Fire behavior identification method and device for fire fighting truck, electronic equipment and storage medium

Abstract

The invention provides a fire behavior identification method and device for a fire fighting truck, electronic equipment and a storage medium, and the method comprises the steps: collecting an original fire image of a fire scene based on an image collection device on the fire fighting truck, and carrying out the first feature extraction of the original fire image after preprocessing operation through convolution operation, the method comprises the steps of obtaining fire image features of multiple channels, compressing the fire image features of the multiple channels to obtain global fire behavior information, and performing secondary feature extraction on the global fire behavior information through a dynamic convolution kernel and a channel attention mechanism to obtain fire behavior recognition features; flame edges and smoke edges in the fire behavior recognition features are calculated adaptively based on a large kernel, feature optimization is carried out on the fire behavior recognition features, the flame height and the flame shape are obtained according to the fire behavior recognition features after feature optimization and a preset fire behavior recognition model, and the fire behavior level of the fire scene is determined based on the flame height and the flame shape. Therefore, the technical problem that fire scene features are difficult to identify accurately is solved.

Description

Technical Field

[0001] The present application relates to the field of image recognition technology, and in particular to a method, device, electronic equipment and storage medium for identifying fire on a fire truck. Background Art

[0002] In recent years, with the acceleration of urbanization and the increase in building density, the frequency of fires has gradually increased. Once a fire occurs, it not only causes huge losses to property, but also seriously threatens people's lives. Therefore, fire prevention and timely and effective fire rescue have become particularly important. Traditional fire identification and response methods rely on manual observation and simple sensor data, which not only have a slow response speed, but are also easily affected by environmental factors, making it difficult to make accurate judgments in time. In actual fire rescue scenarios, fire trucks need to quickly identify and analyze the fire in order to formulate scientific rescue plans. However, the fire scene is complex and changeable, and the detection and identification of the fire is often interfered by various factors such as smoke and light. Although the existing fire recognition technology has made certain progress, the recognition accuracy and stability in complex environments are still great challenges. On the one hand, under different lighting conditions and smoke interference, the fire characteristics may be difficult to be accurately captured by traditional methods; on the other hand, the real-time requirements of the fire scene are high, and traditional methods are difficult to meet the needs of rapid response. Summary of the invention

[0003] In view of the shortcomings of the related technologies mentioned above, the present application provides a fire identification method, device, electronic equipment and storage medium for a fire truck to solve the technical problem that it is difficult to accurately identify the characteristics of a fire scene.

[0004] The present application provides a fire identification method for a fire truck, which includes: collecting the original fire image of the fire scene based on the image acquisition device on the fire truck; performing image preprocessing operation on the original fire image, and extracting the features of the original fire image after the preprocessing operation through a convolution operation to obtain fire image features of multiple channels, and the fire image features include flame morphology features and smoke diffusion features; compressing the fire image features of multiple channels to obtain global fire information, and performing secondary feature extraction on the global fire information through a dynamic convolution kernel and a channel attention mechanism to obtain fire identification features; calculating the flame edge and smoke edge in the fire identification features based on a large core adaptive module to optimize the fire identification features. According to the fire identification features after the feature optimization and the preset fire identification model, the flame height and flame shape are obtained, and the fire level of the fire scene is determined based on the flame height and the flame shape, and the fire level is reported to the fire control center to generate an emergency decision corresponding to the fire level.

[0005] In one embodiment of the present application, performing image preprocessing operations on the original fire image includes: performing cropping operations and rotation operations on the original fire image to obtain original fire images at multiple angles and multiple scales; performing grayscale operations on the original fire images at multiple angles and multiple scales to obtain the original fire image after the preprocessing operation.

[0006] In one embodiment of the present application, extracting fire image features of multiple channels from the original fire image after the preprocessing operation through a convolution operation includes: sliding the convolution kernel on the original fire image according to a preset step size, and obtaining a pixel value of the fire image feature each time the convolution kernel is slid; and obtaining the fire image features of multiple channels through the sliding operation of multiple convolution kernels, wherein the fire image features include flame morphology features and smoke diffusion features.

