Logistics storage video monitoring optimization scheduling method and system in Internet of Things environment

Abstract

The invention discloses a logistics storage video monitoring optimization scheduling method and system in an Internet of Things environment. The method comprises the following steps: acquiring dynamic information of a logistics storage area; inputting the dynamic information into an enhanced image processing model to obtain high-quality image data, and generating a computing resource scheduling scheme; predicting a cargo flow trend and planning a monitoring strategy by using the computing resource scheduling scheme, and ensuring sufficient computing resource support in a high-flow area; when an abnormal event is detected, adjusting the working state of a camera or increasing the monitoring density of a specific area according to a preset task priority and an optimized resource allocation scheme rapid response; and collecting operation effect data as feedback, and continuously learning and optimizing an intelligent computing resource allocation algorithm. By implementing the method provided by the invention, the burden of the data center can be reduced, the real-time performance is improved, the automation and anomaly detection capabilities are improved, the system resource allocation is optimized, the storage cost is reduced, and the system efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of Internet of Things, and more specifically to a method and system for optimizing and scheduling logistics warehousing video monitoring under an Internet of Things environment. Background Art

[0002] In modern logistics and warehousing environments, video surveillance systems are an important part of ensuring cargo safety, improving operational efficiency, and optimizing inventory management. However, with the popularity of high-definition cameras and the sharp increase in video data volume, traditional video surveillance systems face a series of challenges, especially in resource management, data storage, real-time monitoring, and intelligent scheduling. The following is a detailed description of these issues: With the popularity of HD and UHD cameras, the image quality and data volume of video surveillance have shown exponential growth. HD cameras bring a large amount of video data, and traditional systems face storage and processing pressure. The surge in data volume affects the real-time performance and quality of the system. Large amounts of data require efficient storage and transmission strategies. Otherwise, video delays or loss may occur, affecting the monitoring effect. Traditional systems operate on a fixed plan and fail to adjust resources according to real-time needs. Intelligent scheduling can dynamically optimize the monitoring frequency according to environmental changes. The lack of this mechanism may lead to uneven resource allocation. Monitoring of key areas may be affected, resulting in blind spots. HD video takes up a lot of storage space and has high transmission bandwidth requirements. Traditional storage and transmission solutions are difficult to carry large amounts of data. Unoptimized storage and transmission will cause system delays and data loss. Large amounts of monitoring data need to be efficiently managed to ensure stability. Traditional systems can only record passively and lack automatic anomaly detection functions. Warehouse environments need to identify potential safety hazards and operational anomalies. Video analysis technology can automatically identify suspicious activities and reduce manual intervention. Intelligent analysis improves monitoring efficiency and accuracy. Video data requires high real-time performance, and traditional systems may have delays. The centralized architecture causes data transmission delays, affecting monitoring results. Real-time processing and response are critical to warehouse safety. The system needs to optimize its architecture to improve response speed. The amount of video data is huge, and manual backtracking is time-consuming and inefficient. Intelligent retrieval technology can quickly locate the time and place of the incident. Video labeling and indexing can improve data management efficiency. Intelligent backtracking reduces manual intervention and improves event processing efficiency. The integration of monitoring systems with other management systems often encounters compatibility issues. Data sharing and collaboration between different systems are not smooth. The system needs to have higher compatibility to promote the collaborative work of various systems. Optimizing integration can improve overall management efficiency.

[0003] Therefore, it is necessary to design a new method to reduce the burden on data centers, improve real-time performance, enhance automation and anomaly detection capabilities, optimize system resource allocation, reduce storage costs, and improve system efficiency. Summary of the invention

[0004] The purpose of the present invention is to overcome the defects of the prior art and provide a method and system for optimizing the scheduling of logistics warehousing video monitoring in an Internet of Things environment.

[0005] To achieve the above purpose, the present invention adopts the following technical solution: a logistics warehousing video monitoring optimization scheduling method under the Internet of Things environment, comprising: Obtain dynamic information of logistics and warehousing areas; Inputting the dynamic information into an enhanced image processing model to obtain high-quality image data, wherein the enhanced image processing model is trained based on logistics scenarios and automatically adjusts RGB color values ​​according to changes in illumination within a specific time period; Establishing a task graph structure according to the type and priority of the high-quality image data analysis task, and generating a computing resource scheduling solution using a hierarchical attention graph neural network and a particle swarm optimization model; Use the computing resource scheduling scheme to predict cargo flow trends and plan monitoring strategies to ensure sufficient computing resources support in high-flow areas; When an abnormal event is detected, it can respond quickly based on the pre-set task priority and optimized resource allocation plan, adjust the camera working status or increase the monitoring density of a specific area; Collect operational performance data as feedback, and continuously learn and optimize intelligent computing resource allocation algorithms.

[0006] Its further technical solution is: the dynamic information includes the video stream captured by each camera, the real-time data provided by the environmental sensor, and the cargo location determined by tracking the cargo movement path through RFID tags or barcode scanners and combining with the GPS positioning system.

[0007] The further technical solution is: the training process of the enhanced image processing model includes: Collect image data in logistics warehousing environments, which should cover different time periods, weather conditions, and different types of lighting conditions; Annotate the collected image data, including object bounding boxes and category labels; Generate image samples under synthetic lighting conditions by combining the annotated image data with image enhancement technology to obtain a sample set; Build the basic model; Define the loss function and optimizer; The loss function and the optimizer are used to train the basic model using the sample set to obtain an enhanced image processing model.

[0008] The further technical solution is as follows: the basic model includes a convolutional neural network, a light perception module is added to the convolutional neural network, and the light perception model receives environmental information such as light intensity by introducing an additional input layer, or uses an attention mechanism to focus on areas with large light changes in the image; A color adjustment layer is added to the convolutional neural network, which is responsible for dynamically modifying the RGB color value of the image according to the information provided by the illumination perception module. The color adjustment layer is based on traditional color space conversion theory or directly predicts the target color parameters through an end-to-end learning method.

[0009] Its further technical solution is: the task graph structure is established according to the type and priority of the high-quality image data analysis task, and a computing resource scheduling solution is generated using a hierarchical attention graph neural network and a particle swarm optimization model, including: performing task classification and priority setting according to the high-quality image data to obtain task characteristics; Constructing a task graph structure according to the task characteristics; In the task graph structure, a hierarchical attention graph neural network is applied to update the task features to obtain a task feature matrix; The particle swarm optimization model is used in combination with the task feature matrix to generate a computing resource scheduling solution.

[0010] A further technical solution is: in the task graph structure, a hierarchical attention graph neural network is applied to update the task features to obtain a task feature matrix, including: Starting from the task graph structure, extracting an initial task feature vector, wherein the task feature vector includes task type, priority, estimated execution time and resource requirement information; According to the initial task feature vector, a graph neural network with a hierarchical attention mechanism is used to iteratively update the features of the nodes in the task graph structure, capture the complex interaction patterns between tasks, and highlight the task paths that are critical to the overall goal, so as to obtain a task feature matrix.

[0011] A further technical solution is: using the computing resource scheduling scheme to predict cargo flow trends and plan monitoring strategies to ensure sufficient computing resource support in high-flow areas, including: Preprocessing and feature extraction are performed on the dynamic information to obtain extraction results; Inputting the extraction results and the high-quality image data into a prediction model to identify potential trend patterns to obtain a flow trend prediction result; Based on the traffic trend prediction results, the computing resources required for different areas in the future are estimated to support the effective execution of video surveillance and other related tasks. In combination with the computing resource scheduling plan, the corresponding computing resources are pre-deployed in the predicted high-traffic areas to ensure that all monitoring equipment in these areas can obtain the necessary support to perform efficient image processing and data analysis.

