusion unit can construct multi-scale feature representations. Combined with the environmental parameter attention mechanism and data augmentation module in the multi-dimensional parameter analysis unit, it can effectively capture the spatiotemporal correlation features during transportation and improve the sensitivity of anomaly detection. Through the cascaded processing of the three data processing modules, transportation environment features can be gradually extracted and optimized. Finally, through parameter integration and decision mapping in the state analysis module, accurate transportation state assessment results can be output.

[0112] This system optimizes model performance through a specific environmental anomaly discrimination loss function. Combined with dynamic standardization and multi-dimensional filtering techniques, it enhances the model's adaptability to complex transportation environments. Different sized convolutional kernels can extract multi-granular features, while feature dimensionality reduction and expansion operations balance computational efficiency and feature representation capabilities. Introducing frequency factors and confidence weighting mechanisms alleviates the problem of imbalanced data distribution and improves the reliability of evaluation results. The modular design of the overall architecture allows for flexible configuration adjustments to meet the monitoring needs of different transportation scenarios.

[0113] like Figure 2 As shown in some embodiments, a method for real-time monitoring and management of tobacco leaf transportation includes:

[0114] Step 1: Collect transportation environment data and preprocess it to obtain standardized environmental parameters;

[0115] Step 2: Use the basic data fusion unit to perform feature fusion on the standardized environmental parameters to obtain the first output parameter;

[0116] Step 3: Input the first output parameter into multiple sequentially connected multidimensional parameter analysis units to obtain the second output parameter, which serves as the input data for the multidimensional data fusion unit; capture the spatiotemporal correlation features of the transportation environment through the multidimensional perception capability of the multidimensional parameter analysis unit;

[0117] Step 4: Use the multi-dimensional data fusion unit to perform multi-scale feature fusion processing on the input data of this unit to obtain the third output parameter;

[0118] Step 5: Input the third output parameter into multiple sequentially connected multidimensional parameter analysis units to obtain the fourth output parameter, and input it into the multidimensional data fusion unit;

[0119] Step Six: Repeat Steps Four and Five until the preset number of iterations is reached, then output the fifth output parameter;

[0120] Step 7: Input the fifth output parameter into the status analysis module to obtain the transportation environment status assessment result.

[0121] It is understandable that the steps and references recorded in this method for real-time monitoring and management of tobacco leaf transportation are... Figure 1 The modules described correspond to those in the real-time monitoring and management system for tobacco leaf transportation. Therefore, the modules, features, and beneficial effects described above for the real-time monitoring and management system for tobacco leaf transportation also apply to the real-time monitoring and management method for tobacco leaf transportation and the operations it includes, and will not be repeated here.

[0122] The following is for reference. Figure 3 The diagram illustrates a structural schematic of an electronic device 300 suitable for implementing some embodiments of the present invention. The electronic devices in some embodiments of the present invention may include, but are not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 3 The terminal device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments of the present invention.

[0123] like Figure 3As shown, the electronic device 300 may include a processing unit (e.g., a central processing unit, a graphics processing unit, etc.) 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage device 308 into a random access memory (RAM) 303. The RAM 303 also stores various programs and data required for the operation of the electronic device 300. The processing unit 301, ROM 302, and RAM 303 are interconnected via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0124] Typically, the following devices can be connected to I / O interface 305: input devices 306 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 307 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 308 including, for example, magnetic tapes, hard disks, etc.; and communication devices 309. Communication device 309 allows electronic device 300 to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 3 An electronic device 300 with various devices is shown; however, it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed alternatively. Figure 3 Each box shown can represent a device or multiple devices as needed.

[0125] Furthermore, the storage medium in the embodiments of this application stores program instructions capable of implementing all the above methods. These program instructions can be stored in the storage medium in the form of a software product, including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks, or terminal devices such as computers, servers, mobile phones, and tablets.

[0126] The above description is merely an explanation of some preferred embodiments of the present invention and the technical principles employed. Those skilled in the art should understand that the scope of the invention as described in the embodiments of the present invention is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described inventive concept. For example, technical solutions formed by substituting the above-described features with (but not limited to) technical features with similar functions disclosed in the embodiments of the present invention.

Claims

1. A real-time monitoring and management system for tobacco leaf transportation, characterized in that, include: An environmental sensor array configured to collect transportation environment data and output standardized environmental parameters; The tobacco leaf transportation status analysis model consists of three sequentially connected data processing modules. Each data processing module contains a sequentially connected data fusion unit and at least two multi-dimensional parameter analysis units, wherein the data fusion unit of the first data processing module receives the standardized environmental parameters. The data fusion unit is configured to perform feature fusion processing on the input data and output it to the corresponding multidimensional parameter analysis unit; The multidimensional parameter analysis unit includes an environmental parameter attention module and a data augmentation module. The environmental parameter attention module is configured to generate feature vectors through multidimensional filtering, generate superimposed features after performing dynamic standardization and spatiotemporal correlation analysis on the feature vectors, generate aggregate parameters based on extreme value sampling and mean sampling, and generate weighted feature data through weight allocation. The data augmentation module is configured to perform feature dimensionality reduction, multidimensional filtering, nonlinear activation, noise suppression, and feature dimensionality increase operations sequentially on the input data obtained after adding the weighted feature data and the input data of the multidimensional parameter analysis unit and standardizing the input data to generate augmented feature data; the output of the multidimensional parameter analysis unit is jointly determined by the weighted feature data and the augmented feature data, and serves as the input of the next processing unit; The status analysis module is connected to the end-of-line data processing module and is configured to perform feature integration and decision mapping on the output data of the tobacco transportation status analysis model to generate transportation status assessment results.