[0007] In one embodiment of the present application, the fire image features of multiple channels are compressed to obtain global fire information, and the global fire information is subjected to secondary feature extraction through a dynamic convolution kernel and a channel attention mechanism to obtain fire identification features, including: compressing the fire image features of the multiple channels based on a preset image compression algorithm to obtain the global fire information; inputting the global fire information into a dynamic convolution kernel for feature extraction, wherein the convolution kernel size of the dynamic convolution kernel is determined based on the number of channels of the global fire information; calculating the attention weights of each channel of the global fire information after feature extraction in the dynamic convolution kernel, and recalibrating the features of each channel of the global fire information after feature extraction in the dynamic convolution kernel based on the attention weights to obtain the fire identification features.

[0008] In one embodiment of the present application, determining the fire intensity level of the fire scene based on the flame height and the flame shape includes: obtaining a flame height score of the fire scene according to the flame height; if the flame shape is a stable combustion type, determining the flame intensity score of the fire scene as a preset first intensity score; if the flame shape is a rapid spread type, determining the flame intensity score of the fire scene as a preset second intensity score; if the flame shape is an explosion type, determining the flame intensity score of the fire scene as a preset third intensity score; obtaining a fire intensity level score based on a preset flame height weight, a preset flame intensity weight, the flame height score and the flame intensity score, and determining the fire intensity level of the fire scene according to the fire intensity level score.

[0009] In one embodiment of the present application, determining the fire level of the fire scene according to the fire level score includes: if the fire level score is less than or equal to a preset first level score, the fire level of the fire scene is a minor fire. If the fire level score is greater than the preset first level score and less than the preset second level score, the fire level of the fire scene is a medium fire. If the fire level score is greater than or equal to the preset second level score, the fire level of the fire scene is a major fire.

[0010] In one embodiment of the present application, the flame edge and the smoke edge in the fire identification feature are calculated based on the large core adaptive module to optimize the fire identification feature, including: decomposing a large core convolution into a first small core convolution and a second small core convolution, and performing deep feature extraction on the flame edge and the smoke edge in the fire identification feature based on the first small core convolution and the second small core convolution respectively; splicing the features extracted by the first small core convolution and the second small core convolution to obtain the fire identification feature after feature optimization.

[0011] An embodiment of the present application also provides a fire truck flame recognition device, which includes: an image acquisition module, which is used to acquire the original fire image of the fire scene based on the image acquisition device on the fire truck; a first extraction module, which is used to perform image preprocessing operations on the original fire image, and perform a feature extraction on the original fire image after the preprocessing operation through a convolution operation to obtain fire image features of multiple channels, and the fire image features include flame morphology features and smoke diffusion features; a second extraction module, which is used to compress the fire image features of multiple channels to obtain global fire information, and perform secondary feature extraction on the global fire information through a dynamic convolution kernel and a channel attention mechanism to obtain a fire recognition feature; a feature optimization module, which is used to calculate the flame edge and smoke edge in the fire recognition feature based on a large core adaptive module to optimize the fire recognition feature. A fire identification module is used to obtain the flame height and flame shape according to the fire identification features after the feature optimization and a preset fire identification model, and determine the fire level of the fire scene based on the flame height and the flame shape, and report the fire level to the fire control center to generate an emergency decision corresponding to the fire level.

[0012] An embodiment of the present application also provides an electronic device, comprising: one or more processors; a storage device for storing one or more programs, and when the one or more programs are executed by the one or more processors, the electronic device implements the fire identification method for a fire truck as described in any of the above embodiments.

[0013] An embodiment of the present application further provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor of a computer, the computer is enabled to execute a fire identification method for a fire truck as described in any of the above embodiments.

