Fire extinguishing risk assessment intelligent prevention and control method and system

Abstract

The application discloses a fire-fighting risk assessment intelligent prevention and control method and system, and relates to the technical field of intelligent fire fighting, which comprises the following steps: collecting original visible light images and environmental sensing data of a monitoring area; performing reflection suppression and signal enhancement processing on the original visible light images to output pure fire characteristic images; calculating an environmental complexity coefficient according to the pure fire characteristic images and the environmental sensing data; assigning dynamic weights to the pure fire characteristic images and the environmental sensing data based on the environmental complexity coefficient, and generating a regional fire situation comprehensive score; obtaining historical fire situation sample data and historical environmental state data, constructing a risk level cloud model based on the historical fire situation sample data and the historical environmental state data, constructing the environmental complexity coefficient which quantitatively represents the interference degree of the current scene, converting the abstract environmental interference factors into calculable numerical indexes, and giving the system the ability to objectively perceive the environmental interference level.

Description

Technical Field

[0001] This invention relates to the field of smart fire protection technology, and in particular to a smart prevention and control method and system for fire extinguishing risk assessment. Background Technology

[0002] With the rapid development of IoT, big data analytics, and AI technologies, smart fire protection has transformed from traditional passive alarm systems to proactive monitoring and intelligent prevention and control. Existing fire risk assessment and prevention technologies mainly rely on automatic fire alarm systems, which collect on-site data through smoke and heat detectors or infrared image sensors and make logical judgments based on preset fixed thresholds. In recent years, computer vision-based flame recognition algorithms and multi-sensor fusion technologies have gradually become research hotspots. Some advanced systems have begun to try to use digital twin technology to build urban safety supervision platforms to visualize and simulate fire risks.

[0003] Existing fire monitoring front-end sensing technologies often lack in-depth purification mechanisms for complex lighting and reflection environments. This makes it easy for strong light reflections or environmental artifacts in the monitoring images to be misjudged as flame features, seriously affecting the purity and accuracy of front-end feature extraction. Traditional data fusion strategies often use fixed weight allocation mechanisms, which cannot perceive changes in the complexity of the environment itself. When the monitoring area is obscured by dense smoke, has violent airflow disturbances, or has a cluttered background, the confidence levels of visual data and sensor data will dynamically shift. Fixed-weight fusion methods cannot adaptively adjust, which can easily lead to false alarms or missed alarms in the system. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides an intelligent prevention and control method for fire extinguishing risk assessment to address the deep purification mechanism for complex lighting and reflection environments. This results in strong light reflections or environmental artifacts in the monitoring screen being easily misjudged as flame features, seriously affecting the purity and accuracy of front-end feature extraction. Traditional data fusion strategies often use fixed weight allocation mechanisms, which cannot perceive changes in the complexity of the environment itself. When the monitoring area is obscured by dense smoke, has violent airflow disturbances, or has a cluttered background, the confidence of visual data and sensor data will dynamically shift. The fixed weight fusion method cannot adaptively adjust, which can easily lead to false alarms or missed alarms in the system.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides an intelligent prevention and control method for fire extinguishing risk assessment, which includes collecting raw visible light images and environmental sensor data of the monitored area; The original visible light image is subjected to reflection suppression and signal enhancement processing to output a clean fire feature image; The environmental complexity coefficient is calculated based on the clean fire feature image and environmental sensor data. Based on the environmental complexity coefficient, dynamic weights are assigned to the clean fire feature image and environmental sensor data, and a comprehensive score for regional fire conditions is generated. Historical fire sample data and historical environmental status data are acquired. A risk level cloud model is constructed based on the historical fire sample data and historical environmental status data. The comprehensive score of the regional fire is converted into a feature vector and input into the risk level cloud model to output a qualitative risk level. When the qualitative risk level triggers the control threshold, a matter-element model is established, and several candidate control schemes are generated by extension transformation. Calculate the overall superiority value of each candidate prevention and control scheme and select the optimal response strategy to drive the on-site prevention and control equipment to perform action response.

