Gas station fire control method and system, readable storage medium and computer

Abstract

The invention provides a gas station fire control method and system, a readable storage medium and a computer. The method comprises the following steps: inputting each first fire scene image into a convolutional neural network model for model optimization to obtain a convolutional neural network optimization model; performing model optimization on the image detection model based on each second fire scene image to obtain an image detection optimization model; performing data processing on the feature extraction results of the first fire scene images and the second fire scene images by using an image registration algorithm to fuse data; constructing a corresponding fire-fighting strategy set according to the fire-fighting facility information and the fire-fighting task set; and performing model fusion on the fire-fighting strategy set, the fusion data, the convolutional neural network optimization model and the image detection optimization model to construct a fire-fighting detection model, and realizing fire-fighting control of the target gas station through the fire-fighting detection model. The fire-fighting control of the target gas station is realized through the fire-fighting detection model, the task processing efficiency is improved, and the distribution cost is reduced.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a gas station fire control method, system, readable storage medium and computer. Background Art

[0002] With the rapid development of science and technology and the improvement of people's living standards, vehicles have become one of the indispensable tools in people's lives. With the increase in the number of vehicles, the number of gas stations is also gradually increasing.

[0003] As a new research and development product, firefighting robots can be better applied in areas with special characteristics such as gas stations. On the one hand, they can replace humans to complete firefighting work to reduce the harm of disasters to rescue personnel. On the other hand, they can quickly detect fires in gas stations and improve rescue efficiency. However, the current control of firefighting robots is to manually collect relevant data, use the data to process it in the terminal, and control the robot to perform firefighting work at the fire scene according to the processing results in the terminal. However, this method has large errors, and the task allocation of the firefighting robot can only be executed according to the settings of the staff, which greatly reduces the work efficiency and increases the cost of use. Summary of the invention

[0004] Based on this, the purpose of the present invention is to provide a gas station fire control method, system, readable storage medium and computer to at least solve the deficiencies in the above-mentioned technology.

[0005] The present invention provides a gas station fire control method, comprising: Acquire a plurality of first fire scene images and a second fire scene image, and construct a convolutional neural network model, and input each of the first fire scene images into the convolutional neural network model so that the convolutional neural network model is optimized to obtain a convolutional neural network optimization model; Constructing an image detection model, and optimizing the image detection model based on each of the second fire scene images to obtain an image detection optimization model; Extracting features from each of the first fire scene images and each of the second fire scene images, and performing data processing on the feature extraction results using an image registration algorithm to obtain corresponding fused data; Obtaining firefighting facility information of the areas where several gas stations are located, and constructing a firefighting task set, and constructing a corresponding firefighting strategy set according to the firefighting facility information and the firefighting task set; The fire strategy set, the fusion data, the convolutional neural network optimization model and the image detection optimization model are fused to construct a fire detection model, and fire control of the target gas station is achieved through the fire detection model.

[0006] Furthermore, the step of inputting each of the first fire scene images into the convolutional neural network model so as to optimize the convolutional neural network model to obtain a convolutional neural network optimization model includes: Inputting each of the first fire scene images into the convolutional neural network model, so that the convolutional neural network model performs feature extraction on each of the first fire scene images to obtain a corresponding one-dimensional feature map and a two-dimensional feature map; A hybrid attention algorithm is added to the backbone network module of the convolutional neural network model, and the one-dimensional feature map and the two-dimensional feature map are deeply convolved. The deep convolution result is used to optimize the processed convolutional neural network model to obtain a convolutional neural network optimization model.

[0007] Furthermore, the step of optimizing the image detection model based on each of the second fire scene images to obtain an image detection optimization model includes: Performing image processing on each of the second fire scene images to construct a data matrix of each of the second fire scene images; Based on the data matrix, weight calculation and image segmentation are performed on each of the second fire scene images to obtain a plurality of segmented images, and clustering processing is performed on each of the segmented images to obtain a corresponding clustering result; The clustering results are input into the image detection model for model optimization to obtain an image detection optimization model.

[0008] Furthermore, the steps of extracting features from each of the first fire scene images and each of the second fire scene images, and performing data processing on the feature extraction results using an image registration algorithm to obtain corresponding fused data include: Performing data cleaning and data processing on each of the first fire scene images and each of the second fire scene images to obtain a corresponding first processed image and a second processed image; The first processed image and the second processed image are corrected and registered by using an image registration algorithm, and the registered first processed image and the second processed image are feature fused to obtain corresponding fused data.