[0014] Beneficial effects of the present application: The embodiments of the present application provide a method, device, electronic device and storage medium for fire identification on a fire truck. The method collects the original fire image of the fire scene based on the image acquisition device on the fire truck, performs image preprocessing operation on the original fire image, and performs a feature extraction on the original fire image after the preprocessing operation through a convolution operation to obtain fire image features of multiple channels, the fire image features include flame shape and smoke diffusion features, the fire image features of multiple channels are compressed to obtain global fire information, and the global fire information is subjected to secondary feature extraction through a dynamic convolution kernel and a channel attention mechanism to obtain fire identification features, and fire identification is adaptively calculated based on a large core. The flame edge and smoke edge in the feature are used to optimize the fire identification feature, and the flame height and flame shape are obtained according to the fire identification feature after feature optimization and the preset fire identification model. The fire level at the fire scene is determined based on the flame height and flame shape, and the fire level is reported to the fire control center to generate an emergency strategy corresponding to the fire level. Through this method, the fire image is subjected to multiple feature extraction based on the dynamic convolution kernel and channel attention mechanism and feature optimization based on the large kernel adaptation to obtain the fire identification feature, thereby improving the accuracy and real-time performance of fire identification. At the same time, the identified fire level is uploaded in time, so as to quickly generate the corresponding emergency strategy method, making the fire rescue mission more efficient and reliable.

[0015] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a schematic diagram of an implementation environment of a fire truck fire identification method shown in an exemplary embodiment of the present application;

[0017] Figure 2 is a flow chart of a method for identifying fire in a fire truck shown in an exemplary embodiment of the present application;

[0018] Figure 3 is a schematic diagram of an emergency decision making process shown in an exemplary embodiment of the present application;

[0019] Figure 4 is a block diagram of a fire identification device for a fire truck shown in an exemplary embodiment of the present application;

[0020] Figure 5It is a structural schematic diagram of an electronic device shown in an exemplary embodiment of the present application. DETAILED DESCRIPTION

[0021] The following describes the embodiments of the present application through specific examples, and those skilled in the art can easily understand other advantages and effects of the present application from the contents disclosed in this specification. The present application can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present application. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.

[0022] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present application, and thus the drawings only show components related to the present application rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed at will, and the component layout may also be more complicated.

[0023] It should be noted that in this application, "first", "second", etc. are only used to distinguish similar objects, and are not used to limit the order or precedence of similar objects. The variations of "including", "having", etc. described above indicate that the scope covered by the subject of the word is not exclusive except for the examples shown by the word.

[0024] It is understood that the various numbers, step numbers, etc. recorded in this application are distinguished for the convenience of description and are not used to limit the scope of this application. The size of the numbers in this application does not mean the order of execution. The execution order of each process should be determined by its function and internal logic.

[0025] In the following description, a large number of details are discussed to provide a more thorough explanation of the embodiments of the present application. However, it is obvious to those skilled in the art that the embodiments of the present application can be implemented without these specific details. In other embodiments, well-known structures and devices are shown in the form of block diagrams rather than in detail to avoid making the embodiments of the present application difficult to understand.

[0026] The embodiments of the present application respectively propose a fire truck fire identification method, a fire truck fire identification device, an electronic device, a computer-readable storage medium and a computer program product, which will be described in detail below.

[0027] See also Figure 1 , Figure 1 It is a schematic diagram of an implementation environment of a fire identification method for a fire truck shown in an exemplary embodiment of the present application.

[0028] like Figure 1 As shown, the implementation environment may include an image acquisition device 101 and an image recognition device 102, wherein the image acquisition device 101 is a camera on a fire truck, used to acquire original fire images of a fire scene, and the image recognition device 102 is mounted on the fire truck, which may be an onboard computer on the fire truck, and the image recognition device 102 stores a fire intensity recognition method for a fire truck, which includes extracting features from the original fire image, and obtaining a flame height and a flame shape based on the extracted fire intensity recognition features, and determining a fire intensity level at the fire scene based on the flame height and the flame shape, and reporting the fire intensity level to a fire control center to generate an emergency decision corresponding to the fire intensity level.

[0029] See also Figure 2 , Figure 2 FIG. 1 is a flowchart of a fire identification method for a fire truck according to an exemplary embodiment of the present application. The method can be applied to Figure 1 The implementation environment shown is as follows. The method may also be applicable to other exemplary implementation environments and be specifically executed by devices in other implementation environments. This embodiment does not limit the implementation environment to which the method is applicable.

