Chemical industry park gas leakage auxiliary decision-making method and device, system storage medium and electronic equipment

Through the improved YOLOv5s network model and CA attention mechanism, combined with video data, the generalization ability and stability of gas leakage detection methods in chemical parks is solved, and more accurate and stable gas leakage detection and emergency response decisions are achieved.

CN118968384APending Publication Date: 2024-11-15XINJIANG UNIVERSITY
View PDF 6 Cites 0 Cited by

Patent Information

Application Number
CN202411128550.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-16
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

The existing gas leak detection methods in chemical parks are based on specific sensors, with limited generalization capabilities, which are difficult to adapt to different scenarios, and are sensitive to environmental changes and have poor stability, making them difficult to effectively apply in chemical parks.

Method used

The improved YOLOv5s network model is used for deep learning, combined with the CA attention mechanism, and the location and degree of gas leakage are obtained through video data, matching emergency response decisions, and improving detection accuracy and stability.

Benefits of technology

It improves the accuracy and stability of gas leakage detection, can quickly and effectively obtain emergency response decisions, and is suitable for the rapid disposal of gas leakage in chemical parks.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118968384A_ABST
    Figure CN118968384A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of chemical leakage detection, in particular to a chemical industrial park gas leakage auxiliary decision-making method and device, a system storage medium and electronic equipment, and the method comprises the steps: obtaining to-be-detected data of a chemical industrial park; inputting the to-be-detected picture into the gas leakage detection model to obtain an identification result of the gas leakage position and degree; and matching the gas leakage position and degree of the to-be-detected picture with a preset emergency response decision set to obtain a corresponding emergency response decision. The improved YOLOv5s network model is used for completing gas leakage detection, important features can be better captured and utilized, the expression ability of the features is enhanced, the calculation amount is low, the complexity and the calculation amount of the model cannot be increased, and therefore the accuracy and the stability of gas leakage detection are improved, and the detection efficiency is improved. And matching with a preset emergency response decision set is carried out to obtain a corresponding emergency response decision, so that effective support is provided for rapid disposal of gas leakage in the chemical industry park.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of chemical leakage detection, and is a chemical park gas leakage auxiliary decision-making method, device, system storage medium and electronic equipment. Background Art

[0002] Gas leaks in chemical parks may lead to a variety of serious consequences. Different types of gas leaks may cause serious harm to human health. For example, toxic gases such as ammonia, chlorine, sulfur dioxide, etc. may cause breathing difficulties, eye irritation, skin burns and even death. In addition, gas leaks may pollute the surrounding environment, leading to soil pollution, water pollution, harm to plants and animals, and even have long-term effects on the ecosystem. Some gases are also flammable or explosive. If they leak and encounter an ignition source, they may cause a fire or explosion, resulting in casualties and damage to facilities. Therefore, gas leaks may cause factory shutdowns, equipment damage, property losses, and legal liabilities, causing significant economic losses to the company.

[0003] At present, most gas leak detection methods are based on specific sensors and multiple technologies. The existing problems include: first, the detection requirements of different scenarios are different, so the generalization ability is limited, and specific gas sensors are required for leak detection. This method is difficult to adapt to new scenarios; second, it is sensitive to environmental changes. When the meteorological conditions change greatly, the sensor needs to be manually calibrated, and the parameters need to be manually set during the detection process. Therefore, this method has low robustness and poor stability, and is difficult to apply to chemical park scenarios. Summary of the invention

[0004] The present invention provides a chemical park gas leakage auxiliary decision-making method, device, system storage medium and electronic equipment, which overcome the shortcomings of the above-mentioned prior art and can effectively solve the problems that the existing gas leakage detection method based on specific sensors has specific scene requirements and is sensitive to environmental changes, resulting in poor generalization ability and poor stability.