[0009] Furthermore, the firefighting facility information of the areas where several gas stations are located is obtained, and a firefighting task set is constructed. The steps of constructing a corresponding firefighting strategy set according to the firefighting facility information and the firefighting task set include: Obtaining firefighting facility information in the areas where several gas stations are located, wherein the firefighting facility information includes the number information of firefighting robots and corresponding type information; A firefighting robot set is constructed according to the quantity information of the firefighting robots and the corresponding type information, and a task sequence is constructed according to the firefighting robot set. The constructed task sequence is strategically combined with the firefighting robot set to construct a corresponding firefighting strategy set.

[0010] The present invention also provides a gas station fire control system, comprising: A first model building module is used to obtain a plurality of first fire scene images and a second fire scene image, and to build a convolutional neural network model, and to input each of the first fire scene images into the convolutional neural network model so as to optimize the convolutional neural network model to obtain a convolutional neural network optimization model; A second model building module is used to build an image detection model, and optimize the image detection model based on each of the second fire scene images to obtain an image detection optimization model; A data fusion module, used for extracting features from each of the first fire scene images and each of the second fire scene images, and performing data processing on the feature extraction results using an image registration algorithm to obtain corresponding fused data; A fire strategy building module is used to obtain fire facility information of the areas where several gas stations are located, and to build a fire task set, and to build a corresponding fire strategy set according to the fire facility information and the fire task set; The fire control module is used to fuse the fire strategy set, the fusion data, the convolutional neural network optimization model and the image detection optimization model to construct a fire detection model, and realize fire control of the target gas station through the fire detection model.

[0011] Furthermore, the first model building module includes: A feature extraction unit, used for inputting each of the first fire scene images into the convolutional neural network model, so that the convolutional neural network model performs feature extraction on each of the first fire scene images to obtain a corresponding one-dimensional feature map and a two-dimensional feature map; A model optimization unit is used to add a hybrid attention algorithm to the backbone network module of the convolutional neural network model, and perform deep convolution on the one-dimensional feature map and the two-dimensional feature map, and optimize the processed convolutional neural network model using the deep convolution result to obtain a convolutional neural network optimization model.

[0012] Furthermore, the second model building module includes: An image processing unit, used for performing image processing on each of the second fire scene images to construct a data matrix of each of the second fire scene images; An image segmentation unit, used for performing weight calculation and image segmentation on each of the second fire scene images based on the data matrix to obtain a plurality of segmented images, and performing clustering processing on each of the segmented images to obtain a corresponding clustering result; A clustering processing unit is used to input the clustering result into the image detection model for model optimization to obtain an image detection optimization model.

[0013] Furthermore, the data fusion module includes: Performing data cleaning and data processing on each of the first fire scene images and each of the second fire scene images to obtain a corresponding first processed image and a second processed image; The first processed image and the second processed image are corrected and registered by using an image registration algorithm, and the registered first processed image and the second processed image are feature fused to obtain corresponding fused data.

[0014] Furthermore, the fire fighting strategy building module includes: Obtaining firefighting facility information in the areas where several gas stations are located, wherein the firefighting facility information includes the number information of firefighting robots and corresponding type information; A firefighting robot set is constructed according to the quantity information of the firefighting robots and the corresponding type information, and a task sequence is constructed according to the firefighting robot set. The constructed task sequence is strategically combined with the firefighting robot set to construct a corresponding firefighting strategy set.

[0015] The present invention also provides a readable storage medium on which a computer program is stored. When the program is executed by a processor, the above-mentioned gas station fire control method is implemented.

[0016] The present invention also proposes a computer, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned gas station fire control method when executing the computer program.