2. The real-time monitoring and management system for tobacco leaf transportation according to claim 1, characterized in that, The data fusion unit in the first data processing module is the basic data fusion unit, while the data fusion units in other data processing modules are multi-dimensional data fusion units. The basic data fusion unit includes processing standardized environmental parameters using a first filtering network to obtain primary features, and performing data standardization processing on the primary features to obtain the output data of the basic data fusion unit. The multidimensional data fusion unit includes performing a second filtering network and feature dimensionality reduction processing on the output data of the previous data processing module to obtain a first processing result and a second processing result; performing channel calibration processing on the first processing result to obtain a third processing result; and inputting the third processing result and the second processing result into the feature splicing module to obtain the output data of the multidimensional data fusion unit.

3. The real-time monitoring and management system for tobacco leaf transportation according to claim 1, characterized in that, The multidimensional parameter analysis unit further includes a first data standardization layer, a first feedback network, a second data standardization layer, a second feedback network, and a feature normalization layer; The first input data is processed by the first data standardization layer and then processed to obtain the second input data from the multidimensional parameter analysis unit. The environmental parameter attention module processes the second input data to obtain weighted feature data; The weighted feature data is added to the first input data through the first feedback network, and the result is added to the second data standardization layer. The input data of this layer is processed using the second data normalization layer to obtain the third input data; The data augmentation module processes the third input data to obtain augmented feature data; The enhanced feature data is added to the summation result through the second feedback network and then added to the feature normalization layer to obtain the output data of the multidimensional parameter analysis unit.

4. The real-time monitoring and management system for tobacco leaf transportation according to claim 1, characterized in that, The tobacco leaf transportation status analysis model includes three data processing modules connected in sequence. The second data processing module has three multi-dimensional parameter analysis units connected in sequence, and the remaining data processing modules have two multi-dimensional parameter analysis units connected in sequence.

5. The real-time monitoring and management system for tobacco leaf transportation according to claim 1, characterized in that, The state analysis module includes a parameter integration layer and a decision mapping layer. The parameter integration layer performs 1×1 parameter aggregation on the input data of the state analysis module in terms of feature dimensions. The transportation status assessment result is obtained by nonlinearly mapping the output of the parameter integration layer using the decision mapping layer.

6. The real-time monitoring and management system for tobacco leaf transportation according to claim 1, characterized in that, The constructed environmental anomaly discrimination loss function is used to optimize the tobacco leaf transportation status analysis model. The expression of the environmental anomaly discrimination loss function is as follows: in, Indicates the first Environmental anomalies in each sample The model represents the samples Correctly identified as the actual state confidence level The model represents the samples Predicted as The characteristic score of the state Indicates the number of categories of transportation environment status. Indicates the category to which sample i belongs. frequency factor, , Distributed parameters representing environmental parameters This indicates that the model predicts the value of sample i as follows: The feature score, where e is the natural constant.

7. The real-time monitoring and management system for tobacco leaf transportation according to claim 1, characterized in that, The multi-dimensional filtering window size of the environmental parameter attention module is 5×5×3. In the spatiotemporal correlation analysis, a 7×7×5 three-dimensional convolution kernel is used with a dilation interval of 2. A fully connected layer is used for the linear transformation of the aggregation parameters. The multi-dimensional filtering window size of the data augmentation module is 3×3×2, and the ReLU function is applied to the nonlinear activation layer.

8. The real-time monitoring and management system for tobacco leaf transportation according to claim 7, characterized in that, The feature vectors are subjected to dynamic standardization and spatiotemporal correlation analysis, respectively, and the results of the two processes are superimposed to obtain the superimposed feature: in, Indicates superposition features, This represents the multi-dimensional filtering results of the environmental parameter attention module. This indicates dynamic standardization processing. This indicates dilated convolution processing. This indicates an overlay operation.

9. The real-time monitoring and management system for tobacco leaf transportation according to claim 2, characterized in that, The kernel size of the first filter network is 5×5×2, the kernel size of the second filter network is 5×5×1, and the sliding step size is 1. The feature dimensionality reduction process uses a 2×2×1 convolution kernel with a stride of 1.

10. A method for real-time monitoring of tobacco leaf transportation using the real-time monitoring and management system for tobacco leaf transportation according to any one of claims 1-9, characterized in that, include: Step 1: Collect transportation environment data and preprocess it to obtain standardized environmental parameters; Step 2: Use the basic data fusion unit to perform feature fusion on the standardized environmental parameters to obtain the first output parameter; Step 3: Input the first output parameter into multiple sequentially connected multidimensional parameter analysis units to obtain the second output parameter, which serves as the input data for the multidimensional data fusion unit; Capture the spatiotemporal correlation characteristics of the transportation environment through the multidimensional perception capabilities of the multidimensional parameter analysis unit; Step 4: Use the multi-dimensional data fusion unit to perform multi-scale feature fusion processing on the input data of this unit to obtain the third output parameter; Step 5: Input the third output parameter into multiple sequentially connected multidimensional parameter analysis units to obtain the fourth output parameter, and input it into the multidimensional data fusion unit; Step Six: Repeat Steps Four and Five until the preset number of iterations is reached, then output the fifth output parameter; Step 7: Input the fifth output parameter into the status analysis module to obtain the transportation environment status assessment result.

Images