[0005] One of the technical solutions of the present invention is achieved through the following measures: a chemical park gas leakage auxiliary decision-making method, comprising: Obtain the data to be detected in the chemical park, determine whether the data to be detected is video data, and if so, uniformly sample the data to be detected to obtain multiple images to be detected; Input the image to be detected into the gas leakage detection model to obtain the recognition result of the gas leakage location and degree, wherein the gas leakage detection model is obtained by deep learning the improved YOLOv5s network model using a number of samples, and each of the several samples includes a gas leakage image and a label identifying the gas leakage location and degree; The gas leakage location and degree of the gas leakage in the image to be detected are matched with the preset emergency response decision set to obtain the corresponding emergency response decision.

[0006] The following are further optimizations and / or improvements to the above technical solutions: The deep learning process of the above gas leak detection model includes: Obtain several image samples of different gas leakage degrees, label each image sample, and divide them into training set and test set according to the proportion after labeling, wherein the labeling information includes the gas leakage location and gas leakage degree; The improved YOLOv5s network model is deeply learned using the training set to obtain a gas leakage detection model, wherein the improved YOLOv5s network model is based on the YOLOv5s network model with the CA attention mechanism added; The gas leak detection model is tested and evaluated using the test set, and the optimal gas leak detection model is output.

[0007] The above-mentioned image samples of several gas leakage levels are as follows: Set multiple camera positions at multiple shooting distances and obtain video data of a fixed duration at each camera position; Taking the shooting time of the video data as the reference variable of the gas leakage degree, N pictures with different leakage degrees are sampled from the video data of each shooting distance and each shooting position to form a picture sample library; Preprocess all image samples in the image sample library.

[0008] The above preprocesses all sampled image samples, including: Perform affine transformation, filtering, translation, and rotation on all image samples to expand the number of image samples; The expanded image samples are subjected to different degrees of sharpening and noise processing.

[0009] In the above, labelimg is used to mark each image sample, and the gas leakage position and gas leakage degree thereof are marked.

[0010] The second technical solution of the present invention is achieved by the following measures: a chemical park gas leakage auxiliary decision-making device, comprising: The detection data acquisition unit acquires the data to be detected in the chemical park, determines whether the data to be detected is video data, and if so, uniformly samples the data to be detected to obtain multiple images to be detected; The detection unit inputs the image to be detected into the gas leakage detection model to obtain a gas leakage degree recognition result, wherein the gas leakage detection model is obtained by deep learning an improved YOLOv5s network model using a plurality of samples, and each of the plurality of samples includes a gas leakage image and a label identifying a gas leakage location and degree; The decision matching unit matches the gas leakage position and gas leakage degree of the image to be detected with a preset emergency response decision set to obtain a corresponding emergency response decision.

[0011] The following are further optimizations and / or improvements to the above technical solutions: The above also includes a model building unit, including: The sample acquisition module acquires a number of image samples with different gas leakage degrees, labels each image sample, and divides it into a training set and a test set in proportion after labeling, wherein the labeling information includes the gas leakage location and the gas leakage degree; The deep learning module uses the training set to perform deep learning on the improved YOLOv5s network model to obtain a gas leakage detection model, wherein the improved YOLOv5s network model is based on the YOLOv5s network model and adds the CA attention mechanism; The model testing module uses the test set to test and evaluate the gas leak detection model and outputs the optimal gas leak detection model.

[0012] The third technical solution of the present invention is achieved by the following measures: a chemical park gas leakage auxiliary decision-making system, characterized in that it includes: A processing device, in which a chemical park gas leakage auxiliary decision-making device is stored and executed, and a corresponding visual interface is provided; The interactive device is used for the operator to communicate with the processing device, so that the operator provides the processing device with the data to be detected in the chemical park through the interactive unit.

[0013] The present invention is based on deep learning of the improved YOLOv5s network model to obtain a gas leakage detection model, and uses the gas leakage detection model to detect gas leakage in a chemical park, wherein the improved YOLOv5s network model adds a CA attention mechanism, and the CA attention mechanism not only considers channel information, but also considers direction-related position information. By dynamically adjusting the weight of the channel, the network can better capture and utilize important features, enhance the expressiveness of the features, and the low amount of calculation will not increase the complexity and amount of the model, thereby improving the accuracy and stability of gas leakage detection. Further, based on matching with a pre-set emergency response decision set, the corresponding emergency response decision is quickly and effectively obtained, providing effective support for the rapid disposal of gas leakage in chemical parks. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Attached Figure 1 It is a flow chart of the gas leakage decision-making assistance method of the present invention.