[0030] like Figure 2 As shown, in an exemplary embodiment, the fire truck fire identification method includes at least steps S210 to S250, which are described in detail as follows:

[0031] Step S210: collecting original fire images of the fire scene based on the image collection device on the fire truck.

[0032] In one embodiment of the present application, the fire scene is photographed by a camera installed on a fire truck to obtain an original fire image.

[0033] Step S220, performing image preprocessing operation on the original fire image, and performing feature extraction on the original fire image after the preprocessing operation through a convolution operation to obtain fire image features of multiple channels, wherein the fire image features include flame shape and smoke diffusion features.

[0034] In one embodiment of the present application, performing image preprocessing operations on the original fire image includes: performing cropping operations and rotation operations on the original fire image to obtain original fire images at multiple angles and multiple scales; performing grayscale operations on the original fire images at multiple angles and multiple scales to obtain the original fire image after the preprocessing operation.

[0035] In step S230, the fire image features of multiple channels are compressed to obtain global fire intensity information, and the global fire intensity information is subjected to secondary feature extraction through a dynamic convolution kernel and a channel attention mechanism to obtain fire intensity recognition features.

[0036] In one embodiment of the present application, extracting fire image features of multiple channels from an original fire image after a preprocessing operation through a convolution operation includes: sliding a convolution kernel on the original fire image according to a preset step size, and obtaining a pixel value of the fire image feature each time the convolution kernel is slid; and obtaining fire image features of multiple channels through sliding operations of multiple convolution kernels, wherein the fire image features include flame morphology features and smoke diffusion features.

[0037] In one embodiment of the present application, fire image features of multiple channels are compressed to obtain global fire information, and the global fire information is subjected to secondary feature extraction through a dynamic convolution kernel and a channel attention mechanism to obtain fire identification features, including: compressing the fire image features of multiple channels based on a preset image compression algorithm to obtain global fire information; inputting the global fire information into a dynamic convolution kernel for feature extraction, wherein the convolution kernel size of the dynamic convolution kernel is determined based on the number of channels of the global fire information; calculating the attention weights of each channel of the global fire information after feature extraction in the dynamic convolution kernel, and recalibrating the features of each channel of the global fire information after feature extraction in the dynamic convolution kernel based on the attention weights to obtain fire identification features.

[0038] In one embodiment of the present application, performing secondary feature extraction on the global fire information through a dynamic convolution kernel and a channel attention mechanism includes: adaptively adjusting the filter size value of the channel attention mechanism based on the number of channels of the global feature information includes:

[0039]

[0040] In formula (1), k is the filter size value, C is the number of channels, and η,b are preset linear mapping parameters.

[0041] Step S240 , adaptively calculating the flame edge and the smoke edge in the fire intensity recognition feature based on the large core, so as to optimize the fire intensity recognition feature.

[0042] In one embodiment of the present application, the flame edge and the smoke edge in the fire identification feature are calculated based on the large core adaptive module to optimize the fire identification feature, including: decomposing a large core convolution into a first small core convolution and a second small core convolution, and performing deep feature extraction on the flame edge and the smoke edge in the fire identification feature based on the first small core convolution and the second small core convolution respectively; splicing the features extracted by the first small core convolution and the second small core convolution to obtain the fire identification feature after feature optimization.

[0043] In one embodiment of the present application, determining the number of decomposed kernels is a key to the large kernel adaptive module, and the increase in kernel size and dilation rate ensures that the receptive field expands quickly enough. In the present application, the kernel size k, the dilation rate d and the receptive field R of the i-th convolution in the large kernel adaptive module are defined as follows:

[0044] k i-1 ≤k i , d 1 =1,d i-1 <d i ≤R i-1 Formula (2)

[0045] R 1 =k 1 ,R i =d i (k i -1)+R i-1 Formula (3)

[0046] In equations (2) and (3), k is the kernel size, d is the dilation rate, and R is the receptive field.