[0007] As a preferred embodiment of the intelligent prevention and control method for fire extinguishing risk assessment described in this invention, the specific steps of performing reflection suppression and signal enhancement processing on the original visible light image to output a clean fire feature image are as follows: The RGB three-channel components of the original visible light image are obtained, the transmittance estimate of each pixel is calculated using the dark channel prior algorithm, and an atmospheric light value matrix is ​​constructed. The original visible light image is convolved using a multi-scale retinal cortex enhancement algorithm to separate the illumination component and the reflection component. The illumination component is then suppressed using the transmittance estimate to obtain a preliminary dehazed image. The preliminary dehazed image is subjected to histogram equalization to enhance high-frequency edge information in the image, eliminate artifact interference caused by ambient light reflection, and output the clean fire feature image.

[0008] As a preferred embodiment of the intelligent prevention and control method for fire extinguishing risk assessment described in this invention, the specific steps for calculating the environmental complexity coefficient based on the pure fire feature image and environmental sensor data are as follows: The clean fire feature image is converted to grayscale, the average grayscale value of the image pixels is calculated, and the texture complexity entropy value of the clean fire feature image is calculated using the Shannon entropy formula. The texture complexity entropy value characterizes the degree of clutter in the image background. Obtain the temperature sampling sequence per unit time from the environmental sensing data, and calculate the standard deviation of the temperature sampling sequence as a parameter of temperature fluctuation intensity. The real-time smoke concentration value from the environmental sensing data is obtained, an environmental complexity calculation model is constructed, and the environmental complexity coefficient is calculated. The formula for calculating the environmental complexity coefficient is as follows: ; in, This represents the environmental complexity coefficient. The texture complexity entropy value is... This refers to the parameter indicating the severity of temperature fluctuations. The grayscale mean value is... The real-time smoke concentration value is... It is a natural constant. It is the natural logarithm function.

[0009] As a preferred embodiment of the intelligent fire risk assessment and control method of the present invention, the specific steps of allocating dynamic weights to the pure fire feature image and environmental sensor data based on the environmental complexity coefficient and generating a comprehensive regional fire score are as follows: Based on the environmental complexity coefficient, a visual weight function and an environmental weight function are constructed. When the environmental complexity coefficient increases, the output value of the visual weight function is reduced and the output value of the environmental weight function is increased. The square root of the mean gray level and the entropy of the texture complexity of the pure fire feature image is extracted as a visual feature factor, and the sum of the real-time smoke concentration and the real-time ambient temperature in the environmental sensing data is extracted as an environmental feature factor. The visual feature factors are weighted using the visual weight function, and the environmental feature factors are weighted using the environmental weight function. The summation yields the comprehensive regional fire score. The formula for calculating the comprehensive regional fire score is as follows: ; in, The overall score for regional fire situation, The grayscale mean value is... The environmental complexity coefficient is... The texture complexity entropy value is... The real-time smoke concentration value is... The real-time ambient temperature value. It is a natural constant.

[0010] As a preferred embodiment of the intelligent prevention and control method for fire extinguishing risk assessment described in this invention, the steps of acquiring historical fire sample data and historical environmental state data, constructing a risk level cloud model based on the historical fire sample data and historical environmental state data, converting the regional fire comprehensive score into a feature vector and inputting it into the risk level cloud model, and outputting a qualitative risk level are as follows: Statistical analysis was performed on the historical fire sample data and historical environmental state data to calculate the expected value, entropy value and hyperentropy value corresponding to each preset risk level, and to construct a positive cloud generator model. The comprehensive fire score of the region is used as a cloud droplet input to the positive cloud generator model to generate a temporary entropy value that follows a normal distribution. The temporary entropy value introduces a random perturbation factor to simulate the uncertainty of the environment. The degree of certainty of the overall fire situation score in the area relative to each preset risk level is calculated, and the preset risk level corresponding to the maximum degree of certainty is selected as the qualitative risk level. The formula for calculating the degree of certainty is as follows: ; in, For certainty, The overall score for the fire situation in the area is calculated as follows: The expected value, The temporary entropy value is... It is an exponential function.

