Hazardous waste logistics information management system and method

Abstract

The invention relates to the field of logistics information management, and particularly discloses a dangerous waste logistics information management system and method.The method comprises the steps that transfer joint order image information, dangerous waste characteristic basic information and dangerous waste transportation information collected by a camera are obtained firstly, and then a deep learning technology is utilized to obtain dangerous waste logistics information; and feature extraction and correlation analysis are carried out on the three parts, and finally a hazardous waste logistics information report is generated through a generator, so that the overall process of hazardous waste management is optimized, the processing efficiency and accuracy are improved, and safer and sustainable waste management is realized.

Description

Technical Field

[0001] The present application relates to the field of logistics information management, and more specifically, to a hazardous waste logistics information management system and method. Background Art

[0002] Hazardous waste refers to waste that is potentially harmful due to its chemical, physical or biological properties, such as toxic substances, carcinogens, corrosive substances or infectious biological products. This type of waste requires special management and control to comply with strict laws, regulations and environmental standards.

[0003] Traditional processing methods usually involve collecting and filling out transfer forms, which require repeated filling out of the transfer form content during each transfer process. The operation is inconvenient and inefficient, with a high probability of human error, and it is difficult to ensure the transparency and traceability of information.

[0004] Therefore, a hazardous waste logistics information management system and method are desired. Summary of the invention

[0005] In order to solve the above technical problems, this application is proposed. The embodiment of this application provides a hazardous waste logistics information management system and method, which first obtains the transfer form image information, basic information of hazardous waste characteristics and hazardous waste transportation information collected by the camera, and then uses deep learning technology to perform feature extraction and correlation analysis on the three, and finally generates a hazardous waste logistics information report through a generator to optimize the overall process of hazardous waste management, improve processing efficiency and accuracy, and thus achieve safer and more sustainable waste management.

[0006] According to one aspect of the present application, a hazardous waste logistics information management system is provided, comprising:

[0007] The hazardous waste logistics information acquisition module is used to obtain the transfer form image information collected by the camera, basic information on hazardous waste characteristics and hazardous waste transportation information;

[0008] A hazardous waste logistics information extraction module, used to extract a transfer form-related feature vector and a hazardous waste information-related feature vector from the transfer form image information collected by the camera, the basic information of the hazardous waste characteristics and the hazardous waste transportation information;

[0009] The hazardous waste logistics information report generation module is used to generate a hazardous waste logistics information report based on the transfer form association feature vector and the hazardous waste information association feature vector.

[0010] According to another aspect of the present application, a method for managing hazardous waste logistics information is provided, comprising:

[0011] Obtain transfer form image information, basic information on hazardous waste characteristics, and hazardous waste transportation information collected by the camera;

[0012] Extracting a transfer form-related feature vector and a hazardous waste information-related feature vector from the transfer form image information collected by the camera, the basic information of hazardous waste characteristics, and the hazardous waste transportation information;

[0013] Based on the transfer form association feature vector and the hazardous waste information association feature vector, a hazardous waste logistics information report is generated.

[0014] Compared with the prior art, the present application provides a hazardous waste logistics information management system and method, which first obtains the transfer form image information, basic information on hazardous waste characteristics and hazardous waste transportation information collected by a camera, and then uses deep learning technology to perform feature extraction and correlation analysis on the three. Finally, a generator is used to generate a hazardous waste logistics information report to optimize the overall process of hazardous waste management, improve processing efficiency and accuracy, and thus achieve safer and more sustainable waste management. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0016] Figure 1 It is a block diagram of a hazardous waste logistics information management system according to an embodiment of the present application.

[0017] Figure 2 It is a block diagram of a hazardous waste logistics information extraction module in a hazardous waste logistics information management system according to an embodiment of the present application.

[0018] Figure 3 It is a block diagram of the transfer form image information feature extraction unit in the hazardous waste logistics information management system according to an embodiment of the present application.

[0019] Figure 4 It is a block diagram of a hazardous waste logistics information report generation module in a hazardous waste logistics information management system according to an embodiment of the present application.

[0020] Figure 5 Flow chart of a hazardous waste logistics information management method according to an embodiment of the present application.

[0021] Figure 6 is a block diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0022] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0023] Figure 1 FIG. 1 is a block diagram of a hazardous waste logistics information management system according to an embodiment of the present application. Figure 1 As shown, according to the hazardous waste logistics information management system 100 of the embodiment of the present application, it includes: a hazardous waste logistics information acquisition module 110, which is used to acquire the transfer form image information, basic information of hazardous waste characteristics and hazardous waste transportation information collected by the camera; a hazardous waste logistics information extraction module 120, which is used to extract the transfer form association feature vector and the hazardous waste information association feature vector from the transfer form image information, the basic information of hazardous waste characteristics and the hazardous waste transportation information collected by the camera; a hazardous waste logistics information report generation module 130, which is used to generate a hazardous waste logistics information report based on the transfer form association feature vector and the hazardous waste information association feature vector.

[0024] In the above-mentioned hazardous waste logistics information management system 100, the hazardous waste logistics information acquisition module 110 is used to obtain the transfer form image information, basic information on hazardous waste characteristics and hazardous waste transportation information collected by the camera. It should be understood that hazardous waste is waste that may cause potential harm to the environment or health due to its chemical, physical or biological properties, such as toxic substances, carcinogens, corrosive substances or infectious biological products. In order to meet strict regulations and environmental protection standards, these wastes need to be managed and handled specially. Traditional waste treatment methods usually need to be managed by filling out transfer forms. Each waste transfer requires manual filling of these transfer forms, which is not only cumbersome and inefficient, but also prone to human errors. Therefore, it becomes very difficult to ensure the transparency and traceability of information. Therefore, in the technical solution of the present application, by obtaining the transfer form image information, basic information on hazardous waste characteristics and hazardous waste transportation information collected by the camera, and combining deep learning technology to generate a hazardous waste logistics information report, the overall hazardous waste management process can be optimized, the processing efficiency and accuracy can be improved, so as to achieve a safer and more sustainable waste management goal.