[0017] The gas station fire control method, system, readable storage medium and computer of the present invention obtain a first fire scene image and a second fire scene image, extract features from the two different fire scene images, and use the feature extraction results to optimize the constructed convolutional neural network model and image detection model to improve the accuracy of feature extraction, so that the model can more effectively capture important feature information to improve the perception ability of the model; by processing the fire facility information of the area where the gas station is located, a corresponding fire strategy set is constructed according to the fire facility information and the constructed fire task set, and the fire strategy set, fusion data, convolutional neural network optimization model and image detection optimization model are model fused to construct a fire detection model, and the fire detection model is used to achieve fire control of the target gas station, thereby improving the efficiency of task processing and reducing allocation costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 is a flow chart of a gas station fire control method in a first embodiment of the present invention; Figure 2 for Figure 1 Detailed flow chart of step S101; Figure 3 for Figure 1 Detailed flow chart of step S102; Figure 4 for Figure 1 Detailed flow chart of step S103; Figure 5 for Figure 1 Detailed flow chart of step S104; Figure 6 is a structural block diagram of a gas station fire control system in a second embodiment of the present invention; Figure 7 FIG. 4 is a structural block diagram of a computer in a third embodiment of the present invention.

[0019] The following specific implementation manner will further illustrate the present invention in conjunction with the above-mentioned drawings. DETAILED DESCRIPTION

[0020] In order to facilitate the understanding of the present invention, the present invention will be described more fully below with reference to the relevant drawings. Several embodiments of the present invention are given in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive.

[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which the present invention belongs. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0022] Embodiment 1 See also Figure 1 , which shows a gas station fire control method in a first embodiment of the present invention, and the method specifically includes steps S101 to S105: S101, obtaining a plurality of first fire scene images and a second fire scene image, and constructing a convolutional neural network model, inputting each of the first fire scene images into the convolutional neural network model, so that the convolutional neural network model is optimized to obtain a convolutional neural network optimization model; For further information, see Figure 2 , the step S101 specifically includes steps S1011~S1012: S1011, inputting each of the first fire scene images into the convolutional neural network model, so that the convolutional neural network model performs feature extraction on each of the first fire scene images to obtain a corresponding one-dimensional feature map and a two-dimensional feature map; S1012, adding a hybrid attention algorithm to the backbone network module of the convolutional neural network model, and performing deep convolution on the one-dimensional feature map and the two-dimensional feature map, and optimizing the processed convolutional neural network model using the deep convolution result to obtain a convolutional neural network optimization model.

[0023] In the specific implementation, the fire scene image is collected by a preset image acquisition device, wherein the image acquisition device includes but is not limited to a camera, an infrared image sensor and other devices with image acquisition function. In this embodiment, the visible light image of the fire scene is collected by a camera, and the near infrared image of the same fire scene is collected by an infrared camera; Further, a convolutional neural network model is constructed. In order to avoid the uneven distribution of sample number data in the visible light image, each first fire scene image is input into the convolutional neural network model, and the channel attention algorithm and the spatial attention algorithm in the convolutional neural network model are used to perform image processing on each first fire scene image, so as to obtain a one-dimensional feature map and a two-dimensional feature map corresponding to each first fire scene image. It can be understood that by adaptively learning the channel correlation and spatial importance of the input feature map, the accuracy of feature extraction is improved, so that the model can more effectively capture important feature information, so as to improve the perception ability of the model; Specifically, a hybrid attention algorithm is added to the backbone network module of the convolutional neural network model (in this application, the algorithm is obtained by using the deep separable convolution optimization CBAM algorithm), and the one-dimensional feature map and the two-dimensional feature map obtained above are used for deep convolution. The deep convolution result is used to optimize the processed convolutional neural network model to obtain the convolutional neural network optimization model. It can be understood that by adding the hybrid attention algorithm, the structure of the backbone network is richer, and the network can maintain better stability to changes and disturbances in the input image, thereby increasing the generalization ability of the model.

[0024] S102, constructing an image detection model, and optimizing the image detection model based on each of the second fire scene images to obtain an image detection optimization model; For further information, see Figure 3 , the step S102 specifically includes steps S1021 to S1023: S1021, performing image processing on each of the second fire scene images to construct a data matrix of each of the second fire scene images; S1022, performing weight calculation and image segmentation on each of the second fire scene images based on the data matrix to obtain a plurality of segmented images, and performing clustering processing on each of the segmented images to obtain a corresponding clustering result; S1023, inputting the clustering result into the image detection model for model optimization to obtain an image detection optimization model.