[0012] The present invention also provides a logistics warehousing video monitoring optimization scheduling system under the Internet of Things environment, including: A dynamic information acquisition unit, used to acquire dynamic information of the logistics storage area; An image enhancement unit, configured to input the dynamic information into an enhanced image processing model to obtain high-quality image data, wherein the enhanced image processing model is trained based on logistics scenarios and automatically adjusts RGB color values ​​according to changes in illumination within a specific time period; A solution generation unit, used to establish a task graph structure according to the type and priority of the high-quality image data analysis task, and generate a computing resource scheduling solution using a hierarchical attention graph neural network and a particle swarm optimization model; A planning unit, used to use the computing resource scheduling scheme to predict cargo flow trends and plan monitoring strategies to ensure sufficient computing resource support in high-flow areas; The adjustment unit is used to quickly respond to abnormal events according to the pre-set task priority and optimized resource allocation plan, adjust the camera working status or increase the monitoring density of a specific area; The collection unit is used to collect operation effect data as feedback, and continuously learn and optimize the intelligent computing resource allocation algorithm.

[0013] The present invention further provides a computer device, which includes a memory and a processor. The memory stores a computer program, and the processor implements the above method when executing the computer program.

[0014] The present invention also provides a storage medium, wherein the storage medium stores a computer program, and the computer program implements the above method when executed by a processor.

[0015] Compared with the prior art, the present invention has the following beneficial effects: the present invention optimizes image quality through an enhanced image processing model, automatically adjusts RGB color values ​​under changing illumination, reduces data redundancy and storage pressure during image processing, and improves image data quality; based on high-quality image data analysis task priorities and types, a computing resource scheduling scheme is generated using a hierarchical attention graph neural network and a particle swarm optimization model to achieve real-time task optimization and reduce computing resource waste; predicts cargo flow trends and plans monitoring strategies to ensure that high-flow areas receive sufficient computing resource support, avoids excessive consumption of computing resources, and improves system response speed; when an abnormal event occurs, quickly responds and adjusts the camera's working status or increases monitoring density to ensure that potential problems are discovered and handled in a timely manner, and improves automation and anomaly detection capabilities; by collecting and feeding back operating effect data, the intelligent computing resource allocation algorithm is continuously optimized to continuously improve system efficiency, reduce storage costs, and reduce the burden on data centers.

[0016] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying any creative work.

[0018] Figure 1 A schematic diagram of an application scenario of a logistics warehousing video monitoring optimization scheduling method under an Internet of Things environment provided by an embodiment of the present invention; Figure 2 A schematic diagram of a flow chart of a logistics warehousing video monitoring optimization scheduling method under an Internet of Things environment provided by an embodiment of the present invention; Figure 3 A schematic block diagram of a logistics warehousing video monitoring optimization and scheduling system under an Internet of Things environment provided by an embodiment of the present invention; Figure 4 A schematic block diagram of a computer device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0019] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] It should be understood that when used in this specification and the appended claims, the terms "include" and "comprises" indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0021] It should also be understood that the terms used in this specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0022] It should be further understood that the term "and / or" used in the present description and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0023] See also Figure 1 and Figure 2 , Figure 1 Schematic diagram of application scenario of the logistics warehousing video surveillance optimization scheduling method under the Internet of Things environment provided by an embodiment of the present invention. Figure 2 A schematic flow chart of a logistics warehousing video surveillance optimization scheduling method under an Internet of Things environment provided by an embodiment of the present invention. The logistics warehousing video surveillance optimization scheduling method under the Internet of Things environment is applied to a server, which interacts with sensors, cameras and terminals for data, optimizes image quality through an enhanced image processing model, automatically adjusts RGB color values ​​under illumination changes, reduces data redundancy and storage pressure during image processing, and improves image data quality; based on high-quality image data analysis task priority and type, hierarchical attention graph neural network and particle swarm optimization model are used to generate computing resource scheduling schemes to achieve real-time task optimization and reduce computing resource waste; predict cargo flow trends and plan monitoring strategies to ensure that high-flow areas receive sufficient computing resource support, avoid excessive consumption of computing resources, and improve system response speed; when abnormal events occur, quickly respond and adjust the camera working state or increase monitoring density to ensure timely discovery and handling of potential problems, and improve automation and abnormality detection capabilities; by collecting and feeding back operation effect data, continuously optimize intelligent computing resource allocation algorithms, continuously improve system efficiency, reduce storage costs, and reduce the burden on data centers.

[0024] Figure 2 1 is a flow chart of a method for optimizing and scheduling logistics warehousing video monitoring in an Internet of Things environment provided by an embodiment of the present invention. Figure 2 As shown, the method includes the following steps S110 to S160.

[0025] S110. Obtain dynamic information of the logistics storage area.

[0026] In this embodiment, the dynamic information includes video streams captured by various cameras, real-time data provided by environmental sensors, and the location of the goods determined by tracking the movement path of the goods through RFID tags or barcode scanners and combining with the GPS positioning system.

[0027] Specifically, comprehensive and real-time data is collected to reflect the current status within the logistics warehousing environment, providing a basis for subsequent image processing, task analysis and resource scheduling.

[0028] Camera video stream: Multiple high-definition cameras deployed in the logistics warehouse area capture continuous video images of key locations. The video content covers activities such as cargo loading and unloading, handling, storage, and the operations of staff.

[0029] Environmental sensors are used to monitor physical parameters such as temperature, humidity, light intensity, etc., and upload this information to the central control system in real time. This data is very important for understanding the conditions inside the warehouse, especially when it is necessary to adjust image quality or perform anomaly detection.

[0030] RFID tag / barcode scanner and GPS positioning system: Utilize the RFID tag or barcode installed on the goods, and read its unique identifier through the scanner when the goods enter or leave the warehouse or move. Combined with GPS positioning technology, the location change path of each piece of goods can be accurately tracked, realizing visual management of the entire logistics process.

[0031] All sensors and monitoring devices are connected to a unified IoT platform, which automatically sends updated measurements at regular intervals. When specific events occur (such as cargo arriving at a designated location or unusual activity being detected), the system immediately triggers additional data recording to ensure that important moments are not missed.

[0032] The collected data contains noise or incomplete parts, so it needs to be cleaned and formatted before further processing. The video stream is initially filtered to remove unnecessary background information; the environmental sensor data is smoothed to eliminate the impact of transient fluctuations.

[0033] Data from different sources must be accurately synchronized according to timestamps to ensure the correct time relationship between them. All kinds of data are integrated into a comprehensive database so that they can be easily accessed and used in subsequent steps.

[0034] Implement strict security measures during transmission and storage to protect sensitive information from unauthorized access or tampering. Comply with relevant laws and regulations to ensure that personal privacy is properly protected.

[0035] In summary, step S110 aims to build an efficient and reliable data acquisition system, which can not only capture the rich dynamic details in the logistics warehousing area, but also lay a solid foundation for the subsequent task analysis and intelligent decision-making. This step is the premise and starting point of the entire logistics warehousing video surveillance optimization scheduling method, and its accuracy directly affects the performance of the entire system.

[0036] S120. Input the dynamic information into an enhanced image processing model to obtain high-quality image data, wherein the enhanced image processing model is trained based on logistics scenarios and automatically adjusts RGB color values ​​according to lighting changes within a specific time period.

[0037] In this embodiment, the training process of the enhanced image processing model includes: Collect image data in logistics warehousing environments, which should cover different time periods, weather conditions, and different types of lighting conditions; Annotate the collected image data, including object bounding boxes and category labels; Generate image samples under synthetic lighting conditions by combining the annotated image data with image enhancement technology to obtain a sample set; Build the basic model; Define the loss function and optimizer; The loss function and the optimizer are used to train the basic model using the sample set to obtain an enhanced image processing model.