[0015] Attached Figure 2 It is a schematic diagram of the flow chart of the gas leakage detection model construction method of the present invention.

[0016] Attached Figure 3 This is a schematic diagram of partial training results of a model according to an embodiment of the present invention.

[0017] Attached Figure 4 It is a schematic diagram of the comparison results of the accuracy of various leakage levels according to an embodiment of the present invention.

[0018] Attached Figure 5 It is a schematic diagram of the structure of the gas leakage decision-making auxiliary device of the present invention.

[0019] Attached Figure 6 A schematic diagram of an image detection entry interface of the present invention.

[0020] Attached Figure 7 A schematic diagram of a video detection entry interface of the present invention.

[0021] Attached Figure 8 The figure is a schematic diagram of an image detection result interface of the present invention.

[0022] Attached Fig. 9 A schematic diagram of a video detection result interface of the present invention. DETAILED DESCRIPTION

[0023] The present invention is not limited by the following embodiments, and specific implementation methods can be determined based on the technical solution of the present invention and actual conditions.

[0024] The present invention will be further described below in conjunction with embodiments and drawings: Embodiment 1: As attached Figure 1As shown, the embodiment of the present invention discloses a chemical park gas leakage auxiliary decision-making method, comprising: Step S110, obtaining the data to be detected in the chemical park, determining whether the data to be detected is video data, and if so, uniformly sampling the data to be detected to obtain a plurality of images to be detected; In the above steps, the data to be detected at the key monitoring locations of the chemical park can be obtained through monitoring cameras. The data to be detected here can be image data or video data. Therefore, when obtaining the data to be detected in the chemical park, it is necessary to determine whether the data to be detected is video data. If it is video data, it is necessary to uniformly sample the data to be detected according to the set sampling frequency to obtain multiple pictures to be detected. If it is picture data, no processing is required.

[0025] Step S120, input the image to be detected into the gas leakage detection model to obtain the recognition result of the gas leakage location and degree, wherein the gas leakage detection model is obtained by deep learning the improved YOLOv5s network model using a plurality of samples, and each of the plurality of samples includes a gas leakage image and a label identifying the gas leakage location and degree.

[0026] Step S130, matching the gas leakage position and gas leakage degree of the image to be detected with a preset emergency response decision set to obtain a corresponding emergency response decision.

[0027] In the above steps, the emergency response decision set is pre-set, wherein a corresponding emergency response decision is set for each gas leakage level. Furthermore, the emergency response decision for each gas leakage level can be updated, deleted, and the like.

[0028] The present invention discloses a chemical park gas leakage auxiliary decision method, based on deep learning of an improved YOLOv5s network model to obtain a gas leakage detection model, and using the gas leakage detection model to detect gas leakage in the chemical park, wherein the improved YOLOv5s network model adds a CA attention mechanism, and the CA attention mechanism not only considers channel information, but also considers direction-related position information. By dynamically adjusting the weight of the channel, the network can better capture and utilize important features, enhance the expression ability of the features, and the low amount of calculation will not increase the complexity and amount of the model, thereby improving the accuracy of gas leakage detection, and further, based on matching with a pre-set emergency response decision set, the corresponding emergency response decision is quickly and effectively obtained, providing effective support for the rapid disposal of gas leakage in the chemical park.

[0029] Embodiment 2: As attached Figure 2As shown, the embodiment of the present invention discloses a gas leakage auxiliary decision-making method for a chemical park, which is a further optimization of the above embodiment, wherein the deep learning process of the gas leakage detection model further includes: Step S210, obtaining a number of image samples with different gas leakage degrees, labeling each image sample, and dividing them into a training set and a test set in proportion after labeling, wherein the labeling information includes the gas leakage location and the gas leakage degree.

[0030] In the above steps, several image samples of different gas leakage degrees are obtained, including: (1) Setting multiple camera positions at multiple shooting distances and obtaining video data of a fixed length at each camera position; For example, five shooting distances are set, namely 4.6m, 6.9m, 9.8m, 12.6m, and 15.6m, and five shooting positions are set at each shooting distance. Video data with a fixed duration of 24min to 26min is acquired at each shooting distance and each shooting position.