[0025] Specifically, the transfer form image information collected by the camera provides real-time visual data, including the packaging and labeling of hazardous waste, which is crucial to ensure the safe transportation of hazardous waste. The basic information on the characteristics of hazardous waste covers key data such as the type, quantity, and chemical properties of the waste, which helps to assess the hazardousness and treatment requirements of the waste. At the same time, the hazardous waste transportation information records the transportation route, transportation method and related safety measures of the waste, and provides comprehensive tracking and management of the waste transportation process. The acquisition of this information can not only effectively manage the entire life cycle of hazardous waste and ensure that it meets environmental and safety standards, but also provide the necessary basis and data support for regulatory authorities, as well as provide key references when responding to emergencies and planning waste treatment strategies. By collecting and integrating this information through cameras, accurate management and monitoring of the transportation and treatment process of hazardous waste can be achieved, thereby improving overall environmental safety and management efficiency.

[0026] In the above-mentioned hazardous waste logistics information management system 100, the hazardous waste logistics information extraction module 120 is used to extract the transfer form association feature vector and the hazardous waste information association feature vector from the transfer form image information, the basic information of hazardous waste characteristics and the hazardous waste transportation information collected by the camera. It should be understood that the purpose of extracting the transfer form association feature vector and the hazardous waste information association feature vector from the transfer form image information, the basic information of hazardous waste characteristics and the hazardous waste transportation information collected by the camera is to achieve in-depth analysis and comprehensive management of hazardous waste logistics data. Extracting the transfer form association feature vector and the hazardous waste information association feature vector from the transfer form image information, the basic information of hazardous waste characteristics and the hazardous waste transportation information collected by the camera not only helps to achieve effective management and monitoring of hazardous waste logistics data, but also provides important data support and decision-making basis for environmental protection and safety management.

[0027] Figure 2 FIG. 1 is a block diagram of a hazardous waste logistics information extraction module in a hazardous waste logistics information management system according to an embodiment of the present application. Figure 2As shown, in a specific embodiment of the present application, the hazardous waste logistics information extraction module 120 includes: a transfer form image information feature extraction unit 121, which is used to extract features from the transfer form image information collected by the camera to obtain the transfer form association feature vector; a hazardous waste characteristic feature extraction unit 122, which is used to extract features from the basic information of hazardous waste characteristics to obtain a global hazardous waste characteristic basic information feature vector; a hazardous waste transportation information feature extraction unit 123, which is used to extract features from the hazardous waste transportation information to obtain a hazardous waste transportation information semantic association feature vector; a hazardous waste information feature association unit 124, which is used to associate the global hazardous waste characteristic basic information feature vector with the hazardous waste transportation information semantic association feature vector to obtain the hazardous waste information association feature vector.

[0028] It should be understood that the transfer form associated feature vector contains important features extracted from the image, such as the packaging status, marking status, and transport container characteristics of hazardous waste. This information is converted into a numerical data representation through image processing algorithms and feature extraction methods. The feature extraction process first optimizes the transfer form image data into a form suitable for further analysis through image preprocessing techniques such as image enhancement and denoising. Then, algorithms such as target detection and image segmentation are used to extract hazardous waste and related information from the image to form image regions of interest. Next, representative features such as shape, color, texture and other visual features are extracted from these regions through deep learning models or traditional feature extraction methods.

[0029] Furthermore, the purpose of extracting the basic information of hazardous waste characteristics to obtain the feature vector of global basic information of hazardous waste characteristics is to extract key features from huge data so as to more effectively manage and analyze the chemical characteristics and treatment requirements of hazardous waste. The basic information of hazardous waste characteristics covers key data such as chemical composition, toxicity level, physical state, treatment method, etc., which are of great significance for assessing the hazard level of waste, formulating safe treatment measures, and complying with environmental regulations. The feature extraction process includes converting the basic information of hazardous waste characteristics into numerical or vectorized form for further processing and analysis by computer systems. First, the original text description or attribute data is split and standardized into multiple data items or feature labels through word segmentation and semantic understanding technology. Then, these data items are integrated and arranged into a global basic information feature matrix using data mining and statistical analysis methods. In the process of feature extraction, deep learning models such as convolutional neural networks may be used to learn and abstract the basic information of characteristics and extract representative and discriminative feature vectors. These feature vectors can reflect the key characteristics of hazardous waste characteristics, such as the main components of its chemical composition, the correlation between different characteristics, etc., and provide a basis for subsequent data analysis and model building. By extracting the feature vectors of basic information of global hazardous waste characteristics, it is possible to quickly analyze and compare a large amount of hazardous waste data, and support the optimization design of environmental monitoring, waste classification, and treatment processes. These feature vectors not only help to accurately assess the degree of hazard and difficulty of waste treatment, but also provide a scientific basis for enterprises and regulatory agencies to ensure that waste management meets environmental protection and safety standards. Therefore, the application of feature extraction technology in hazardous waste management is of great significance for improving management efficiency and reducing environmental risks.