[0025] In the specific implementation, an image detection model is pre-constructed, and image processing is performed on each second fire scene image. The similarity matrix of the image is constructed using Euclidean distance, and the similarity matrix is ​​normalized to obtain a corresponding normalized Laplace matrix. The corresponding adjacency matrix is ​​constructed using K maximum eigenvalue vectors in the normalized Laplace matrix. Specifically, weights of each second fire scene image are calculated according to the adjacency matrix obtained above, and the image is segmented according to the minimum uniform weight to cut it into different sub-graphs. Each sub-graph is clustered to obtain a corresponding clustering result, and the clustering result is input into the above-constructed image detection model for model optimization. The clustering result is combined with the image detection model using a sparse subspace clustering algorithm, high-dimensional data is represented in a low-dimensional manner, and data points are distributed to different subspaces through clustering, thereby improving the processing efficiency of the image detection model.

[0026] S103, extracting features from each of the first fire scene images and each of the second fire scene images, and performing data processing on the feature extraction results using an image registration algorithm to obtain corresponding fused data; For further information, see Figure 4 , the step S103 specifically includes steps S1031~S1032: S1031, performing data cleaning and data processing on each of the first fire scene images and each of the second fire scene images to obtain a corresponding first processed image and a second processed image; S1032: Perform image correction and registration on the first processed image and the second processed image using an image registration algorithm, and perform feature fusion on the registered first processed image and the second processed image to obtain corresponding fused data.

[0027] In the specific implementation, each first fire scene image and each second fire scene image is subjected to data cleaning and data enhancement, the purpose of which is to eliminate redundancy in the data, fill in gaps, improve the quality and reliability of the data, provide better input for subsequent image fusion, and improve the fusion effect and visual perception quality.

[0028] Specifically, an image registration algorithm is used to perform image registration on the first processed image and the second processed image to eliminate distortion and deformation in the image, a reference coordinate system is pre-constructed, and the first processed image and the second processed image are mapped to the same coordinate system in different directions or angles using the reference coordinate system (in this embodiment, the image registration algorithm includes but is not limited to a feature matching algorithm and a least squares fitting algorithm), and feature areas of the image after image registration are extracted, and feature fusion is performed on the feature areas. Correlation and complementary features are extracted from the feature space information of the two images, and feature-level fusion is performed using the extracted features, thereby making up for the shortcomings of a single feature, reducing feature redundancy and noise, and improving the robustness and generalization ability of the model.

[0029] Specifically, a multi-channel fusion algorithm is constructed, and the feature areas obtained above are used to restore data samples through a step-by-step denoising process in the multi-channel fusion algorithm. The multi-channel fusion algorithm is used to generate a diffusion model to realize data fusion processing, and the feature-level fusion results obtained above are integrated with the multi-channel fusion results of this time to obtain the corresponding fused data.

[0030] S104, obtaining firefighting facility information of the areas where several gas stations are located, and constructing a firefighting task set, and constructing a corresponding firefighting strategy set according to the firefighting facility information and the firefighting task set; For further information, see Figure 5 , the step S104 specifically includes steps S1041-S1042: S1041, obtaining firefighting facility information in the areas where several gas stations are located, wherein the firefighting facility information includes the number information of firefighting robots and corresponding type information; S1042, construct a firefighting robot set according to the number information of the firefighting robots and the corresponding type information, and construct a task sequence according to the firefighting robot set, and strategically combine the constructed task sequence with the firefighting robot set to construct a corresponding firefighting strategy set.

[0031] In the specific implementation, the firefighting facility information of the area where several gas stations are located is obtained, wherein the firefighting facility information includes the number information corresponding to the firefighting robots pre-placed in the area, the type information of the firefighting robots, and the driving speed of the firefighting robots. The number information, type information, and driving speed are used to construct a firefighting robot set. , among which, fire fighting robot Depend on , , Three sets of data show that, represents the number of firefighting robots, Indicates the type of fire fighting robot. Indicates the driving speed of the fire-fighting robot, constructs a corresponding task sequence according to the fire-fighting robot set, generates the corresponding rescue task quantity, processing level, execution time and other related information for the fire-fighting robot set, and combines various types of rescue task information to obtain the corresponding task sequence; Specifically, define the time set corresponding to the firefighting robot set , which represents the time when the firefighting robot starts to perform tasks according to its preset path set, where ; Define the time window set corresponding to the fire fighting robot set , , the time window set represents the fire fighting robot A task can only be executed during the time period in which the task exists, where Indicates the task The firefighting robot can only perform the task during this period of time. , Indicates the task The end time of Indicates the task The start time of The bidding algorithm is used to process the constructed task sequence and the firefighting robot set to obtain the corresponding bidding set. ,in, Indicates the owner of the bidding set, the fire robot The task number obtained by the current bidding, Indicates the number of elements in the bidding set, that is, the number of tasks, for example: Firefighting robot The first task currently obtained through bidding is Task 2.