[0031] (2) Taking the shooting time of the video data as the reference variable of the gas leakage degree, N pictures with different leakage degrees are sampled from the video data of each shooting distance and each shooting position to form a picture sample library; For example, the leakage degree is divided into 7 levels, and the number of related pictures of different leakage degrees is N=m*8, and the number of picture samples in the picture sample library is m*8*5*5.

[0032] (3) Preprocess all image samples in the image sample library.

[0033] The above steps preprocess all sampled image samples, including: (a) Perform affine transformation, filtering, translation, and rotation on all image samples to expand the number of image samples. This step is used to expand the number of image samples, increase the richness of image samples, prevent overfitting of the model in deep learning, and improve the generalization ability of the model.

[0034] (b) The expanded image samples are subjected to different degrees of sharpening and noise processing. This step is used to simulate image samples with different clarity.

[0035] In the above steps, labelimg is used to mark each image sample to mark the gas leakage position and gas leakage degree therein. In specific implementation, the marked gas leakage degree levels may include Leaklevel_0, Leaklevel_1, Leaklevel_2, Leaklevel_3, Leaklevel_4, Leaklevel_5, Leaklevel_6, and Leaklevel_7.

[0036] Step S220, using the training set to perform deep learning on the improved YOLOv5s network model to obtain a gas leakage detection model, wherein the improved YOLOv5s network model is based on the YOLOv5s network model and adds a CA attention mechanism.

[0037] The above-mentioned improved YOLOv5s network model adds the CA attention mechanism on the basis of the YOLOv5s network model. The CA attention mechanism not only considers the channel information, but also the direction-related position information. By dynamically adjusting the weights of the channels, the network can better capture and utilize important features, enhance the expression ability of features, and the low computational complexity will not increase the complexity and computational complexity of the model, thereby improving the accuracy of gas leak detection.

[0038] Step S230, using the test set to test and evaluate the gas leakage detection model, and outputting the optimal gas leakage detection model.

[0039] In this embodiment, based on the improved YOLOv5s deep learning framework, the hardware and software conditions are: Ubuntu18.04 operating system, python3.8, pytorch1.10, Nvidia RTX A5000 GPU, 24GB RAM. The model is trained using the prepared data set, and the training process uses the CIOU loss function to supervise the training.

[0040] Development environment, as described below: (1) Hardware environment Processor: Intel(R) Core(TM) i5-9300H CPU @ 2.40GHz or equivalent; Memory: 512 MB RAM; Hard disk: 512 MB free hard disk space; Graphics Card: High-color display adapter with 1024x768 resolution.

[0041] (2) Software environment Operating system: Windows 7, macOS 10.11; Recommended configuration: Windows 10 or later, macOS 10.14 or later.

[0042] (3) Runtime library configuration matplotlib>=3.2.2, numpy>=1.18.5, opencv-python>=4.1.2, Pillow==8.4.0, PyYAML>=5.3.1, requests>=2.23.0, sci y>=1.4.1, torch>=1.7.0, torchvision>=0.8.1, tqdm>=4.41.0, tensorboard>=2.4.1, pandas>=1.1.4, seaborn>=0.11.0.

[0043] After 300 epochs of training, the model was tested and evaluated on the test set. The following indicators were used to evaluate the training results of the model: precision P (Precision), recall R (Recall) and average precision mAp. The model evaluation results are shown in Table 1. The model with the highest accuracy was selected as the gas leak detection model used. The experimental results are as follows: Some training results are shown in the attached figure. Figure 3 The accuracy comparison results of each leakage level are shown in the attached Figure 4 As shown, it can be seen that the best detection is no leakage, and the worst detection is level 3 leakage.