[0030] Furthermore, the purpose of extracting features from hazardous waste transportation information to obtain semantically associated feature vectors of hazardous waste transportation information is to extract key information from complex transportation data to support intelligent management and comprehensive analysis of the waste transportation process. Hazardous waste transportation information involves key data such as the starting point, end point, transportation route, transportation method, and safety measures of the waste, which are essential to ensure the safe transportation of the waste and compliance with regulatory requirements. The feature extraction process includes converting hazardous waste transportation information into structured or semantic feature vectors for further processing and understanding by the system. First, the transportation information is normalized and standardized through data cleaning and preprocessing techniques to ensure data consistency and accuracy. Then, the key words, entities, and relationships in the information are extracted using natural language processing technology or knowledge graph construction methods to form semantically associated feature vectors. In the process of feature extraction, machine learning models or rule-based methods may be used to analyze and mine the semantic associations in the transportation information. For example, by identifying key words such as waste type, transportation method, safety measures, etc., logical connections and associated features between these key information are established.

[0031] In particular, the purpose of associating the global basic information feature vector of hazardous waste characteristics and the semantic association feature vector of hazardous waste transportation information to obtain the hazardous waste information association feature vector is to integrate the chemical characteristics and transportation process data of the waste, so as to achieve comprehensive analysis and management of the entire life cycle of the waste. The global basic information feature vector of hazardous waste characteristics reflects important characteristics of the waste, such as chemical composition, toxicity level, and physical state, while the semantic association feature vector of hazardous waste transportation information covers key information such as the transportation route, safety measures, and transportation method of the waste. Comprehensive analysis of the association between waste characteristics and transportation information helps to assess the risks and challenges that waste may face during transportation. For example, combining the chemical composition characteristics of the waste and the transportation route information, it is possible to quickly identify the types of waste and transportation links that may cause safety problems, and take targeted safety measures and management strategies.

[0032] Figure 3 FIG. 1 is a block diagram of a transfer form image information feature extraction unit in a hazardous waste logistics information management system according to an embodiment of the present application. Figure 3 As shown, in a specific embodiment of the present application, the transfer link image information feature extraction unit 121 includes: a transfer link image data preprocessing subunit 1211, used to perform data preprocessing on the transfer link image information collected by the camera to obtain multiple transfer link image regions of interest; a transfer link associated non-local convolution subunit 1212, used to pass the multiple transfer link image regions of interest through a transfer link non-local neural network as a feature encoder to obtain the transfer link associated feature vector.

[0033] It should be understood that the transfer form image may contain a lot of background information and insignificant parts, while the region of interest refers to the important area in the image related to a specific object or event, such as the mark of hazardous waste, the packaging status, or the characteristics of the transport container. The data preprocessing stage includes techniques such as image enhancement, denoising, and size normalization to ensure the quality and consistency of image data. These steps help to eliminate noise in the image, enhance the contrast and clarity of the image, and thus improve the accuracy of subsequent detection and extraction of the region of interest. The preprocessed images are analyzed using computer vision techniques such as target detection, image segmentation, and feature extraction. Target detection technology can identify hazardous waste marks or specific areas in the image, and image segmentation can divide the image into several sub-regions to extract the target region of interest. Feature extraction methods are used to extract meaningful features such as shape, color, texture, etc. from these regions for subsequent analysis and processing.

[0034] Furthermore, the purpose of using the transfer list non-local neural network as a feature encoder to obtain the transfer list associated feature vector is to improve the data analysis and decision support capabilities of the waste transportation process. The region of interest of the transfer list image may contain rich information, such as waste marking, packaging status, and characteristics of the transport container, which are crucial for safety supervision and transportation management. The transfer list non-local neural network as a feature encoder has the ability to effectively extract complex relationships and patterns in the image. The non-local neural network combines the cutting-edge technology of deep learning and can automatically discover and extract important features in the region of interest, such as shape, texture, color, etc., by learning the correlation between each pixel in the image. By taking the region of interest as input, the transfer list non-local neural network can convert each region of interest into a feature vector in a high-dimensional feature space. These feature vectors not only reflect the unique attributes of each region, but also contain the semantic association and spatial location information between regions, thereby forming the associated feature vector of the transfer list. Specifically, the first point convolution processing, the second point convolution processing and the third point convolution processing are respectively performed on the regions of interest of the multiple transferred linked images to obtain a first feature map, a second feature map and a third feature map; the position-weighted sum between the first feature map and the second feature map is calculated to obtain a fused feature map; the fused feature map is input into a Softmax function to map the feature values ​​of each position in the fused feature map to a probability space to obtain a normalized fused feature map; the position-point multiplication between the normalized fused feature map and the third feature map is calculated to obtain a re-fused feature map; the re-fused feature map is embedded in a Gaussian similarity function to obtain a global similarity feature map; the global similarity feature map is subjected to a fourth point convolution processing to adjust the number of channels of the global similarity feature map to obtain a channel-adjusted global similarity feature map; and the position-weighted sum of the channel-adjusted global similarity feature map and the regions of interest of the multiple transferred linked images is calculated to obtain the transferred linked association feature vector.

[0035] In a specific embodiment of the present application, the transfer link image data preprocessing subunit 1211 includes: performing image division on the transfer link image information collected by the camera to obtain multiple transfer link image areas; passing the multiple transfer link image areas through a transfer link target detection network to obtain the multiple transfer link image regions of interest.