[0038] In this embodiment, the original video stream is processed by an enhanced image processing model to adapt to changes under different lighting conditions, thereby outputting clearer and more accurate image data to provide support for subsequent tasks (such as object recognition, path planning, etc.).

[0039] The model can automatically adjust the RGB color value according to the lighting changes in a specific time period, ensuring good visual effects even in low-light or high-contrast environments. Advanced machine learning algorithms such as convolutional neural networks (CNNs) are used to improve image quality.

[0040] In order to make the model widely applicable, a large amount of image data needs to be collected from logistics warehousing environments. These images should cover different time periods of the day (morning, noon, evening), various weather conditions (sunny, cloudy, rainy), and different lighting conditions (natural light, artificial light). Ensure that the collected data is representative and maintains a certain proportion under various conditions to avoid the model being biased towards a specific situation.

[0041] Draw detailed bounding boxes for key elements in each image (such as goods, shelves, people, etc.) and assign corresponding category labels (such as "carton", "pallet", "forklift driver"). This step is crucial for training the model to correctly identify and classify targets. Semi-automated annotation tools can be used to improve efficiency, but manual review is still required to ensure accuracy.

[0042] Image enhancement techniques (such as brightness adjustment, contrast enhancement, noise removal, etc.) are combined with labeled real image data to generate a series of synthetic image samples under extreme lighting conditions (such as extremely dark or extremely bright environments). This method not only increases the number of training samples, but also improves the model's ability to cope with complex lighting environments.

[0043] According to the task requirements, select a suitable CNN or other type of deep learning network as the basic model architecture, such as ResNet, EfficientNet, etc. They have demonstrated excellent performance in various computer vision tasks.

[0044] Determine the number of network layers, filter size, activation function and other hyperparameters to prepare for subsequent training.

[0045] Define an appropriate loss function to measure the difference between the predicted result and the true label. Common ones include cross entropy loss, mean square error loss, etc. For this task, you may consider adding additional constraints to encourage the model to generate a more natural image color distribution. Select an efficient optimization algorithm (such as Adam, SGD) to minimize the loss function, accelerate convergence and find the global optimal solution.

[0046] Divide the constructed sample set into multiple small batches, input them into the model in turn for forward propagation to calculate the loss value, and then back propagate to update the weight parameters. Regularly evaluate the performance of the model on an independent validation set, adjust the hyperparameters or improve the model structure according to the results until satisfactory accuracy is achieved.

[0047] The enhanced image processing model obtained through the above training can receive real-time video streams from the camera and quickly output high-quality image data after optimization processing. These images have better color reproduction and detail expression, which helps to improve the accuracy and reliability of subsequent analysis tasks.

[0048] This step is one of the key technical links in the entire logistics warehousing video surveillance optimization scheduling method. It is directly related to whether the system can effectively extract valuable information and make correct decisions.

[0049] In this embodiment, the basic model includes a convolutional neural network, and a light perception module is added to the convolutional neural network. The light perception model receives environmental information such as light intensity by introducing an additional input layer, or uses an attention mechanism to focus on areas with large light changes in the image; A color adjustment layer is added to the convolutional neural network, which is responsible for dynamically modifying the RGB color value of the image according to the information provided by the illumination perception module. The color adjustment layer is based on traditional color space conversion theory or directly predicts the target color parameters through an end-to-end learning method.

[0050] The introduction of the light perception module and color adjustment layer in the enhanced image processing model is an important improvement for optimizing image quality under complex and changeable lighting conditions in logistics scenarios. The addition of the light perception module is to improve the ability of the convolutional neural network (CNN) to understand images under different lighting conditions, thereby improving the quality of the output image.

[0051] In order to make CNN aware of information such as the light intensity in the environment, one or more additional input layers are added to the network architecture. These layers can receive external information such as light intensity data provided by light sensors, timestamps (for inferring the time of day), data returned by weather forecast APIs, etc.

[0052] Another approach is to embed an attention mechanism in CNN, which allows the network to automatically focus on those image areas that are more affected by lighting changes. By learning which parts need more attention, the network can self-adjust the weight distribution during training to more effectively handle local lighting differences.

[0053] The information provided by the illumination perception module can help CNN better understand the background conditions of the input image, and then more accurately extract useful features. It also provides necessary parameters and guidance for the subsequent color adjustment layer, ensuring that the color adjustment process is both scientific and efficient.

[0054] Using classic color theory, such as the conversion from RGB to Lab color space, to achieve dynamic adjustment of color values. This method relies on predefined mathematical formulas and has the advantages of fast calculation speed and good stability.

[0055] Through deep learning technology, the network can learn how to adjust the color of the image according to the changes in lighting conditions. This method does not require pre-setting complex conversion rules, but allows the model to learn the optimal color mapping relationship from a large number of training samples.

[0056] After receiving the information transmitted by the light perception module, the color adjustment layer can immediately adjust the RGB color value of the image to compensate for the color deviation caused by light changes. This ensures that the adjusted image not only meets the requirements in terms of technical indicators, but also has a natural and realistic visual effect without obvious color distortion.

[0057] The above improvements are integrated into the existing CNN framework to form a more intelligent and adaptable enhanced image processing model. This model can not only significantly improve the quality of images in logistics and warehousing environments, but may also be applicable to other fields, such as autonomous driving, security monitoring, etc., and any application scenario with high requirements for image quality and challenges of lighting changes can benefit from it.

[0058] In this embodiment, a photosensitive sensor is installed near the image acquisition device (such as a camera) to directly measure the ambient light intensity. These sensors can provide accurate lighting information in real time, including but not limited to parameters such as brightness and color temperature. Ensure that the data of the photosensitive sensor can be synchronized with the image frame, and perform necessary preprocessing on the collected data, such as filtering and denoising, unit conversion, etc., for subsequent use.

[0059] A subnetwork dedicated to estimating lighting conditions is trained, which can extract lighting-related features from the input image, such as shadow distribution, highlight area, etc., and then infer the current lighting state. Considering that lighting changes may be reflected at different scales, multi-scale convolution kernels can be used to capture lighting patterns from local details to the overall scene.

[0060] The introduction of the attention mechanism enables CNN to automatically identify and strengthen the attention of areas that are greatly affected by lighting during the learning process. By adjusting the attention weights, the network can focus more on the lighting characteristics of key parts, thereby improving the ability to understand complex lighting environments. Combined with the lighting perception results, a dynamic feature map is generated to guide subsequent color adjustment layers or other processing steps to respond to lighting changes more accurately.

[0061] Using data from external sources such as weather forecast services and geolocation information as supplementary information sources helps the network better predict and adapt to different lighting conditions. For example, the sun altitude angle can be estimated based on the geolocation and timestamp to indirectly understand the lighting conditions.

[0062] If you choose to obtain lighting information through external sensors, you can add one or more additional input channels to the existing CNN architecture to specifically receive this information. The advantage of doing this is that the lighting data can be involved in the calculation process of the entire network together with other image features. For situations where lighting estimation relies entirely on image content, the lighting perception function can be embedded in a specific layer of the CNN to form an independent but closely related submodule. This module is responsible for analyzing the input image and outputting estimates of light intensity and distribution for use by subsequent layers.

[0063] Specifically, the lighting perception module and the color adjustment layer work together to ensure that the image maintains good visual effects under different lighting conditions. Specifically, this is achieved through the following aspects: The light perception module is responsible for analyzing the lighting conditions in the environment or image, including but not limited to brightness, color temperature, direction of light, etc. It can extract relevant information based on data from external sensors or directly from the image content. The acquired light information is converted into a form suitable for subsequent processing, such as standardized values, feature vectors, etc., so that it can be understood and used by the color adjustment layer.