[0044] Embodiment 3: As attached Figure 5 As shown, the embodiment of the present invention discloses a chemical park gas leakage auxiliary decision-making device, comprising: The detection data acquisition unit acquires the data to be detected in the chemical park, determines whether the data to be detected is video data, and if so, uniformly samples the data to be detected to obtain multiple images to be detected; The detection unit inputs the image to be detected into the gas leakage detection model to obtain a gas leakage degree recognition result, wherein the gas leakage detection model is obtained by deep learning an improved YOLOv5s network model using a plurality of samples, and each of the plurality of samples includes a gas leakage image and a label identifying a gas leakage location and degree; A decision matching unit matches the gas leakage location and gas leakage degree of the image to be detected with a preset emergency response decision set to obtain a corresponding emergency response decision; Model building unit, including: The sample acquisition module acquires a number of image samples with different gas leakage degrees, labels each image sample, and divides it into a training set and a test set in proportion after labeling, wherein the labeling information includes the gas leakage location and the gas leakage degree; The deep learning module uses the training set to perform deep learning on the improved YOLOv5s network model to obtain a gas leakage detection model, wherein the improved YOLOv5s network model is based on the YOLOv5s network model and adds the CA attention mechanism; The model testing module uses the test set to test and evaluate the gas leak detection model and outputs the optimal gas leak detection model.

[0045] Embodiment 4: The embodiment of the present invention discloses a chemical park gas leakage auxiliary decision-making system, which is characterized by comprising: A processing device, in which the chemical park gas leakage auxiliary decision-making device as described in the above embodiment is stored and executed, and a corresponding visual interface is provided; The interactive device is used for the operator to communicate with the processing device, so that the operator provides the processing device with the data to be detected in the chemical park through the interactive unit.

[0046] The processing device may be a computer, in which a chemical park gas leakage auxiliary decision-making device is embedded, and a visualization interface is provided for the chemical park gas leakage auxiliary decision-making device. Further, the visualization interface for entering the chemical park gas leakage auxiliary decision-making device may include a picture detection entry interface and a video detection entry interface, as shown in the attached Figure 6 , 7 The corresponding test result interface can be shown in the attached Figure 8 , 9 shown.

[0047] Embodiment 5: The embodiment of the present invention discloses a storage medium, on which a computer program that can be read by a computer is stored, and the computer program is configured to execute a chemical park gas leakage auxiliary decision-making method when running.

[0048] The above storage medium may include, but is not limited to: a USB flash drive, a read-only memory, a mobile hard disk, a magnetic disk or an optical disk, and other media that can store computer programs.

[0049] Embodiment 6: The embodiment of the present invention discloses an electronic device, including a processor and a memory, wherein the memory stores a computer program, and the computer program is loaded and executed by the processor to implement a method for auxiliary decision-making of gas leakage in a chemical park.

[0050] The processor may be a central processing unit (CPU), a general purpose processor, a digital signal processor (DSP), an ASIC, an FPGA or other programmable logic device, a transistor logic device, a hardware component or any combination thereof. It may implement or execute various exemplary logic blocks, modules and circuits described in conjunction with the disclosure of this application. It may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and the like. The memory may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory, a mobile hard disk, a magnetic disk or an optical disk.

[0051] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes. The schemes in the embodiments of the present application may be implemented in various computer languages, for example, object-oriented programming language Java and literal scripting language JavaScript, etc.

[0052] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0053] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0054] The above technical features constitute the best embodiment of the present invention, which has strong adaptability and best implementation effect. Non-essential technical features can be added or reduced according to actual needs to meet the requirements of different situations.

Claims

1. A chemical park gas leakage auxiliary decision-making method, characterized in that: include: Obtain the data to be detected in the chemical park, determine whether the data to be detected is video data, and if so, uniformly sample the data to be detected to obtain multiple images to be detected; Input the image to be detected into the gas leakage detection model to obtain the recognition result of the gas leakage location and degree, wherein the gas leakage detection model is obtained by deep learning the improved YOLOv5s network model using a number of samples, and each of the several samples includes a gas leakage image and a label identifying the gas leakage location and degree; The gas leakage location and degree of the gas leakage in the image to be detected are matched with the preset emergency response decision set to obtain the corresponding emergency response decision.