[0036] It should be understood that the transfer slip image may contain complex scenes and multiple regions, such as different types of waste containers, labels and transport vehicles, etc. These regions can be segmented through image segmentation for independent analysis and processing. Image segmentation technology can perform regional segmentation based on specific features or predefined boundaries in the image. These features may be color, texture, shape, etc., or the segmentation process can be automated based on pixel intensity or contextual information. Through effective image segmentation algorithms, complex transfer slip images can be decomposed into multiple independent regions, each of which contains specific waste information or transportation details. The acquisition of multiple transfer slip image regions helps to optimize subsequent data analysis and decision support. Each image region can be extracted and analyzed separately to identify important information such as waste type, transportation status, safety signs, etc., so as to more accurately evaluate the compliance and safety of the waste transportation process.

[0037] Furthermore, the purpose of obtaining multiple transfer list image regions through the transfer list object detection network is to realize the automatic recognition and extraction of key information in the waste transportation process. The transfer list image region may contain various waste containers, labels and transportation tools, and the object detection network can effectively locate and identify the target objects or key features in these regions, such as waste markings, packaging status or specific features of transportation containers. The transfer list object detection network is based on deep learning technology and learns and recognizes different targets in waste transportation images by training a large number of labeled data sets. This network can automatically detect the target objects that may exist in each transfer list image region, such as different types of waste containers or transportation tools, based on the visual features and semantic information of the image. The transfer list image regions of interest extracted by the object detection network have high precision and accuracy, which can effectively reduce manual intervention and error rate. These regions of interest are regarded as a collection of key information and can provide detailed data on various aspects of the waste transportation process, such as important information such as waste type, quantity, packaging status, and safety identification. The transfer list object detection network supports simultaneous detection and recognition of multiple categories of targets and is suitable for complex scenes and diverse waste transportation images. It can effectively cope with different lighting conditions, angle changes and background interference, and improve the comprehensiveness and accuracy of data in the waste management process. Specifically, the transfer link object detection network is used to process the multiple transfer link image regions using the following object detection formula to obtain the multiple transfer link image interest regions;

[0038] Among them, the target detection formula is:

[0039] ROI=(cls(ψ det ,B),Regr(ψ det ,B))

[0040] Among them, ψ det are multiple transfer image regions, B is the anchor box, ROI is the region of interest of multiple transfer image regions, cls(ψ det ,B) indicates classification, Regr(ψ det ,B) indicates regression.

[0041] In a specific embodiment of the present application, the hazardous waste characteristic feature extraction unit 122 includes: segmenting the basic information of hazardous waste characteristics to obtain multiple basic information data items of hazardous waste characteristics; arranging the multiple basic information data items of hazardous waste characteristics in a two-dimensional matrix to obtain a global basic information feature matrix of hazardous waste characteristics; and passing the global basic information feature matrix of hazardous waste characteristics through a global basic information convolutional neural network of hazardous waste characteristics to obtain the global basic information feature vector of hazardous waste characteristics.

[0042] It should be understood that the characteristic information of hazardous waste usually includes multiple aspects such as the physical properties, chemical composition, and hazardous characteristics of the waste. By segmenting this information, complex characteristic descriptions can be broken down into more specific and accurate data items. For example, physical properties can be divided into particle size, density, shape, etc.; chemical composition can be divided into main chemical substances, content ratio, etc.; hazardous characteristics can be divided into toxicity, corrosiveness, flammability, etc. Word segmentation enables each data item to independently represent a key feature, which is crucial for waste classification, management, and the formulation of safety measures. For example, by analyzing the chemical composition data items of waste, its potential impact on the environment and human health can be evaluated, and the correct treatment and transportation plans can be formulated in a targeted manner.

[0043] Furthermore, the purpose of arranging multiple basic information data items of hazardous waste characteristics in a two-dimensional matrix to obtain a global basic information feature matrix of hazardous waste characteristics is to comprehensively and systematically display the multi-dimensional information of waste characteristics, so as to facilitate a comprehensive understanding and analysis of key aspects such as the degree of hazard, treatment requirements and environmental impact of waste. Arranging multiple data items in a two-dimensional matrix can organize and structure the waste characteristic information in rows and columns. For example, rows can represent different types or batches of hazardous wastes, and columns can represent specific characteristics of wastes, such as chemical composition, toxicity level, corrosiveness, etc. This arrangement allows each data item to have its own independent position in the matrix, which is easy to find and compare. The global basic information feature matrix of hazardous waste characteristics can provide a comprehensive data perspective to help managers and decision makers comprehensively assess the risks and management needs of wastes. Through the matrix, the differences and similarities of different wastes in various characteristics can be seen at a glance, providing a scientific basis for formulating effective waste management strategies. The establishment of a feature matrix is ​​conducive to data standardization and comparative analysis. Wastes from different sources and types may have different characterization methods. By arranging them in a unified matrix, the format and expression of waste characteristic data can be standardized, promoting data exchange and sharing, thereby strengthening global and international cooperation and communication.

[0044] Furthermore, the purpose of obtaining the feature vector of the basic information of global hazardous waste characteristics by passing the feature matrix of the basic information of global hazardous waste characteristics through the convolutional neural network of the basic information of global hazardous waste characteristics is to convert the complex multidimensional data structure into a compact and representative vector form, so as to facilitate more efficient data analysis, pattern recognition and predictive modeling. The feature matrix of the basic information of global hazardous waste characteristics usually contains a large number of data items and dimensions, each of which may represent the characteristic description of different wastes. Through the convolutional neural network (CNN), key features can be effectively extracted from these data and converted into feature vectors with a fixed length. The convolutional neural network can gradually extract and refine the spatial and temporal correlation features in the waste characteristic matrix through multi-layer convolution and pooling operations. The representation of these features in the feature vector can better capture the potential connections and patterns between waste characteristics, which helps to improve the expressiveness of data and the utilization of information. The feature vector obtained by the convolutional neural network of the basic information of global hazardous waste characteristics has high abstraction and generalization ability. This enables the feature vector to not only simplify the data structure, but also more effectively cope with the diversity and complexity of waste characteristic descriptions, providing a more stable and reliable data foundation for subsequent data analysis and application. Specifically, each layer of the global hazardous waste characteristics basic information convolutional neural network is used to perform convolution processing, mean pooling processing based on the local feature matrix and nonlinear activation processing on the input data in the forward pass of the layer to output the global hazardous waste characteristics basic information feature vector from the last layer of the global hazardous waste characteristics basic information convolutional neural network, wherein the input of the global hazardous waste characteristics basic information convolutional neural network is the global hazardous waste characteristics basic information feature matrix.