[0011] As a preferred embodiment of the intelligent prevention and control method for fire extinguishing risk assessment described in this invention, the step of establishing a matter-element model when the qualitative risk level triggers the control threshold and generating several candidate prevention and control schemes using extension transformation includes the following steps: When the qualitative risk level exceeds the preset control threshold, a matter-element model is constructed that includes the control target, fire-fighting equipment, and motion values. The matter-element model is subjected to permutation and addition / deletion transformations, and the corresponding fire-fighting equipment combination is matched according to the current fire type. The fire-fighting equipment combination includes a sprinkler system, a smoke exhaust fan, and a fireproof roller shutter. The fire-fighting equipment combination is logically arranged to generate several candidate prevention and control schemes, including single equipment actions and multi-equipment linkage actions.

[0012] As a preferred embodiment of the intelligent prevention and control method for fire extinguishing risk assessment described in this invention, the specific steps of calculating the comprehensive superiority value of each candidate prevention and control scheme and selecting the optimal response strategy to drive the on-site prevention and control equipment to perform action response are as follows: Obtain the estimated fire extinguishing efficiency coefficient, response rate factor, estimated energy consumption cost, and equipment response delay time for each of the candidate prevention and control schemes. A comprehensive excellence evaluation model is constructed, using the product of the estimated fire extinguishing efficiency coefficient and the response rate factor as the numerator, and the weighted sum of the estimated energy consumption cost and the equipment response delay time as the denominator, to calculate the comprehensive excellence value of each candidate prevention and control scheme. The formula for calculating the comprehensive excellence value is as follows: ; in, The overall excellence score, The estimated fire extinguishing efficiency coefficient is... The response rate factor, For the estimated energy consumption cost, The device response delay time. This is the energy consumption balance adjustment constant. This is the time balance adjustment constant; The candidate prevention and control scheme with the highest comprehensive excellence value is selected as the optimal handling strategy, and control commands are generated to drive the on-site prevention and control equipment to perform actions and responses.

[0013] Secondly, the present invention provides an intelligent prevention and control system for fire extinguishing risk assessment, including a data acquisition module, an image processing module, a complexity calculation module, a dynamic assessment module, a risk simulation module, a scheme generation module, and a strategy execution module; The data acquisition module is used to acquire raw visible light images and environmental sensor data of the monitored area; The image processing module is used to perform reflection suppression and signal enhancement processing on the original visible light image and output a clean fire feature image; The complexity calculation module is used to calculate the environmental complexity coefficient based on the clean fire feature image and environmental sensor data. The dynamic evaluation module assigns dynamic weights to the clean fire feature image and environmental sensor data based on the environmental complexity coefficient, and generates a comprehensive score for the regional fire situation. The risk simulation module is used to acquire historical fire sample data and historical environmental status data, construct a risk level cloud model based on the historical fire sample data and historical environmental status data, convert the regional fire comprehensive score into a feature vector and input it into the risk level cloud model, and output a qualitative risk level. The scheme generation module is used to establish a matter-element model when the qualitative risk level triggers the control threshold, and to generate several candidate control schemes by using extension transformation. The strategy execution module is used to calculate the comprehensive superiority value of each candidate prevention and control scheme and select the optimal handling strategy to drive the on-site prevention and control equipment to perform action response.

[0014] Thirdly, the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, wherein when the computer program is executed by the processor, it implements any step of the intelligent prevention and control method for fire extinguishing risk assessment as described in the first aspect of the present invention.

[0015] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the intelligent prevention and control method for fire extinguishing risk assessment as described in the first aspect of the present invention.