[0064] Based on the information provided by the light perception module, the color adjustment layer can apply predefined rules to adjust the color of the image. For example, increasing the brightness in low-light environments, or adjusting the white balance to compensate for the blue cast at high color temperatures (such as daylight). More advanced systems may use machine learning models to dynamically decide how to adjust the color. These models can learn the best color adjustment scheme based on historical data and automatically adjust parameters for different lighting conditions.

[0065] After completing an adjustment, the color adjustment layer can evaluate the adjustment effect in some way and feed the result back to the light perception module. This helps to continuously improve the accuracy of light estimation and the effect of color adjustment. As time goes by and more data is accumulated, the entire system can continuously optimize its performance through online learning or offline training, enabling it to provide more accurate color adjustment services in various complex lighting environments.

[0066] Some applications may require special processing for specific types of scenes (such as indoor, outdoor, night scenes, etc.). At this time, the light perception module can help identify the current scene type and notify the color adjustment layer to adopt the corresponding strategy. Allow users to set some basic color adjustment parameters such as contrast, saturation, etc. according to their personal preferences. The light perception module and the color adjustment layer should jointly consider these preferences to meet the personalized needs of users while ensuring image quality.

[0067] In this embodiment, the illumination perception module can extract relevant information based on data from external sensors or directly from image content. In order to improve the accuracy of illumination estimation, it is recommended to integrate the two information sources using the following method: First, the system makes a preliminary assessment of the current lighting conditions by analyzing the brightness, contrast, and other visual features in the input image. This step can quickly provide a rough overview of the lighting environment. Then, the preliminary estimate is calibrated and refined using the precise light intensity measurements provided by the photosensitive sensor installed near the camera. The photosensitive sensor can directly capture the actual light level of the environment, including parameters such as brightness, color temperature, and direction, thereby providing a more accurate basis for subsequent color adjustments. Finally, the light perception module will integrate information from image analysis and photosensitive sensors to generate a comprehensive light status report. The color adjustment layer dynamically adjusts the RGB color values ​​of the image based on this report to ensure good visual effects even under complex and changing lighting conditions.

[0068] Assume that in a smart camera system, the light perception module detects that the current environment is dusk and there is obvious backlighting. It passes this information to the color adjustment layer, which performs the following operations based on it: Increase overall brightness to improve visibility of dark details. Adjust white balance to reduce color deviation caused by changes in color temperature. Apply HDR (High Dynamic Range) technology to balance the difference between bright and dark areas. If there are objects in shadow, it may also be necessary to locally enhance contrast to make foreground objects clearer.

[0069] Through such a collaborative working mechanism, the lighting perception module and the color adjustment layer can effectively cope with complex and changing lighting conditions and provide users with a high-quality visual experience.

[0070] S130. Establish a task graph structure according to the type and priority of the high-quality image data analysis task, and generate a computing resource scheduling plan using a hierarchical attention graph neural network and a particle swarm optimization model.

[0071] In one embodiment, the above-mentioned step S130 may include steps S131 - S134 .

[0072] S131, performing task classification and priority setting according to the high-quality image data to obtain task characteristics; S132, constructing a task graph structure according to the task characteristics; S133. In the task graph structure, apply a hierarchical attention graph neural network to update task features to obtain a task feature matrix.

[0073] In one embodiment, the above-mentioned step S133 may include steps S1331 - S1332 .

[0074] S1331, starting from the task graph structure, extracting an initial task feature vector, wherein the task feature vector includes task type, priority, estimated execution time and resource requirement information; S1332. Based on the initial task feature vector, a graph neural network with a hierarchical attention mechanism is used to iteratively update the features of the nodes in the task graph structure, capture the complex interaction patterns between tasks, and highlight the task paths that are critical to the overall goal, so as to obtain a task feature matrix.

[0075] S134. Generate a computing resource scheduling plan by using the particle swarm optimization model in combination with the task feature matrix.

[0076] In this embodiment, the input high-quality image data is first analyzed to determine the type of each task (e.g., edge detection, color correction, object recognition, etc.) and its relative importance or urgency (i.e., priority). The result of this process is the definition of the unique characteristics of each task, which are critical for subsequent resource allocation.

[0077] Based on the task characteristics obtained in S131 and the dependencies between them, a task graph structure is created to represent all tasks to be performed and their relationships. This helps to visualize the logical order between different tasks and possible concurrent operations. Specifically, a corresponding node is created in the task graph for each image analysis task. For tasks involving multi-stage processing, subtask nodes are further subdivided. Task nodes are connected by directed edges to express the sequence or dependency between tasks. For example, only after the object recognition in a specific area is completed, the quantity statistics of the goods in the area can be started. Appropriate weights are given to nodes and edges to reflect the time required for the task, resource consumption, and its importance in the entire workflow. This helps to plan the task execution path more accurately later.

[0078] Starting from the constructed task graph, the initial task feature vector is extracted. These features include information such as task type, priority, estimated execution time and resource requirements. A graph neural network with a hierarchical attention mechanism is used to iteratively update the features of the nodes in the task graph. HAGNN can capture the complex interaction patterns between tasks and highlight those task paths that are critical to the overall goal, thereby generating a more optimized task feature matrix. Specifically, the basic attributes of each task, such as task type, priority, estimated execution time and required computing resource information, are extracted from the task graph structure. This information forms the basis of the task feature vector. Using a graph neural network with a hierarchical attention mechanism (HAGNN), the node features in the task graph structure are iteratively updated multiple times. This not only captures the complex interaction patterns between tasks, but also highlights those paths or nodes that are critical to completing the entire task. The final output is an optimized task feature matrix.

[0079] The last step is to combine the previously obtained task feature matrix and apply the particle swarm optimization (PSO) algorithm to determine how to optimally allocate available computing resources to each task. PSO is a heuristic search algorithm that imitates the foraging behavior of a flock of birds to find the global optimal solution in the solution space. In this context, PSO is used to find the best resource allocation strategy to ensure that efficiency and effectiveness are maximized while meeting all task requirements. Specifically, in the PSO framework, each "particle" represents a possible task scheduling solution. The position of the particle corresponds to which edge computing device each task is assigned to and its start time. A reasonable fitness function is designed to quantify the quality of each solution. The fitness function takes into account multiple dimensions such as task completion efficiency and resource utilization to ensure that the optimal solution is selected. The PSO model receives the task feature matrix from HAGNN as one of the inputs, and uses these features to guide the search direction of the particle swarm, accelerate convergence and increase the probability of finding the global optimal solution. After multiple iterations, the particle swarm continuously adjusts its position and finally determines a scheduling solution that can both meet the task priority and efficiently utilize existing resources.

[0080] The optimal task scheduling scheme derived from the PSO model is applied to the actual environment to reasonably allocate computing resources to each image analysis task. The flexibility of the system is maintained, and when new tasks are added or the status of existing tasks changes, it can respond quickly and recalculate the optimal scheduling scheme. The execution of tasks is continuously tracked, performance indicator data is collected, and a feedback mechanism is formed to guide the optimization of resource allocation strategies in similar scenarios in the future.

[0081] In summary, the above process not only takes into account the characteristics of the image data analysis task itself, but also fully combines advanced machine learning algorithms and intelligent optimization technologies, aiming to achieve efficient computing resource management and ensure the fast and accurate completion of image processing tasks in the logistics warehousing system.