[0045] In a specific embodiment of the present application, the hazardous waste transportation information feature extraction unit 123 includes: passing the hazardous waste transportation information through a converter-based hazardous waste transportation information semantic encoder to obtain multiple hazardous waste transportation information semantic feature vectors; and cascading the multiple hazardous waste transportation information semantic feature vectors to obtain the hazardous waste transportation information semantic association feature vector.

[0046] It should be understood that hazardous waste transportation involves a variety of information, such as transportation routes, transportation methods, transportation conditions, waste types and quantities, etc. Through the transformer-based semantic encoder, this information can be encoded and understood in a natural language processing manner, so that each transportation information can be mapped to a high-dimensional semantic space. Semantic encoders based on transformer architecture, such as BERT or GPT, can use pre-trained language models and self-attention mechanisms to perform deep learning representations of hazardous waste transportation information. These models can capture the complex dependencies and semantic associations between words and sentences in the text, and generate feature vectors with rich semantic information. Through the generation of semantic feature vectors, semantic understanding and similarity comparison of waste transportation information can be achieved. These vectors can not only reveal the relationship between different transportation information, but also be used to identify abnormal or risk events, and improve the safety and efficiency of the transportation process. Specifically, the hazardous waste transportation information is segmented to obtain a word sequence; the embedding layer of the converter-based hazardous waste transportation information semantic encoder is used to map each word in the word sequence into a word embedding vector to obtain a sequence of word embedding vectors; the converter-based Bert model of the converter-based hazardous waste transportation information semantic encoder is used to perform global contextual semantic encoding on the sequence of word embedding vectors to obtain multiple hazardous waste transportation information semantic feature vectors.

[0047] Furthermore, the purpose of cascading multiple semantic feature vectors of hazardous waste transportation information to obtain semantic association feature vectors of hazardous waste transportation information is to expand the semantic representation of a single transportation information into an overall representation containing a wider range of information, thereby more comprehensively capturing the association and comprehensive characteristics of the waste transportation process. A single semantic feature vector of hazardous waste transportation information usually contains the semantically important features of the information, such as transportation mode, starting point, end point, transportation conditions, etc. By cascading multiple such vectors, the associations between different transportation information can be integrated and displayed. For example, multiple transportation information with similar starting points and end points can be cascaded to form a larger semantic space. The cascading operation enables the semantic association feature vectors of hazardous waste transportation information to better reflect the overall characteristics and trends of the waste transportation process. These feature vectors not only contain the details of a single transportation information, but also take into account the interactions and influences between multiple information, thereby providing a more comprehensive and integrated information basis to support higher-level data analysis and decision-making. Cascading multiple semantic feature vectors also helps to identify and analyze patterns and anomalies in the waste transportation process. By comparing and analyzing the semantic association feature vectors of hazardous waste transportation information, we can discover common transportation routes, frequently occurring changes in transportation conditions, and potential transportation risk factors, providing more accurate and effective support for preventing and handling accidents.

[0048] In the above-mentioned hazardous waste logistics information management system 100, the hazardous waste logistics information report generation module 130 is used to generate a hazardous waste logistics information report based on the transfer form association feature vector and the hazardous waste information association feature vector. It should be understood that the transfer form association feature vector reflects the key information in the waste transportation process, such as the transportation starting point, end point, route, transportation method and conditions. This information is structured and encoded in the form of semantic feature vectors, so that each transportation data can be effectively understood and analyzed. The hazardous waste information association feature vector covers the characteristic description of the waste, such as chemical composition, toxicity level, treatment requirements, etc. These feature vectors provide important information from the perspective of waste, which helps to assess the degree of danger and environmental impact of waste, and provide a scientific basis for the formulation of waste management and treatment strategies. By comprehensively analyzing the association feature vectors of the transfer form and waste information, it is possible to fully understand and evaluate the key information and characteristics in the waste logistics process, and provide substantial support and guidance for environmental protection and waste management decisions.

[0049] Figure 4 FIG. 1 is a block diagram of a hazardous waste logistics information report generation module in a hazardous waste logistics information management system according to an embodiment of the present application. Figure 4 As shown, in a specific embodiment of the present application, the hazardous waste logistics information report generation module 130 includes: a hazardous waste logistics information feature fusion unit 131, which is used to fuse the transfer form association feature vector and the hazardous waste information association feature vector to obtain a hazardous waste logistics information generation feature vector; a hazardous waste logistics information feature optimization unit 132, which is used to perform group distribution constraints on the hazardous waste logistics information generation feature vector based on inherent modal information to obtain an optimized hazardous waste logistics information generation feature vector; a logistics information report generation unit 133, which is used to pass the optimized hazardous waste logistics information generation feature vector through a generator to generate a hazardous waste logistics information report.