[0032] The above data are integrated to calculate the fire fighting robot In the task Complete the task after adding it to the bidding set Total score obtained: ; ; In the formula, Fire fighting robot By bidding collection The total score of completing the tasks in order, , Fire fighting robots Complete the task The scores and costs after the fire fighting robot are calculated through the pre-built database. Indicates the task Add to Bid Collection Rear fire fighting robot The total score after completing the task.

[0033] After obtaining the total scores of all firefighting robots, they are sorted according to the score values ​​to construct the corresponding firefighting strategy set.

[0034] S105, performing model fusion on the fire strategy set, the fusion data, the convolutional neural network optimization model and the image detection optimization model to construct a fire detection model, and realizing fire control of the target gas station through the fire detection model.

[0035] In the specific implementation, the fire protection strategy set, fusion data, convolutional neural network optimization model and image detection optimization model obtained above are fused, and the two models are combined using the fusion data to realize image processing of different fire scenes, thereby improving processing efficiency while improving the accuracy and robustness of the model. The fire protection detection model obtained by the combination is used to realize fire control of the target gas station through the fire protection detection model.

[0036] In summary, the gas station fire control method in the above-mentioned embodiment of the present invention obtains the first fire scene image and the second fire scene image, extracts features from the two different fire scene images, and uses the feature extraction results to optimize the constructed convolutional neural network model and image detection model to improve the accuracy of feature extraction, so that the model can more effectively capture important feature information to improve the perception ability of the model; by processing the firefighting facility information in the area where the gas station is located, a corresponding firefighting strategy set is constructed according to the firefighting facility information and the constructed firefighting task set, and the firefighting strategy set, fusion data, convolutional neural network optimization model and image detection optimization model are fused to construct a fire detection model, and the fire detection model is used to realize fire control of the target gas station, thereby improving the efficiency of task processing and reducing allocation costs.

[0037] Embodiment 2 Another aspect of the present invention is to provide a gas station fire control system, please refer to Figure 6 , shown is a gas station fire control system in a second embodiment of the present invention, the system comprises: The first model building module 11 is used to obtain a plurality of first fire scene images and a second fire scene image, and to build a convolutional neural network model, and to input each of the first fire scene images into the convolutional neural network model so as to optimize the convolutional neural network model to obtain a convolutional neural network optimization model; Furthermore, the first model building module 11 includes: A feature extraction unit, used for inputting each of the first fire scene images into the convolutional neural network model, so that the convolutional neural network model performs feature extraction on each of the first fire scene images to obtain a corresponding one-dimensional feature map and a two-dimensional feature map; A model optimization unit is used to add a hybrid attention algorithm to the backbone network module of the convolutional neural network model, and perform deep convolution on the one-dimensional feature map and the two-dimensional feature map, and optimize the processed convolutional neural network model using the deep convolution result to obtain a convolutional neural network optimization model.

[0038] A second model building module 12 is used to build an image detection model, and optimize the image detection model based on each of the second fire scene images to obtain an image detection optimization model; Furthermore, the second model building module 12 includes: An image processing unit, used for performing image processing on each of the second fire scene images to construct a data matrix of each of the second fire scene images; An image segmentation unit, used for performing weight calculation and image segmentation on each of the second fire scene images based on the data matrix to obtain a plurality of segmented images, and performing clustering processing on each of the segmented images to obtain a corresponding clustering result; A clustering processing unit is used to input the clustering result into the image detection model for model optimization to obtain an image detection optimization model.

[0039] A data fusion module 13 is used to extract features from each of the first fire scene images and each of the second fire scene images, and perform data processing on the feature extraction results using an image registration algorithm to obtain corresponding fused data; Furthermore, the data fusion module 13 includes: Performing data cleaning and data processing on each of the first fire scene images and each of the second fire scene images to obtain a corresponding first processed image and a second processed image; The first processed image and the second processed image are corrected and registered by using an image registration algorithm, and the registered first processed image and the second processed image are feature fused to obtain corresponding fused data.