[0047] According to the definitions of formula (2) and formula (3), in this embodiment, a convolution with a kernel of 23 is split into a convolution with a kernel of 5 and a convolution with a kernel of 7, so that the extracted features are more accurate. The splitting operation is implemented by convolutions with different receptive fields, which is expressed as follows:

[0048] y 0 =X,y i+1 =F i (y i ) Formula (4)

[0049] In formula (4), y 0 and X represents the original image, F is the image using kernel k i The depth of the convolution, y i Represents the image features after the i-th convolution, y i+1 Represents y i Image features after deep convolution.

[0050] Step S250, obtain the flame height and flame shape according to the feature-optimized fire identification features and the preset fire identification model, determine the fire level at the fire scene based on the flame height and flame shape, and report the fire level to the fire control center to generate an emergency strategy corresponding to the fire level.

[0051] In one embodiment of the present application, determining the fire intensity level of a fire scene based on flame height and flame shape includes: obtaining a flame height score for the fire scene based on the flame height; if the flame shape is a stable combustion type, determining the flame intensity score for the fire scene as a preset first intensity score; if the flame shape is a rapid spread type, determining the flame intensity score for the fire scene as a preset second intensity score; if the flame shape is an explosion type, determining the flame intensity score for the fire scene as a preset third intensity score; obtaining a fire intensity level score based on a preset flame height weight, a preset flame intensity weight, a flame height score and a flame intensity score, and determining the fire intensity level of the fire scene based on the fire intensity level score.

[0052] In one embodiment of the present application, determining the fire level of the fire scene according to the fire level score includes: if the fire level score is less than or equal to the preset first level score, the fire level of the fire scene is a minor fire. If the fire level score is greater than the preset first level score and less than the preset second level score, the fire level of the fire scene is a medium fire. If the fire level score is greater than or equal to the preset second level score, the fire level of the fire scene is a major fire.

[0053] See also Figure 3 , Figure 3 This is a schematic diagram of the emergency decision-making process shown in an exemplary embodiment of the present application. After the fire at the fire scene is identified, the fire level and fire scene-related images are packaged and uploaded to the fire control center in real time. The fire control center conducts a comprehensive analysis based on the fire level and fire scene-related images to generate an emergency strategy corresponding to the fire level.

[0054] See also Figure 4 , Figure 4 is a block diagram of a fire identification device for a fire truck shown in an exemplary embodiment of the present application. The device can be applied to Figure 1 The implementation environment shown is as follows. The device may also be applicable to other exemplary implementation environments and specifically configured in other devices. This embodiment does not limit the implementation environment to which the device is applicable.

[0055] like Figure 4 As shown, the exemplary fire intensity recognition device for a fire truck includes: an image acquisition module 401 , a first extraction module 402 , a second extraction module 403 , a feature optimization module 404 and a fire intensity recognition module 405 .

[0056] An image acquisition module 401 is used to acquire original fire images of the fire scene based on an image acquisition device on a fire truck;

[0057] The first extraction module 402 is used to perform a picture preprocessing operation on the original fire image, and perform a feature extraction on the original fire image after the preprocessing operation through a convolution operation to obtain fire image features of multiple channels, where the fire image features include flame morphology features and smoke diffusion features;

[0058] The second extraction module 403 is used to compress the fire image features of multiple channels to obtain global fire intensity information, and perform secondary feature extraction on the global fire intensity information through a dynamic convolution kernel and a channel attention mechanism to obtain fire intensity recognition features;

[0059] The feature optimization module 404 is used to calculate the flame edge and the smoke edge in the fire intensity recognition feature based on the large core adaptive module to optimize the fire intensity recognition feature.

[0060] The fire identification module 405 is used to obtain the flame height and flame shape according to the fire identification features after the feature optimization and the preset fire identification model, and determine the fire level at the fire scene based on the flame height and flame shape, and report the fire level to the fire control center to generate an emergency decision corresponding to the fire level.