[0082] In this embodiment, although the PSO model imitates the foraging behavior of a flock of birds, its core advantages lie in its parallel search capability and its ability to avoid falling into local optimality. Specifically: The PSO algorithm allows multiple "particles" (representing different task scheduling schemes) to explore different parts of the solution space at the same time. This parallelism increases the possibility of finding the global optimal solution, especially when facing large-scale or complex optimization problems. PSO helps the system escape the local optimal trap and continue to look for better solutions through the interaction and information sharing mechanism between particles. For example, in the logistics warehousing scenario, when considering both task priority and resource utilization, PSO can help the system balance the relationship between the two goals and find the best allocation strategy that can meet the needs of urgent tasks and make efficient use of existing resources. Suppose that multiple image processing tasks need to be completed in a limited time. The PSO model will quickly determine an optimal task scheduling scheme based on task characteristics (such as type, priority, expected execution time and resource requirements) and existing computing resources. This scheme not only ensures the timely processing of critical tasks, but also maximizes the efficiency of the overall system.

[0083] In addition to deep learning algorithms such as LSTM and GRU, considering the seasonal and cyclical characteristics in logistics scenarios, technologies specifically for time series prediction such as Prophet and ARIMA can also be introduced. In addition, a regular update mechanism should be established to adapt to the ever-changing business environment: For cargo flow forecasting tasks with obvious periodicity, you can choose a dedicated time series forecasting tool such as Prophet. It is good at processing data sets containing trends, seasonality, and holiday effects, and can capture long-term patterns more accurately. For non-periodic short-term fluctuations, neural network models such as LSTM or GRU can be used.

[0084] In order to ensure that the prediction results are always close to the actual situation, the system should set a fixed period (such as weekly or monthly) to retrain the prediction model with the latest collected data. This can not only reflect the latest business dynamics, but also gradually improve the expressiveness of the model.

[0085] Throughout the process, the system needs to continuously track the prediction performance and feed back the actual operation data to the model for further optimization and adjustment. For example, if it is found that the prediction error in a certain time period is large, the learning intensity of this part of the data can be strengthened to improve the prediction accuracy.

[0086] S140. Utilize the computing resource scheduling scheme to predict cargo flow trends and plan monitoring strategies to ensure sufficient computing resource support in high-flow areas.

[0087] In one embodiment, the above-mentioned step S140 may include steps S141 - S143 .

[0088] S141, preprocessing and feature extraction of the dynamic information to obtain extraction results; S142, inputting the extraction result and the high-quality image data into a prediction model to identify potential trend patterns to obtain a flow trend prediction result; S143. Estimate the computing resources required for different areas in the future based on the traffic trend prediction results to support the effective execution of video surveillance and other related tasks. Pre-deploy corresponding computing resources in the predicted high-traffic areas in combination with the computing resource scheduling plan to ensure that all monitoring equipment in these areas can obtain the necessary support to perform efficient image processing and data analysis.

[0089] In this embodiment, first, the system needs to accumulate high-quality image data within a certain period of time, and these data include but are not limited to timestamps of goods entering and leaving the warehouse, types of goods, transportation routes and other information.

[0090] Extract features from the collected data that help analyze cargo flow trends, such as cargo throughput in different time periods, activity in a specific area, etc. This step may also involve preliminary processing of the image data, such as object recognition, trajectory tracking, etc.

[0091] Using machine learning or deep learning algorithms (such as LSTM, GRU, etc.), a cargo flow prediction model is built based on historical data and extracted features. The model can predict cargo flow at a certain moment or time in the future based on the patterns observed in the past.

[0092] The trained model can regularly output forecast results of cargo flow trends in the next few days or even weeks. This will help decision makers understand in advance which periods may see traffic peaks and adjust monitoring strategies accordingly.

[0093] Combined with the traffic trend forecast results, identify areas that are expected to become traffic hotspots. These areas may be cargo loading and unloading areas, channel intersections, or other places prone to congestion. For the identified hotspots, pre-allocate sufficient computing resources according to the previously formulated computing resource scheduling plan. This means ensuring that there are enough edge computing devices available in these areas and that they have been configured to prioritize tasks from these areas. Taking into account the uncertainty in actual operations, a dynamic adjustment mechanism also needs to be established. When real-time monitoring data shows that the actual situation deviates from the forecast, it can respond quickly, re-evaluate resource requirements and make corresponding adjustments.

[0094] Implement new monitoring strategies as planned, increase monitoring efforts during expected high-traffic periods, and ensure that all related equipment is operating normally. Continue to track the performance of various indicators, including task completion efficiency, resource utilization, etc. The collected data is not only used for improvement measures within the current cycle, but also provides a more accurate basis for future traffic forecasts. With the accumulation of more data and the advancement of technology, the prediction model and resource scheduling algorithm are continuously optimized, making the prediction accuracy of the entire system higher and the resource utilization more reasonable.

[0095] For key areas identified as high traffic, ensure that they have the highest priority computing resource support. This may mean deploying additional high-performance computing nodes in these areas, or temporarily deploying idle resources in other areas when necessary. In addition to daily resource planning, you must also be prepared to respond to emergencies. For example, when an unexpected event causes an area to suddenly become a traffic hotspot, you can quickly launch an emergency plan to immediately supplement the required computing power.

[0096] In summary, through the above process, not only can the cargo flow trend be effectively predicted, but also the monitoring strategy can be reasonably planned to ensure that key areas receive sufficient computing resource support during peak periods, thereby improving the operational efficiency and service quality of the entire logistics warehousing system.

[0097] In addition, feature engineering is performed based on historical data, and a cargo flow prediction model is trained using machine learning or deep learning algorithms. In this step, the "computing resource scheduling solution" ensures that computing resources for data processing and model training are sufficient and efficiently utilized.

[0098] Use the trained model to predict areas that may become traffic peaks in the future. This step relies on powerful computing power to quickly process large amounts of data and generate predictions. Based on the predictions, the "Computing Resource Scheduling Scheme" will allocate additional computing resources to these hot spots in advance. This means that edge servers, cloud services, etc. will be optimally configured to ensure that sufficient processing power and storage space can be provided instantly when needed. When the actual traffic pattern does not match the prediction, the "Computing Resource Scheduling Scheme" allows the system to dynamically adjust resource allocation. For example, if a high traffic suddenly occurs in an unexpected area, some idle resources can be quickly transferred to that area through an automated scheduler to ensure that its monitoring and processing needs are met.

[0099] For predicted high-traffic areas, the "Computing Resource Scheduling Plan" guides how to optimally deploy monitoring equipment and technologies, such as high-definition cameras and smart sensors. Ensure that these devices can obtain the necessary computing support to perform image recognition, data analysis, and other tasks. Continuously monitor the performance of various indicators and feed back actual operating data to the "Computing Resource Scheduling Plan" to improve future resource allocation decisions. In addition, this also helps to continuously optimize the prediction model and improve prediction accuracy.

[0100] The computing resource scheduling solution has designed a flexible resource expansion mechanism that can quickly increase computing resources in specific areas in emergency situations. For example, during promotional activities or when encountering unexpected events, more resources can be deployed in time to cope with the surge in workload. In order to prevent any single point of failure from affecting the stability of the entire system, the "Computing Resource Scheduling Solution" also includes redundancy design and disaster recovery plans to ensure that even if some resources fail, the normal execution of critical tasks can be maintained.

[0101] In this embodiment, the collected dynamic information is first preprocessed, such as cleaning noise, filling missing values, etc., and then useful features are extracted from it. These features may include but are not limited to time period, weather conditions, special events (such as holidays), historical traffic patterns, etc., which are crucial for understanding the current traffic status and predicting future trends.

[0102] The feature extraction results obtained in S141 are input into a pre-trained prediction model together with high-quality image data. The model can be a machine learning or deep learning-based algorithm that is used to identify potential trend patterns. By analyzing this data, the model can output a prediction result about future cargo flow trends, which helps to understand in advance which areas may experience high traffic.