[0050] It should be understood that the purpose of fusing the transfer form association feature vector and the hazardous waste information association feature vector to obtain the hazardous waste logistics information generation feature vector is to integrate the waste transportation process and characteristic information into a unified and comprehensive representation to support intelligent analysis, prediction and decision-making of waste logistics management. The feature vector comprehensively considers the dynamic information of the spatial and temporal dimensions in the waste transportation process, as well as the multidimensional characteristics of the waste itself, to form a comprehensive and integrated description.

[0051] In particular, in the technical solution of the present application, considering that hazardous waste logistics information involves multiple aspects such as transfer form image information, basic information on waste characteristics and transportation details, this information needs to be converted into feature vectors that can be effectively described and utilized through feature mining, so as to carry out further data analysis and decision support. Here, the characteristic manifold of the feature vector generated by hazardous waste logistics information is adjusted by using the prior information of the neural network model, which can optimize the representation ability of the feature vector and the structured processing of the data. Among them, the neural network can effectively extract high-level features and complex structures in the data by learning patterns and associations in a large number of data samples, so that the feature vector generated by hazardous waste logistics information is more representative and discriminative. Specifically, in practical applications, some special or edge cases may be encountered, such as atypical waste characteristics or abnormal transportation details, which may have an important impact on the generation and reporting of logistics information. By introducing special instance perturbations in the training stage, the model can better adapt to and handle these special cases, improving the robustness and coping ability of the system.

[0052] Among them, the hazardous waste logistics information generation feature vector is subjected to a group distribution constraint based on the inherent modal information to obtain an optimized hazardous waste logistics information generation feature vector, including: calculating the mean and variance of the hazardous waste logistics information generation feature vector; calculating the difference between the eigenvalue of a predetermined position of the hazardous waste logistics information generation feature vector and the mean of the hazardous waste logistics information generation feature vector to obtain a first intermediate calculation value; multiplying the network parameter matrix by the hazardous waste logistics information generation feature vector and then subtracting the network parameter matrix from the hazardous waste logistics information generation feature vector by position to obtain a first intermediate vector, and calculating the Frobenius norm of the first intermediate vector to obtain a first norm value; adding the first norm value to the hazardous waste logistics information generation feature vector; The square root of the variance of the hazardous waste logistics information generating characteristic vector is taken to obtain a second intermediate calculation value; the first intermediate calculation value is divided by the second intermediate calculation value to obtain a third intermediate calculation value; the squares of each position in the hazardous waste logistics information generating characteristic vector are calculated and then accumulated to obtain a first total value; the square root of the first total value is taken and then divided by the scale of the network parameter matrix, and then the natural exponential function value is calculated to obtain a fourth intermediate calculation value, wherein the scale of the network parameter matrix is ​​equal to the product of the width value and the height value of the network parameter matrix; the fourth intermediate calculation value is multiplied by the weighted hyperparameter and then added to the third intermediate calculation value to obtain the eigenvalue of the predetermined position of the optimized hazardous waste logistics information generating characteristic vector.

[0053] Among them, the hazardous waste logistics information generation feature vector is subjected to group distribution constraints based on inherent modal information to obtain an optimized hazardous waste logistics information generation feature vector, including:

[0054]

[0055] Where V is the feature vector generated by the hazardous waste logistics information, M is the network parameter matrix, α represents the weighted hyperparameter, and v i is the eigenvalue of the i-th eigenvector generated for the hazardous waste logistics information, N is the scale of the network parameter matrix, the scale of the network parameter matrix is ​​equal to the product of the width value of the network parameter matrix and its height value, L is the length of the eigenvector generated for the hazardous waste logistics information, ||·|| F The Frobenius norm of the feature vector generated by the hazardous waste logistics information, represents matrix multiplication, μ represents the mean of the feature set composed of all eigenvalues ​​of the feature vector generated by the hazardous waste logistics information, σ 2 represents the variance of the feature set composed of all eigenvalues ​​of the feature vector generated by the hazardous waste logistics information, v' i The i-th eigenvalue in the eigenvector is generated for the optimized hazardous waste logistics information.

[0056] That is, in the technical solution of the present application, considering that when the neural network model performs feature mining on the hazardous waste logistics information generation feature vector based on the network parameter matrix, if the inherent implicit association between the network output feature and the network parameter matrix can be further considered, the prior information of the neural network model can be further used to modulate the feature manifold of the hazardous waste logistics information generation feature vector to reduce the possibility of the occurrence of special instance disturbances of its feature values. Based on this, in a preferred embodiment, the hazardous waste logistics information generation feature vector is subjected to group distribution constraints based on inherent modal information, which first maps the hazardous waste logistics information generation feature vector to the high-dimensional feature space where the network parameter matrix is ​​located, and uses the compatibility perception factor between the mapping result and the original input as the inherent implicit association descriptor between the network output feature and the network parameter matrix. Then, the statistical characteristics of the feature group of the hazardous waste logistics information generation feature vector are used as the inherent modal information of the hazardous waste logistics information generation feature vector, and the inherent implicit association descriptor is used as the instance disturbance differential adjustment variable, and the attribute values ​​of each position of the hazardous waste logistics information generation feature vector are dynamically standardized. At the same time, the class-standardized modulus data of the hazardous waste logistics information generating feature vector is used as a bias item to constrain the acceptable range of the population distribution of the hazardous waste logistics information generating feature vector. In this way, the attribute influence of the interference distribution of abnormal samples can be suppressed as much as possible, thereby improving the expression accuracy of the hazardous waste logistics information generating feature vector.