[0061] Figure 5 The structure diagram of the computer system suitable for implementing the electronic device of the embodiment of the present application is shown. It should be noted that: Figure 5 The computer system 500 of the electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0062] like Figure 5 As shown, the computer system 500 includes a central processing unit (CPU) 501, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 502 or the program loaded from the storage part 508 to the random access memory (RAM) 503, such as executing the method described in the above embodiment. In the RAM 503, various programs and data required for system operation are also stored. The CPU 501, the ROM 502 and the RAM 503 are connected to each other through the bus 504. The input / output (I / O) interface 505 is also connected to the bus 504.

[0063] The following components are connected to the I / O interface 505: an input section 506 including a keyboard, a mouse, etc.; an output section 507 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker; a storage section 508 including a hard disk, etc.; and a communication section 509 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 509 performs communication processing via a network such as the Internet. A drive 510 is also connected to the I / O interface 505 as needed. A removable medium 511, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 510 as needed so that a computer program read therefrom is installed into the storage section 508 as needed.

[0064] In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication section 509, and / or installed from a removable medium 511. When the computer program is executed by a central processing unit (CPU) 501, various functions defined in the system of the present application are executed.

[0065] It should be noted that the computer-readable medium shown in the embodiment of the present application can be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium can be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, wherein a computer-readable computer program is carried. This propagated data signal can take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. A computer program contained on a computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0066] The flowchart and block diagram in the accompanying drawings illustrate the possible architecture, functions and operations of the system, method and computer program product according to various embodiments of the present application. Wherein, each box in the flowchart or block diagram can represent a module, a program segment, or a part of the code, and the above-mentioned module, program segment, or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0067] The units involved in the embodiments described in this application may be implemented by software or hardware, and the units described may also be set in a processor. The names of these units do not, in some cases, constitute limitations on the units themselves.

[0068] Another aspect of the present application provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor of a computer, the computer executes the fire identification method for a fire truck as described above. The computer-readable storage medium may be included in the electronic device described in the above embodiment, or may exist independently without being assembled into the electronic device.

[0069] Another aspect of the present application also provides a computer program product or a computer program, which includes a computer instruction stored in a computer-readable storage medium. A processor of a computer device reads the computer instruction from the computer-readable storage medium, and the processor executes the computer instruction, so that the computer device executes the fire identification method for a fire truck provided in each of the above embodiments.

[0070] The above embodiments are merely illustrative of the principles and effects of the present application, and are not intended to limit the present application. Anyone familiar with the technology may modify or change the above embodiments without violating the spirit and scope of the present application. Therefore, all equivalent modifications or changes made by a person of ordinary skill in the art without departing from the spirit and technical ideas disclosed in the present application shall still be covered by the claims of the present application.

Claims

1. A method for identifying fire intensity of a fire truck, characterized in that: The fire engine fire identification method comprises: The original fire image of the fire scene is collected based on the image collection device on the fire truck; Performing an image preprocessing operation on the original fire image, and performing a feature extraction on the original fire image after the preprocessing operation through a convolution operation to obtain fire image features of multiple channels, wherein the fire image features include flame morphology features and smoke diffusion features; Compressing the fire image features of multiple channels to obtain global fire intensity information, and performing secondary feature extraction on the global fire intensity information through a dynamic convolution kernel and a channel attention mechanism to obtain fire intensity recognition features; The flame edge and the smoke edge in the fire intensity recognition feature are calculated based on the large core adaptive module to optimize the fire intensity recognition feature. The flame height and flame shape are obtained according to the fire identification features after the feature optimization and the preset fire identification model, and the fire level of the fire scene is determined based on the flame height and the flame shape, and the fire level is reported to the fire control center to generate an emergency decision corresponding to the fire level.

2. The fire truck fire identification method according to claim 1, characterized in that: The image preprocessing operation performed on the original fire image includes: Performing cropping and rotation operations on the original fire image to obtain original fire images at multiple angles and multiple scales; Grayscale operations are performed on original fire images at multiple angles and multiple scales to obtain original fire images after the preprocessing operations.

3. The fire engine fire identification method according to claim 1, characterized in that: Extracting fire image features of multiple channels from the original fire image after the preprocessing operation through convolution operation includes: Sliding the convolution kernel on the original fire image according to a preset step length, and obtaining a pixel value of the fire image feature each time; The fire image features of multiple channels are obtained through sliding operations of multiple convolution kernels, and the fire image features include flame shape features and smoke diffusion features.