[0040] A fire strategy building module 14 is used to obtain fire facility information of the areas where several gas stations are located, and to build a fire task set, and to build a corresponding fire strategy set according to the fire facility information and the fire task set; Furthermore, the fire fighting strategy building module 14 includes: Obtaining firefighting facility information in the areas where several gas stations are located, wherein the firefighting facility information includes the number information of firefighting robots and corresponding type information; A firefighting robot set is constructed according to the quantity information of the firefighting robots and the corresponding type information, and a task sequence is constructed according to the firefighting robot set. The constructed task sequence is strategically combined with the firefighting robot set to construct a corresponding firefighting strategy set.

[0041] The fire control module 15 is used to perform model fusion on the fire strategy set, the fusion data, the convolutional neural network optimization model and the image detection optimization model to construct a fire detection model, and realize fire control of the target gas station through the fire detection model.

[0042] The functions or operation steps implemented when the above modules and units are executed are generally the same as those in the above method embodiments, and will not be repeated here.

[0043] The gas station fire control system provided in the embodiment of the present invention has the same implementation principle and technical effects as those of the aforementioned method embodiment. For the sake of brief description, for matters not mentioned in the system embodiment, reference may be made to the corresponding contents in the aforementioned method embodiment.

[0044] Embodiment 3 The present invention also provides a computer, see Figure 7 , shown is a computer in the third embodiment of the present invention, including a memory 10, a processor 20, and a computer program 30 stored in the memory 10 and executable on the processor 20. When the processor 20 executes the computer program 30, the above-mentioned gas station fire control method is implemented.

[0045] The memory 10 includes at least one type of readable storage medium, which includes a flash memory, a hard disk, a multimedia card, a card-type memory (e.g., an SD or DX memory, etc.), a magnetic memory, a magnetic disk, an optical disk, etc. In some embodiments, the memory 10 may be an internal storage unit of a computer, such as a hard disk of the computer. In other embodiments, the memory 10 may also be an external storage device, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card, etc. Further, the memory 10 may also include both an internal storage unit of the computer and an external storage device. The memory 10 may be used not only to store application software and various types of data installed in the computer, but also to temporarily store data that has been output or is to be output.

[0046] Among them, in some embodiments, the processor 20 can be an electronic control unit (Electronic Control Unit, abbreviated as ECU, also known as a vehicle computer), a central processing unit (Central Processing Unit, CPU), a controller, a microcontroller, a microprocessor or other data processing chip, used to run the program code stored in the memory 10 or process data, such as executing access restriction programs, etc.

[0047] It should be pointed out that Figure 7 The structure shown does not constitute a limitation on the computer. In other embodiments, the computer may include fewer or more components than shown in the figure, or combine certain components, or arrange the components differently.

[0048] The embodiment of the present invention further provides a readable storage medium on which a computer program is stored. When the program is executed by a processor, the fire control method for a gas station as described above is implemented.

[0049] Those skilled in the art will appreciate that the logic and / or steps represented in the flowchart or otherwise described herein, for example, may be considered as an ordered list of executable instructions for implementing logical functions, and may be specifically implemented in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For purposes of this specification, "computer-readable medium" may be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0050] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0051] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or a combination thereof: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0052] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0053] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A gas station fire control method, characterized in that: include: Acquire a plurality of first fire scene images and a second fire scene image, and construct a convolutional neural network model, and input each of the first fire scene images into the convolutional neural network model so that the convolutional neural network model is optimized to obtain a convolutional neural network optimization model; Constructing an image detection model, and optimizing the image detection model based on each of the second fire scene images to obtain an image detection optimization model; Extracting features from each of the first fire scene images and each of the second fire scene images, and performing data processing on the feature extraction results using an image registration algorithm to obtain corresponding fused data; Obtaining firefighting facility information of the areas where several gas stations are located, and constructing a firefighting task set, and constructing a corresponding firefighting strategy set according to the firefighting facility information and the firefighting task set; The fire strategy set, the fusion data, the convolutional neural network optimization model and the image detection optimization model are fused to construct a fire detection model, and fire control of the target gas station is achieved through the fire detection model.