[0057] Furthermore, the purpose of generating hazardous waste logistics information reports by optimizing the feature vectors of hazardous waste logistics information is to use a data-driven approach to transform complex waste logistics information into detailed reports that are easy to understand and use, so as to support environmental protection, safety management and decision-making. By processing the optimized feature vectors of hazardous waste logistics information through the generator, generating hazardous waste logistics information reports can not only transform complex data into practical report forms, but also provide scientific data support and technical guarantees for waste management and environmental protection, and promote the intelligent and efficient development of waste management.

[0058] In summary, the embodiment of the present application first obtains the transfer form image information, basic information on hazardous waste characteristics and hazardous waste transportation information collected by the camera, and then uses deep learning technology to perform feature extraction and correlation analysis on the three. Finally, a generator is used to generate a hazardous waste logistics information report to optimize the overall process of hazardous waste management, improve processing efficiency and accuracy, and thus achieve safer and more sustainable waste management.

[0059] As described above, the hazardous waste logistics information management system 100 according to the embodiment of the present application can be implemented in various terminal devices. In one example, the hazardous waste logistics information management system 100 can be integrated into the terminal device as a software module and / or a hardware module. For example, the hazardous waste logistics information management system 100 can be a software module in the operating system of the terminal device, or can be an application developed for the terminal device; of course, the hazardous waste logistics information management system 100 can also be one of the many hardware modules of the terminal device.

[0060] Alternatively, in another example, the hazardous waste logistics information management system 100 and the terminal device may also be separate devices, and the hazardous waste logistics information management system 100 may be connected to the terminal device via a wired and / or wireless network and transmit interactive information in accordance with an agreed data format.

[0061] Figure 5 FIG. 1 is a flow chart of a hazardous waste logistics information management method according to an embodiment of the present application. Figure 5 As shown, according to the hazardous waste logistics information management method of the embodiment of the present application, it includes: S110, acquiring transfer form image information, basic information on hazardous waste characteristics and hazardous waste transportation information collected by a camera; S120, extracting transfer form association feature vectors and hazardous waste information association feature vectors from the transfer form image information, the basic information on hazardous waste characteristics and the hazardous waste transportation information collected by the camera; S130, generating a hazardous waste logistics information report based on the transfer form association feature vectors and the hazardous waste information association feature vectors.

[0062] Here, those skilled in the art will appreciate that the specific operations of each step in the above-mentioned hazardous waste logistics information management method have been referred to above. Figures 1 to 4 The description of the hazardous waste logistics information management system has been introduced in detail, and therefore, its repeated description will be omitted.

[0063] Below, reference Figure 6 To describe an electronic device according to an embodiment of the present application.

[0064] Figure 6 It is a structural block diagram of an electronic device according to an embodiment of the present invention.

[0065] like Figure 6 As shown, the electronic device 10 includes: a processor 11 and a memory 13. The processor 11 and the memory 13 are connected, such as through a bus 12. Optionally, the electronic device 10 may also include a transceiver 14. It should be noted that in actual applications, the transceiver 14 is not limited to one, and the structure of the electronic device 10 does not constitute a limitation on the embodiments of the present invention.

[0066] The processor 11 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It may implement or execute various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of the present invention. The processor 11 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and the like.

[0067] The bus 12 may include a path for transmitting information between the above components. The bus 12 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus 12 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 6 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0068] The memory 13 is used to store a computer program corresponding to the hazardous waste logistics information management method of the above embodiment of the present invention, and the computer program is controlled and executed by the processor 11. The processor 11 is used to execute the computer program stored in the memory 13 to implement the contents shown in the above method embodiment.

[0069] Among them, the electronic device 10 includes but is not limited to: mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 6 The electronic device 10 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0070] It should be noted that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection portion with one or more wirings (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways if necessary, and then stored in a computer memory.

[0071] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0072] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0073] In the description of the present invention, it is to be understood that the terms “center”, “longitudinal”, “lateral”, “length”, “width”, “thickness”, “up”, “down”, “front”, “back”, “left”, “right”, “vertical”, “horizontal”, “top”, “bottom”, “inside”, “outside”, “clockwise”, “counterclockwise”, “axial”, “radial”, “circumferential”, etc., indicating orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.

[0074] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0075] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements, unless otherwise clearly defined. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0076] In the present invention, unless otherwise clearly specified and limited, a first feature being "above" or "below" a second feature may mean that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium. Moreover, a first feature being "above", "above" or "above" a second feature may mean that the first feature is directly above or obliquely above the second feature, or simply means that the first feature is higher in level than the second feature. A first feature being "below", "below" or "below" a second feature may mean that the first feature is directly below or obliquely below the second feature, or simply means that the first feature is lower in level than the second feature.

[0077] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. A hazardous waste logistics information management system, characterized in that: include: The hazardous waste logistics information acquisition module is used to obtain the transfer form image information collected by the camera, basic information on hazardous waste characteristics and hazardous waste transportation information; A hazardous waste logistics information extraction module, used to extract a transfer form-related feature vector and a hazardous waste information-related feature vector from the transfer form image information collected by the camera, the basic information of the hazardous waste characteristics and the hazardous waste transportation information; The hazardous waste logistics information report generation module is used to generate a hazardous waste logistics information report based on the transfer form association feature vector and the hazardous waste information association feature vector.