4. The fire truck fire identification method according to claim 1, characterized in that: The fire image features of multiple channels are compressed to obtain global fire information, and the global fire information is subjected to secondary feature extraction through a dynamic convolution kernel and a channel attention mechanism to obtain fire identification features including: Compressing the fire image features of the multiple channels based on a preset image compression algorithm to obtain the global fire intensity information; Inputting the global fire intensity information into a dynamic convolution kernel for feature extraction, wherein the convolution kernel size of the dynamic convolution kernel is determined based on the number of channels of the global fire intensity information; The attention weights of each channel of the global fire information after feature extraction in the dynamic convolution kernel are calculated, and the features of each channel of the global fire information after feature extraction in the dynamic convolution kernel are recalibrated based on the attention weights to obtain the fire identification features.

5. The fire identification method for a fire truck according to claim 1, characterized in that: Determining the fire level of the fire scene based on the flame height and the flame shape includes: Obtaining a flame height score of the fire scene according to the flame height; If the flame shape is a stable combustion type, the flame intensity score of the fire scene is determined as a preset first intensity score; If the flame shape is a fast spreading type, the flame intensity score of the fire scene is determined as a preset second intensity score; If the flame shape is an explosion type, the flame intensity score of the fire scene is determined as a preset third intensity score; A fire intensity score is obtained based on a preset flame height weight, a preset flame intensity weight, the flame height score and the flame intensity score, and the fire intensity score of the fire scene is determined according to the fire intensity score.

6. The fire intensity identification method of a fire truck according to claim 5, characterized in that: Determining the fire severity level of the fire scene according to the fire severity level score includes: If the fire intensity score is less than or equal to the preset first level score, the fire intensity at the fire scene is a minor fire. If the fire intensity score is greater than the preset first level score and less than the preset second level score, the fire intensity at the fire scene is a medium fire. If the fire intensity score is greater than or equal to the preset second level score, the fire intensity at the fire scene is a major fire.

7. The fire truck fire identification method according to claim 1, characterized in that: Calculating the flame edge and the smoke edge in the fire intensity recognition feature based on the large core adaptive module to optimize the fire intensity recognition feature includes: Decomposing a large kernel convolution into a first small kernel convolution and a second small kernel convolution, and performing deep feature extraction on the flame edge and the smoke edge in the fire intensity recognition feature based on the first small kernel convolution and the second small kernel convolution respectively; The features extracted by the first small-kernel convolution and the second small-kernel convolution are concatenated to obtain the fire intensity recognition features after feature optimization.

8. A fire identification device for a fire truck, characterized in that: The fire identification device for a fire truck comprises: An image acquisition module, used to acquire original fire images of the fire scene based on an image acquisition device on a fire truck; A first extraction module is used to perform an image preprocessing operation on the original fire image, and perform a feature extraction on the original fire image after the preprocessing operation through a convolution operation to obtain fire image features of multiple channels, wherein the fire image features include flame morphology features and smoke diffusion features; A second extraction module is used to compress the fire image features of multiple channels to obtain global fire intensity information, and perform secondary feature extraction on the global fire intensity information through a dynamic convolution kernel and a channel attention mechanism to obtain fire intensity recognition features; The feature optimization module is used to calculate the flame edge and the smoke edge in the fire intensity recognition feature based on the large core adaptive module to optimize the fire intensity recognition feature. A fire identification module is used to obtain the flame height and flame shape according to the fire identification features after the feature optimization and a preset fire identification model, and determine the fire level of the fire scene based on the flame height and the flame shape, and report the fire level to the fire control center to generate an emergency decision corresponding to the fire level.

9. An electronic device, characterized in that: The electronic device comprises: one or more processors; A storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, enables the electronic device to implement the fire identification method for a fire truck as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed by a processor of a computer, the computer is caused to execute the fire intensity identification method for a fire truck as described in any one of claims 1-7.

Images