[0103] Based on the traffic trend prediction results in S142, the amount of computing resources required in different areas in the future is estimated. In order to support the effective execution of video surveillance and other related tasks, it is necessary to pre-deploy sufficient computing resources in areas where high traffic is expected based on the prediction results. This means not only considering the existing infrastructure conditions, but also flexibly adjusting resource allocation to ensure that all monitoring equipment can obtain the necessary support for efficient image processing and data analysis. In addition, it is also possible to combine the previous computing resource scheduling plan to optimize the efficiency of resource use and avoid resource waste or shortage.

[0104] This predictive maintenance and resource management approach not only improves the responsiveness and service quality of the logistics system, but also enhances the adaptability and stability of the system in the face of emergencies. In this way, companies can more effectively manage and optimize their logistics networks, improve operational efficiency, reduce costs, and provide customers with a better service experience.

[0105] S150. When an abnormal event is detected, respond quickly according to the pre-set task priority and the optimized resource allocation plan, adjust the camera working status or increase the monitoring density of a specific area.

[0106] In this embodiment, when an abnormal event (such as a traffic accident, illegal intrusion, etc.) is detected, the configuration of the monitoring system can be quickly adjusted according to the pre-set task priority and the optimized resource allocation plan to ensure the safety and operational efficiency of key areas.

[0107] Each monitoring task is assigned a priority, which determines which tasks should be handled first when resources are limited. For example, when an emergency is detected, the working status of the relevant camera will be immediately adjusted to the highest priority in order to provide the clearest real-time picture. Based on the previous prediction results of cargo flow trends and other dynamic information, the system has developed an optimized resource allocation plan. This means that when an abnormal event occurs, the system can quickly mobilize the necessary computing resources to where they are needed, such as increasing the number of cameras in a specific area or improving the video resolution to obtain more detailed monitoring information. In order to better capture and analyze abnormal events, change the angle, focal length or sensitivity settings of certain cameras, or start the recording function. In addition, mobile monitoring equipment can be temporarily deployed to enhance the coverage of specific areas.

[0108] In this embodiment, after detecting an abnormal event, not only the working state of the camera should be adjusted, but other types of sensors should also be considered to form a multimodal monitoring system. This can provide more comprehensive information support when different types of emergencies occur: When an abnormal event is detected, the system not only relies on the visual information provided by the video stream, but also combines the physical signals provided by devices such as smoke detectors and thermometers. For example, in a fire warning scenario, in addition to observing the video footage of flames and smoke, you can also refer to the temperature change curve recorded by the thermometer and the alarm signal from the smoke detector to jointly confirm whether there is a fire.

[0109] Based on the comprehensive analysis results of multimodal data, the system can flexibly adjust the monitoring strategy. For example, when a large number of people suddenly gather in a certain area, in addition to improving the resolution and frame rate of the camera, the infrared thermal imager in the nearby area can be activated to obtain a more dimensional human body heat distribution map to more accurately assess the on-site situation.

[0110] A complete set of emergency plans is prepared in advance, covering all possible abnormal situations. Once the corresponding conditions are triggered, the system can immediately execute the preset operation process, such as notifying relevant personnel, blocking dangerous areas, dispatching rescue forces, etc., to ensure that safety measures are quickly in place.

[0111] S160. Collect operation effect data as feedback, and continuously learn and optimize the intelligent computing resource allocation algorithm.

[0112] By collecting and analyzing data on system operation performance as feedback, the intelligent computing resource allocation algorithm is continuously improved, allowing the system to become smarter and more efficient over time.

[0113] The operation effect data includes but is not limited to indicators such as system response time, image quality, recognition accuracy, and resource utilization. These data not only reflect the performance of the current system, but also provide an important reference for subsequent optimization. Using machine learning technology, especially reinforcement learning algorithms, the system learns from each operation and automatically adjusts parameters to adapt to new challenges. As more data accumulates, the system's predictive ability and decision-making level will continue to improve. Based on the collected data, regularly evaluate the effectiveness of existing computing resource allocation algorithms, and introduce new models or improve existing models to ensure that they are always in the best state. This may involve updating prediction models, adjusting task priority rules, and optimizing resource scheduling strategies.

[0114] This self-optimization process is the key to building an adaptive and intelligent logistics monitoring system. It not only improves the flexibility and responsiveness of the system, but also promotes the maximization of resource utilization, thereby bringing greater value to the enterprise. At the same time, it also reflects the important role of modern technology in improving safety and operational efficiency.

[0115] In this example, the decision-making ability of the system is gradually improved through a well-defined reward function and a trial-and-error process: In the logistics warehousing environment, the design of the reward function should focus on the quality of task completion, resource utilization efficiency, and service response speed. For example, successfully completing a high-priority task on time can get a positive reward; while failing to respond to an emergency request in time or wasting too much computing resources will be punished negatively. In this way, the system is encouraged to make better decisions.

[0116] After each operation, the system will adjust its internal parameters based on the actual results and continuously try new resource allocation strategies. As experience accumulates, the system gradually learns how to better balance the relationship between various indicators and find the most appropriate solution. For example, when encountering a traffic peak, the system may try to increase the number of edge servers, observe the impact on task completion time and resource consumption, and then decide whether to promote this practice.

[0117] The entire process forms a self-optimizing closed loop, where the system learns from every operation and automatically adjusts parameters to adapt to new challenges. Over time, the system's predictive capabilities and decision-making levels will continue to improve, bringing greater value to the enterprise. For example, after multiple iterations, the system was able to significantly reduce energy consumption and hardware costs with little impact on service quality.

[0118] For example, suppose there is a large logistics warehouse with multiple high-definition cameras, environmental sensors (monitoring temperature, humidity, etc.), RFID tag readers, and GPS positioning systems deployed inside. The warehouse receives a large amount of goods in and out every day, forming a complex and dynamic working environment.

[0119] The system collects data from various devices around the clock, including video streams, environmental parameters and cargo location information. All data is uploaded to the central server synchronously and pre-processed to remove noise and redundancy. Then, the data is integrated into a comprehensive database, providing a solid foundation for subsequent steps.

[0120] For images taken at different time periods and weather conditions, the system uses a pre-trained enhanced image processing model to automatically adjust the RGB color values ​​to ensure high-quality images even in low-light or high-contrast environments. During this process, the light perception module analyzes the ambient light intensity in real time and guides the color adjustment layer to make corresponding modifications.

[0121] Based on the optimized image data, the system identifies the type of task currently being performed and its priority. Then, a task graph structure is constructed to represent the dependencies between tasks, and the task features are iteratively updated through a hierarchical attention graph neural network. Finally, a particle swarm optimization algorithm is used to determine the best computing resource allocation scheme to ensure that key tasks are processed in a timely manner.

[0122] Using historical data and machine learning algorithms, the system predicts cargo flow trends over the next few days. Based on the forecast results, additional computing resources are deployed in advance in areas where high traffic is expected to occur, ensuring that monitoring equipment in these areas can operate efficiently.

[0123] When the system detects an abnormal event (such as too many people suddenly gathering in a certain area), it will immediately activate the emergency response procedure. Specifically, the working status of the relevant cameras is adjusted to the highest priority, and the monitoring density of the area is increased to quickly obtain more detailed information for subsequent analysis.

[0124] The system regularly evaluates its performance and adjusts its internal algorithms based on actual operating performance data. For example, if a certain area is found to frequently have false positives, the parameters of the lighting perception module or the color adjustment layer can be fine-tuned to reduce such occurrences. Over time, the entire system will become smarter and more efficient.