2. The hazardous waste logistics information management system according to claim 1 is characterized in that: The hazardous waste logistics information extraction module includes: A transfer form image information feature extraction unit, used for extracting features from the transfer form image information collected by the camera to obtain a transfer form associated feature vector; A hazardous waste characteristic feature extraction unit, used for extracting the characteristics of the basic information of hazardous waste characteristics to obtain a global hazardous waste characteristic basic information feature vector; A hazardous waste transportation information feature extraction unit, used to extract features from the hazardous waste transportation information to obtain a semantically associated feature vector of the hazardous waste transportation information; The hazardous waste information feature association unit is used to associate the global hazardous waste characteristic basic information feature vector with the hazardous waste transportation information semantic association feature vector to obtain the hazardous waste information association feature vector.

3. The hazardous waste logistics information management system according to claim 2 is characterized in that: The transfer form image information feature extraction unit comprises: A transfer form image data preprocessing subunit, used for performing data preprocessing on the transfer form image information collected by the camera to obtain a plurality of transfer form image regions of interest; The transfer-link association non-local convolution subunit is used to pass the regions of interest of the multiple transfer-link images through a transfer-link non-local neural network as a feature encoder to obtain the transfer-link association feature vector.

4. The hazardous waste logistics information management system according to claim 3 is characterized in that: The transfer form image data preprocessing subunit includes: Performing image segmentation on the transfer form image information collected by the camera to obtain a plurality of transfer form image areas; The multiple transfer link image regions are passed through a transfer link target detection network to obtain the multiple transfer link image regions of interest.

5. The hazardous waste logistics information management system according to claim 4 is characterized in that: The hazardous waste characteristic feature extraction unit comprises: Segmenting the basic information on characteristics of hazardous waste to obtain a plurality of basic information data items on characteristics of hazardous waste; Arranging a plurality of basic information data items of hazardous waste characteristics in a two-dimensional matrix to obtain a global basic information characteristic matrix of hazardous waste characteristics; The global hazardous waste characteristics basic information feature matrix is ​​passed through a global hazardous waste characteristics basic information convolutional neural network to obtain the global hazardous waste characteristics basic information feature vector.

6. The hazardous waste logistics information management system according to claim 5 is characterized in that: The hazardous waste transportation information feature extraction unit comprises: Passing the hazardous waste transportation information through a converter-based hazardous waste transportation information semantic encoder to obtain a plurality of hazardous waste transportation information semantic feature vectors; The multiple hazardous waste transportation information semantic feature vectors are cascaded to obtain the hazardous waste transportation information semantic association feature vector.

7. The hazardous waste logistics information management system according to claim 6 is characterized in that: The hazardous waste logistics information report generation module includes: A hazardous waste logistics information feature fusion unit, used to fuse the transfer form association feature vector and the hazardous waste information association feature vector to obtain a hazardous waste logistics information generation feature vector; A hazardous waste logistics information feature optimization unit, used for performing group distribution constraints on hazardous waste logistics information generation feature vectors based on inherent modal information to obtain optimized hazardous waste logistics information generation feature vectors; The logistics information report generating unit is used to generate a hazardous waste logistics information report by passing the optimized hazardous waste logistics information generating feature vector through a generator.

8. The hazardous waste logistics information management system according to claim 7 is characterized in that: The hazardous waste logistics information feature optimization unit includes: Calculating the mean and variance of the characteristic vector generated by the hazardous waste logistics information; Calculating the difference between the characteristic value of the predetermined position of the hazardous waste logistics information generating characteristic vector and the mean value of the hazardous waste logistics information generating characteristic vector to obtain a first intermediate calculated value; Multiplying the network parameter matrix by the characteristic vector generated by the hazardous waste logistics information, and then subtracting the matrix from the characteristic vector generated by the hazardous waste logistics information by position to obtain a first intermediate vector, and calculating the Frobenius norm of the first intermediate vector to obtain a first norm value; Add the first norm value to the variance of the characteristic vector generated by the hazardous waste logistics information and then take the square root to obtain a second intermediate calculated value; dividing the first intermediate calculated value by the second intermediate calculated value to obtain a third intermediate calculated value; Calculating the square of each position in the hazardous waste logistics information generating feature vector and then accumulating and summing them to obtain a first total value; Calculate a natural exponential function value after taking the square root of the first sum value and dividing it by the scale of the network parameter matrix to obtain a fourth intermediate calculation value, wherein the scale of the network parameter matrix is ​​equal to the product of the width value of the network parameter matrix and the height value thereof; The fourth intermediate calculated value is multiplied by the weighted hyperparameter and then added to the third intermediate calculated value to obtain the characteristic value of the predetermined position of the characteristic vector generated by optimizing the hazardous waste logistics information.

9. A method for managing hazardous waste logistics information, characterized in that: include: Obtain transfer form image information, basic information on hazardous waste characteristics, and hazardous waste transportation information collected by the camera; Extracting a transfer form-related feature vector and a hazardous waste information-related feature vector from the transfer form image information collected by the camera, the basic information of hazardous waste characteristics, and the hazardous waste transportation information; Based on the transfer form association feature vector and the hazardous waste information association feature vector, a hazardous waste logistics information report is generated.

10. The hazardous waste logistics information management method according to claim 9, characterized in that: Extracting a transfer form-related feature vector and a hazardous waste information-related feature vector from the transfer form image information collected by the camera, the basic information of hazardous waste characteristics, and the hazardous waste transportation information, including: Performing feature extraction on the transfer form image information collected by the camera to obtain the transfer form associated feature vector; Extracting features of the basic information of hazardous waste characteristics to obtain a global basic information feature vector of hazardous waste characteristics; Extracting features of the hazardous waste transportation information to obtain a semantically associated feature vector of the hazardous waste transportation information; The global hazardous waste characteristic basic information feature vector and the hazardous waste transportation information semantic association feature vector are associated to obtain the hazardous waste information association feature vector.

Images