[0016] The beneficial effects of this invention are as follows: by performing targeted reflection suppression and signal enhancement processing on the original visible light image, it effectively filters out specular reflection interference under complex lighting conditions and enhances weak fire characteristics, solves the problem of key visual feature distortion under adverse imaging conditions, deeply integrates the texture information of the image with the fluctuation characteristics of real-time environmental data, constructs an environmental complexity coefficient that quantitatively represents the degree of interference in the current scene, transforms abstract environmental interference factors into calculable numerical indicators, and endows the system with the ability to objectively perceive the level of environmental interference. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A flowchart for intelligent prevention and control methods for fire extinguishing risk assessment; Figure 2 This is a schematic diagram of an intelligent fire prevention and control system for fire extinguishing risk assessment. Detailed Implementation

[0019] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0020] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0021] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0022] Reference Figures 1-2 As an embodiment of the present invention, this embodiment provides an intelligent prevention and control method for fire extinguishing risk assessment, comprising the following steps: S1. Collect raw visible light images and environmental sensor data of the monitored area.

[0023] Furthermore, wide-angle surveillance cameras deployed at high points in the monitoring area are used to capture continuous frames of the monitoring area, obtaining raw visible light images that include the environmental background and potential fire sources. The spatial location of the monitoring area is divided into grids by a distributed sensor network, and the real-time ambient temperature, relative humidity and smoke particle concentration values ​​at each grid node are collected to obtain an environmental sensing data set. The timestamps of the original visible light image and the sampling timestamps of the environmental sensing data set are synchronized using a high-precision crystal oscillator clock signal to eliminate timing errors caused by data transmission delays, thus obtaining time-synchronized multimodal monitoring data.

[0024] It should be noted that by synchronously acquiring and aligning multimodal data, monitoring errors caused by the lag or asynchrony of data from a single sensor can be effectively eliminated, providing high-fidelity and highly consistent basic data support for subsequent accurate assessment of fire conditions.

[0025] S2. Perform reflection suppression and signal enhancement processing on the original visible light image to output a clean fire feature image.

[0026] Furthermore, the RGB three-channel components of the original visible light image are obtained, the transmittance estimate of each pixel is calculated using the dark channel prior algorithm, and an atmospheric light value matrix is ​​constructed. A multi-scale retinal cortex enhancement algorithm is used to perform convolution operations on the original visible light image to separate the illumination component and the reflection component. The illumination component is suppressed using the transmittance estimate to obtain a preliminary dehazed image. Histogram equalization is performed on the initially dehazed image to enhance high-frequency edge information, eliminate artifact interference caused by ambient light reflection, and output a clean fire feature image.

[0027] It should be noted that by combining dark channel prior and retinal cortex enhancement algorithms, not only is the interference of smoke and complex lighting on the image effectively removed, but the weak visual features in the early stage of a fire are also highlighted, which greatly improves the accuracy of fire source identification in low visibility environments.

[0028] S3. Calculate the environmental complexity coefficient based on the pure fire feature image and environmental sensor data.

[0029] Furthermore, the pure fire feature image is converted to grayscale, the average grayscale value of the image pixels is calculated, and the texture complexity entropy value of the pure fire feature image is calculated using the Shannon entropy formula. The texture complexity entropy value characterizes the degree of clutter in the image background. Acquire the temperature sampling sequence per unit time from environmental sensor data, calculate the standard deviation of the temperature sampling sequence, and use it as a parameter for the severity of temperature fluctuations. Obtain real-time smoke concentration values ​​from environmental sensor data, construct an environmental complexity calculation model, and calculate the environmental complexity coefficient. The formula for calculating the environmental complexity coefficient is as follows: ; in, This represents the environmental complexity coefficient. This represents the texture complexity entropy value. This is a parameter representing the severity of temperature fluctuations. The grayscale mean is... This represents the real-time smoke concentration value. It is a natural constant. It is the natural logarithm function.