2. The fire control method for a gas station according to claim 1, characterized in that: The step of inputting each of the first fire scene images into the convolutional neural network model so as to optimize the convolutional neural network model to obtain a convolutional neural network optimization model comprises: Inputting each of the first fire scene images into the convolutional neural network model, so that the convolutional neural network model performs feature extraction on each of the first fire scene images to obtain a corresponding one-dimensional feature map and a two-dimensional feature map; A hybrid attention algorithm is added to the backbone network module of the convolutional neural network model, and the one-dimensional feature map and the two-dimensional feature map are deeply convolved. The deep convolution result is used to optimize the processed convolutional neural network model to obtain a convolutional neural network optimization model.

3. The fire control method for a gas station according to claim 1, characterized in that: The step of optimizing the image detection model based on each of the second fire scene images to obtain an image detection optimization model includes: Performing image processing on each of the second fire scene images to construct a data matrix of each of the second fire scene images; Based on the data matrix, weight calculation and image segmentation are performed on each of the second fire scene images to obtain a plurality of segmented images, and clustering processing is performed on each of the segmented images to obtain a corresponding clustering result; The clustering results are input into the image detection model for model optimization to obtain an image detection optimization model.

4. The fire control method for a gas station according to claim 1, characterized in that: The steps of extracting features from each of the first fire scene images and each of the second fire scene images, and performing data processing on the feature extraction results using an image registration algorithm to obtain corresponding fused data include: Performing data cleaning and data processing on each of the first fire scene images and each of the second fire scene images to obtain a corresponding first processed image and a second processed image; The first processed image and the second processed image are corrected and registered by using an image registration algorithm, and the registered first processed image and the second processed image are feature fused to obtain corresponding fused data.

5. The fire control method for a gas station according to claim 1, characterized in that: The steps of obtaining firefighting facility information of the areas where several gas stations are located and constructing a firefighting task set, and constructing a corresponding firefighting strategy set according to the firefighting facility information and the firefighting task set include: Obtaining firefighting facility information in the areas where several gas stations are located, wherein the firefighting facility information includes the number information of firefighting robots and corresponding type information; A firefighting robot set is constructed according to the quantity information of the firefighting robots and the corresponding type information, and a task sequence is constructed according to the firefighting robot set. The constructed task sequence is strategically combined with the firefighting robot set to construct a corresponding firefighting strategy set.

6. A gas station fire control system, characterized in that: include: A first model building module is used to obtain a plurality of first fire scene images and a second fire scene image, and to build a convolutional neural network model, and to input each of the first fire scene images into the convolutional neural network model so as to optimize the convolutional neural network model to obtain a convolutional neural network optimization model; A second model building module is used to build an image detection model, and optimize the image detection model based on each of the second fire scene images to obtain an image detection optimization model; A data fusion module, used for extracting features from each of the first fire scene images and each of the second fire scene images, and performing data processing on the feature extraction results using an image registration algorithm to obtain corresponding fused data; A fire strategy building module is used to obtain fire facility information of the areas where several gas stations are located, and to build a fire task set, and to build a corresponding fire strategy set according to the fire facility information and the fire task set; The fire control module is used to fuse the fire strategy set, the fusion data, the convolutional neural network optimization model and the image detection optimization model to construct a fire detection model, and realize fire control of the target gas station through the fire detection model.

7. The gas station fire control system according to claim 6, characterized in that: The first model building module includes: A feature extraction unit, used for inputting each of the first fire scene images into the convolutional neural network model, so that the convolutional neural network model performs feature extraction on each of the first fire scene images to obtain a corresponding one-dimensional feature map and a two-dimensional feature map; A model optimization unit is used to add a hybrid attention algorithm to the backbone network module of the convolutional neural network model, and perform deep convolution on the one-dimensional feature map and the two-dimensional feature map, and optimize the processed convolutional neural network model using the deep convolution result to obtain a convolutional neural network optimization model.

8. The gas station fire control system according to claim 6, characterized in that: The second model building module includes: An image processing unit, used for performing image processing on each of the second fire scene images to construct a data matrix of each of the second fire scene images; An image segmentation unit, used for performing weight calculation and image segmentation on each of the second fire scene images based on the data matrix to obtain a plurality of segmented images, and performing clustering processing on each of the segmented images to obtain a corresponding clustering result; A clustering processing unit is used to input the clustering result into the image detection model for model optimization to obtain an image detection optimization model.

9. A readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the gas station fire control method as described in any one of claims 1 to 5 is implemented.

10. A computer comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the gas station fire control method as described in any one of claims 1 to 5 is implemented.

Images