[0125] The above-mentioned logistics warehousing video surveillance optimization scheduling method under the Internet of Things environment optimizes image quality through an enhanced image processing model, automatically adjusts the RGB color value under lighting changes, reduces data redundancy and storage pressure during image processing, and improves image data quality; based on high-quality image data analysis task priority and type, hierarchical attention graph neural network and particle swarm optimization model are used to generate computing resource scheduling solutions to achieve real-time task optimization and reduce computing resource waste; predict cargo flow trends and plan monitoring strategies to ensure that high-flow areas receive sufficient computing resource support, avoid excessive consumption of computing resources, and improve system response speed; when abnormal events occur, respond quickly and adjust the camera working status or increase the monitoring density to ensure timely discovery and handling of potential problems, and enhance automation and anomaly detection capabilities; by collecting and feeding back operation effect data, continuously optimize the intelligent computing resource allocation algorithm, continuously improve system efficiency, reduce storage costs, and reduce the burden on data centers.

[0126] Figure 3 3 is a schematic block diagram of a logistics warehousing video monitoring optimization scheduling system 300 in an Internet of Things environment provided by an embodiment of the present invention. Figure 3 As shown, corresponding to the above logistics warehousing video monitoring optimization scheduling method under the Internet of Things environment, the present invention also provides a logistics warehousing video monitoring optimization scheduling system 300 under the Internet of Things environment. The logistics warehousing video monitoring optimization scheduling system 300 under the Internet of Things environment includes a unit for executing the above logistics warehousing video monitoring optimization scheduling method under the Internet of Things environment, and the system can be configured in a server. Specifically, please refer to Figure 3The logistics warehousing video monitoring optimization and scheduling system 300 under the Internet of Things environment includes a dynamic information acquisition unit 301, an image enhancement unit 302, a solution generation unit 303, a planning unit 304, an adjustment unit 305 and a collection unit 306.

[0127] The dynamic information acquisition unit 301 is used to obtain dynamic information of the logistics storage area; the image enhancement unit 302 is used to input the dynamic information into the enhanced image processing model to obtain high-quality image data, wherein the enhanced image processing model is trained based on the logistics scenario and automatically adjusts the RGB color value according to the illumination change in a specific time period; the solution generation unit 303 is used to establish a task graph structure according to the task type and priority of the high-quality image data analysis, and generate a computing resource scheduling plan using a hierarchical attention graph neural network and a particle swarm optimization model; the planning unit 304 is used to use the computing resource scheduling plan to predict the trend of cargo flow and plan a monitoring strategy to ensure sufficient computing resource support in high-flow areas; the adjustment unit 305 is used to respond quickly to abnormal events when they are detected, adjust the camera working status or increase the monitoring density of a specific area according to the pre-set task priority and the optimized resource allocation plan; the collection unit 306 is used to collect operation effect data as feedback, and continuously learn and optimize the intelligent computing resource allocation algorithm.

[0128] In one embodiment, the system further comprises a training unit, configured to: Collect image data in a logistics warehousing environment, and the images should cover different time periods, weather conditions and different types of lighting conditions; annotate the collected image data, including object bounding boxes and category labels; generate image samples under synthetic lighting conditions by combining the annotated image data with image enhancement technology to obtain a sample set; build a basic model; define a loss function and an optimizer; use the loss function and the optimizer to train the basic model with the sample set to obtain an enhanced image processing model.

[0129] In one embodiment, the solution generation unit 303 is used to: Task classification and priority setting are performed according to the high-quality image data to obtain task characteristics; a task graph structure is constructed according to the task characteristics; in the task graph structure, a hierarchical attention graph neural network is applied to update task characteristics to obtain a task characteristic matrix; and a computing resource scheduling plan is generated by combining the particle swarm optimization model with the task characteristic matrix.

[0130] In one embodiment, the solution generating unit 303 is further configured to: Starting from the task graph structure, an initial task feature vector is extracted, and the task feature vector includes task type, priority, estimated execution time and resource requirement information; based on the initial task feature vector, a graph neural network with a hierarchical attention mechanism is used to iteratively update the features of the nodes in the task graph structure, capture the complex interaction patterns between tasks, and highlight the task paths that are critical to the overall goal, so as to obtain a task feature matrix.

[0131] In one embodiment, the planning unit 304 is used to: The dynamic information is preprocessed and feature extracted to obtain extraction results; the extraction results and the high-quality image data are input into a prediction model to identify potential trend patterns to obtain traffic trend prediction results; based on the traffic trend prediction results, the computing resources required for different areas in the future are estimated to support the effective execution of video surveillance and other related tasks, and the corresponding computing resources are pre-deployed in the predicted high-traffic areas in combination with the computing resource scheduling plan to ensure that all monitoring equipment in these areas can obtain the necessary support to perform efficient image processing and data analysis.

[0132] It should be noted that technical personnel in the relevant field can clearly understand that the specific implementation process of the logistics warehousing video surveillance optimization scheduling system 300 and each unit under the above-mentioned Internet of Things environment can refer to the corresponding description in the aforementioned method embodiment, and for the convenience and conciseness of the description, it will not be repeated here.

[0133] The logistics warehousing video monitoring optimization scheduling system 300 under the above-mentioned Internet of Things environment can be implemented in the form of a computer program. The computer program can be used in Figure 4 Runs on the computer device shown.

[0134] See also Figure 4 , Figure 4 5 is a schematic block diagram of a computer device provided in an embodiment of the present application. The computer device 500 may be a server, wherein the server may be an independent server or a server cluster composed of multiple servers.

[0135] See also Figure 4 The computer device 500 includes a processor 502 , a memory and a network interface 505 connected via a system bus 501 , wherein the memory may include a non-volatile storage medium 503 and an internal memory 504 .

[0136] The non-volatile storage medium 503 can store an operating system 5031 and a computer program 5032. The computer program 5032 includes program instructions, and when the program instructions are executed, the processor 502 can execute a logistics warehousing video monitoring optimization scheduling method under an Internet of Things environment.

[0137] The processor 502 is used to provide computing and control capabilities to support the operation of the entire computer device 500 .

[0138] The internal memory 504 provides an environment for the operation of the computer program 5032 in the non-volatile storage medium 503. When the computer program 5032 is executed by the processor 502, the processor 502 can execute a logistics warehousing video monitoring optimization scheduling method under the Internet of Things environment.

[0139] The network interface 505 is used to communicate with other devices over the network. Figure 4 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present application, and does not constitute a limitation on the computer device 500 to which the solution of the present application is applied. The specific computer device 500 may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0140] The processor 502 is used to run the computer program 5032 stored in the memory to implement the following steps: Acquire dynamic information of logistics storage areas; input the dynamic information into an enhanced image processing model to obtain high-quality image data, wherein the enhanced image processing model is trained based on logistics scenarios and automatically adjusts RGB color values ​​according to changes in illumination within a specific time period; establish a task graph structure based on task types and priorities according to the high-quality image data analysis, and generate a computing resource scheduling plan using a hierarchical attention graph neural network and a particle swarm optimization model; use the computing resource scheduling plan to predict cargo flow trends and plan monitoring strategies to ensure sufficient computing resource support in high-flow areas; when an abnormal event is detected, respond quickly based on pre-set task priorities and optimized resource allocation plans, adjust the camera working status or increase the monitoring density of a specific area; collect operating performance data as feedback, and continuously learn and optimize the intelligent computing resource allocation algorithm.