2. The chemical park gas leakage auxiliary decision-making method according to claim 1 is characterized in that: The deep learning process of the gas leak detection model includes: Obtain several image samples of different gas leakage degrees, label each image sample, and divide them into training set and test set according to the proportion after labeling, wherein the labeling information includes the gas leakage location and gas leakage degree; The improved YOLOv5s network model is deeply learned using the training set to obtain a gas leakage detection model, wherein the improved YOLOv5s network model is based on the YOLOv5s network model with the CA attention mechanism added; The gas leak detection model is tested and evaluated using the test set, and the optimal gas leak detection model is output.

3. The chemical park gas leakage auxiliary decision-making method according to claim 2 is characterized in that: The obtaining of a number of image samples of different gas leakage degrees includes: Set multiple camera positions at multiple shooting distances and obtain video data of a fixed duration at each camera position; Taking the shooting time of the video data as the reference variable of the gas leakage degree, N pictures with different leakage degrees are sampled from the video data of each shooting distance and each shooting position to form a picture sample library; Preprocess all image samples in the image sample library.

4. The chemical park gas leakage auxiliary decision-making method according to claim 3 is characterized in that: The preprocessing of all sampled image samples includes: Perform affine transformation, filtering, translation, and rotation on all image samples to expand the number of image samples; The expanded image samples are subjected to different degrees of sharpening and noise processing.

5. The chemical park gas leakage auxiliary decision-making method according to any one of claims 1 to 4, characterized in that: Use labelimg to label each image sample to identify the gas leakage location and degree.

6. A chemical park gas leakage auxiliary decision-making device using the method as described in any one of claims 1 to 5, characterized in that: include: The detection data acquisition unit acquires the data to be detected in the chemical park, determines whether the data to be detected is video data, and if so, uniformly samples the data to be detected to obtain multiple images to be detected; The detection unit inputs the image to be detected into the gas leakage detection model to obtain a gas leakage degree recognition result, wherein the gas leakage detection model is obtained by deep learning an improved YOLOv5s network model using a plurality of samples, and each of the plurality of samples includes a gas leakage image and a label identifying a gas leakage location and degree; The decision matching unit matches the gas leakage position and gas leakage degree of the image to be detected with a preset emergency response decision set to obtain a corresponding emergency response decision.

7. The chemical park gas leakage auxiliary decision-making device according to claim 6, characterized in that: Also included are model building units, including: The sample acquisition module acquires a number of image samples with different gas leakage degrees, labels each image sample, and divides it into a training set and a test set in proportion after labeling, wherein the labeling information includes the gas leakage location and the gas leakage degree; The deep learning module uses the training set to perform deep learning on the improved YOLOv5s network model to obtain a gas leakage detection model, wherein the improved YOLOv5s network model is based on the YOLOv5s network model and adds the CA attention mechanism; The model testing module uses the test set to test and evaluate the gas leak detection model and outputs the optimal gas leak detection model.

8. A chemical park gas leakage auxiliary decision-making system, characterized in that: include: A processing device, in which the chemical park gas leakage auxiliary decision-making device as described in claims 6 to 7 is stored and executed, and a corresponding visualization interface is provided; The interactive device is used for the operator to communicate with the processing device, so that the operator provides the processing device with the data to be detected in the chemical park through the interactive unit.

9. A storage medium, characterized in that: The storage medium stores a computer program that can be read by a computer, and the computer program is configured to execute the chemical park gas leakage auxiliary decision-making method as described in any one of claims 1 to 5 when running.

10. An electronic device, characterized in that: It comprises a processor and a memory, wherein a computer program is stored in the memory, and the computer program is loaded and executed by the processor to implement the chemical park gas leakage auxiliary decision-making method as described in any one of claims 1 to 5.

Citation Information

Patent Citations

  • Wheat ear detection method based on improved YOLOv5

    CN114973002A

  • Apple leaf disease detection method based on improved Yolov5s model

    CN115273072A

  • Deep learning-based airtight container airtightness water detection method

    CN115760771A

  • Variable data printing defect detection method and system based on YOLOv5s model

    CN117274176A

  • Rice planthopper target detection method, system and equipment based on improved Yolov5 and medium

    CN117351471A