[0141] It should be understood that in the embodiment of the present application, the processor 502 may be a central processing unit (CPU), and the processor 502 may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0142] It can be understood by those skilled in the art that all or part of the processes in the method for implementing the above embodiment can be completed by instructing the relevant hardware through a computer program. The computer program includes program instructions, and the computer program can be stored in a storage medium, which is a computer-readable storage medium. The program instructions are executed by at least one processor in the computer system to implement the process steps of the embodiment of the above method.

[0143] Therefore, the present invention also provides a storage medium. The storage medium may be a computer-readable storage medium. The storage medium stores a computer program, wherein when the computer program is executed by a processor, the processor executes the following steps: Acquire dynamic information of logistics storage areas; input the dynamic information into an enhanced image processing model to obtain high-quality image data, wherein the enhanced image processing model is trained based on logistics scenarios and automatically adjusts RGB color values ​​according to changes in illumination within a specific time period; establish a task graph structure based on task types and priorities according to the high-quality image data analysis, and generate a computing resource scheduling plan using a hierarchical attention graph neural network and a particle swarm optimization model; use the computing resource scheduling plan to predict cargo flow trends and plan monitoring strategies to ensure sufficient computing resource support in high-flow areas; when an abnormal event is detected, respond quickly based on pre-set task priorities and optimized resource allocation plans, adjust the camera working status or increase the monitoring density of a specific area; collect operating performance data as feedback, and continuously learn and optimize the intelligent computing resource allocation algorithm.

[0144] The storage medium may be a USB flash drive, a mobile hard disk, a read-only memory (ROM), a magnetic disk, or an optical disk, etc., which are computer-readable storage media that can store program codes.

[0145] It should be noted that the functions or steps that can be implemented by the above storage medium or computer device can be correspondingly referred to the relevant descriptions on the server side and the client side in the aforementioned method embodiment. In order to avoid repetition, they will not be described one by one here.

[0146] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0147] In the several embodiments provided by the present invention, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are only schematic. For example, the division of each unit is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed.

[0148] The steps in the method of the embodiment of the present invention can be adjusted in order, combined and deleted according to actual needs. The units in the system of the embodiment of the present invention can be combined, divided and deleted according to actual needs. In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0149] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, terminal, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention.

[0150] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A logistics warehousing video monitoring optimization scheduling method under the Internet of Things environment, characterized by: include: Obtain dynamic information of logistics and warehousing areas; Inputting the dynamic information into an enhanced image processing model to obtain high-quality image data, wherein the enhanced image processing model is trained based on logistics scenarios and automatically adjusts RGB color values ​​according to changes in illumination within a specific time period; Establishing a task graph structure according to the type and priority of the high-quality image data analysis task, and generating a computing resource scheduling solution using a hierarchical attention graph neural network and a particle swarm optimization model; Use the computing resource scheduling scheme to predict cargo flow trends and plan monitoring strategies to ensure sufficient computing resources support in high-flow areas; When an abnormal event is detected, it can respond quickly based on the pre-set task priority and optimized resource allocation plan, adjust the camera working status or increase the monitoring density of a specific area; Collect operational performance data as feedback, and continuously learn and optimize intelligent computing resource allocation algorithms.

2. The method for optimizing and scheduling logistics warehousing video monitoring under the Internet of Things environment according to claim 1 is characterized in that: The dynamic information includes video streams captured by various cameras, real-time data provided by environmental sensors, and the location of the goods determined by tracking the movement path of the goods through RFID tags or barcode scanners and combining with the GPS positioning system.

3. The method for optimizing and scheduling logistics warehousing video monitoring under the Internet of Things environment according to claim 2 is characterized in that: The training process of the enhanced image processing model includes: Collect image data in logistics warehousing environments, which should cover different time periods, weather conditions, and different types of lighting conditions; Annotate the collected image data, including object bounding boxes and category labels; Generate image samples under synthetic lighting conditions by combining the annotated image data with image enhancement technology to obtain a sample set; Build the basic model; Define the loss function and optimizer; The loss function and the optimizer are used to train the basic model using the sample set to obtain an enhanced image processing model.

4. The method for optimizing and scheduling logistics warehousing video monitoring under the Internet of Things environment according to claim 3 is characterized in that: The basic model includes a convolutional neural network, and a light perception module is added to the convolutional neural network. The light perception model receives environmental information such as light intensity by introducing an additional input layer, or uses an attention mechanism to focus on areas with large light changes in the image; A color adjustment layer is added to the convolutional neural network, which is responsible for dynamically modifying the RGB color value of the image according to the information provided by the illumination perception module. The color adjustment layer is based on traditional color space conversion theory or directly predicts the target color parameters through an end-to-end learning method.

5. The method for optimizing and scheduling logistics warehousing video monitoring under the Internet of Things environment according to claim 4 is characterized in that: The step of establishing a task graph structure according to the type and priority of the high-quality image data analysis task, and generating a computing resource scheduling solution using a hierarchical attention graph neural network and a particle swarm optimization model, includes: performing task classification and priority setting according to the high-quality image data to obtain task characteristics; Constructing a task graph structure according to the task characteristics; In the task graph structure, a hierarchical attention graph neural network is applied to update the task features to obtain a task feature matrix; The particle swarm optimization model is used in combination with the task feature matrix to generate a computing resource scheduling solution.

6. The method for optimizing and scheduling logistics warehousing video monitoring in an Internet of Things environment according to claim 5 is characterized in that: In the task graph structure, a hierarchical attention graph neural network is applied to update the task features to obtain a task feature matrix, including: Starting from the task graph structure, extracting an initial task feature vector, the task feature vector including task type, priority, estimated execution time and resource requirement information; According to the initial task feature vector, a graph neural network with a hierarchical attention mechanism is used to iteratively update the features of the nodes in the task graph structure, capture the complex interaction patterns between tasks, and highlight the task paths that are critical to the overall goal, so as to obtain a task feature matrix.

7. The method for optimizing and scheduling logistics warehousing video monitoring in an Internet of Things environment according to claim 6 is characterized in that: The computing resource scheduling scheme is used to predict cargo flow trends and plan monitoring strategies to ensure sufficient computing resources support in high-flow areas, including: Preprocessing and feature extraction are performed on the dynamic information to obtain extraction results; Inputting the extraction results and the high-quality image data into a prediction model to identify potential trend patterns to obtain a flow trend prediction result; Based on the traffic trend prediction results, the computing resources required for different areas in the future are estimated to support the effective execution of video surveillance and other related tasks. In combination with the computing resource scheduling plan, the corresponding computing resources are pre-deployed in the predicted high-traffic areas to ensure that all monitoring equipment in these areas can obtain the necessary support to perform efficient image processing and data analysis.

8. Logistics warehousing video monitoring optimization and scheduling system under the Internet of Things environment, characterized by: include: A dynamic information acquisition unit, used to acquire dynamic information of the logistics storage area; An image enhancement unit, configured to input the dynamic information into an enhanced image processing model to obtain high-quality image data, wherein the enhanced image processing model is trained based on logistics scenarios and automatically adjusts RGB color values ​​according to changes in illumination within a specific time period; A solution generation unit, used to establish a task graph structure according to the type and priority of the high-quality image data analysis task, and generate a computing resource scheduling solution using a hierarchical attention graph neural network and a particle swarm optimization model; A planning unit, used to use the computing resource scheduling scheme to predict cargo flow trends and plan monitoring strategies to ensure sufficient computing resource support in high-flow areas; The adjustment unit is used to quickly respond to abnormal events according to the pre-set task priority and optimized resource allocation plan, adjust the camera working status or increase the monitoring density of a specific area; The collection unit is used to collect operation effect data as feedback, and continuously learn and optimize the intelligent computing resource allocation algorithm.

9. A computer device, characterized in that: The computer device comprises a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the method according to any one of claims 1 to 7 is implemented.

10. A storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images