[0030] It should be noted that the environmental complexity coefficient innovatively integrates the image texture entropy and the fluctuation characteristics of environmental physical quantities, which can objectively quantify the interference level of the current monitoring environment and avoid misjudgment caused by sudden environmental changes in traditional methods.

[0031] S4. Based on the environmental complexity coefficient, assign dynamic weights to the pure fire feature images and environmental sensor data, and generate a comprehensive score for regional fire conditions.

[0032] Furthermore, a visual weight function and an environmental weight function are constructed based on the environmental complexity coefficient. When the environmental complexity coefficient increases, the output value of the visual weight function is reduced and the output value of the environmental weight function is increased. The square root of the mean gray level and the entropy of texture complexity in the pure fire feature image is extracted as the visual feature factor, and the sum of the real-time smoke concentration and the real-time ambient temperature in the environmental sensor data is extracted as the environmental feature factor. The visual feature factors are weighted using a visual weighting function, and the environmental feature factors are weighted using an environmental weighting function. The summation yields the comprehensive regional fire score. The formula for calculating the comprehensive regional fire score is as follows: ; in, The overall score for regional fire situation, The grayscale mean is... This represents the environmental complexity coefficient. This represents the texture complexity entropy value. This represents the real-time smoke concentration value. This is the real-time ambient temperature value. It is a natural constant.

[0033] It should be noted that by introducing a dynamic weight allocation mechanism based on environmental complexity, adaptive complementarity between visual features and environmental sensing data is achieved. The system automatically prioritizes sensor data in harsh environments and image features in clear environments, thus ensuring the reliability of the overall fire score under various extreme conditions.

[0034] S5. Obtain historical fire sample data and historical environmental status data, construct a risk level cloud model based on the historical fire sample data and historical environmental status data, and convert the regional fire comprehensive score into a feature vector and input it into the risk level cloud model to output the qualitative risk level.

[0035] Furthermore, statistical analysis is conducted on historical fire sample data and historical environmental status data to calculate the expected value, entropy value, and hyperentropy value corresponding to each preset risk level, and to construct a positive cloud generator model. The regional fire situation comprehensive score is used as cloud droplet input to the positive cloud generator model to generate a temporary entropy value that follows a normal distribution. The temporary entropy value introduces a random perturbation factor to simulate the uncertainty of the environment. The degree of certainty of the overall fire situation score in the area relative to each preset risk level is calculated. The preset risk level corresponding to the maximum degree of certainty is selected as the qualitative risk level. The formula for calculating the degree of certainty is as follows: ; in, For certainty, The overall score for regional fire situation, For the expected value, This is a temporary entropy value. It is an exponential function.

[0036] It should be noted that by using cloud models to handle the ambiguity and randomness in fire assessment, quantitative comprehensive scores can be scientifically transformed into qualitative risk levels, solving the technical problems of blurred boundaries and low fault tolerance in traditional threshold determination methods in unstructured environments.

[0037] S6. When the qualitative risk level triggers the control threshold, establish a matter-element model and use extension transformation to generate several candidate control schemes.

[0038] Furthermore, when the qualitative risk level exceeds the preset control threshold, a matter-element model is constructed that includes the prevention and control target, fire-fighting equipment, and motion values. The matter-element model is subjected to permutation and addition / deletion transformations. The corresponding fire-fighting equipment combination is matched according to the current fire type. The fire-fighting equipment combination includes sprinkler system, smoke exhaust fan and fireproof roller shutter. The fire protection equipment combination is logically arranged to generate several candidate prevention and control schemes, including single equipment actions and multi-equipment linkage actions.

[0039] It should be noted that by using the matter-element transformation theory of extension theory for scheme derivation, the limitations of fixed plans are overcome, and a variety of targeted equipment linkage combinations can be flexibly generated according to the real-time fire situation, thereby improving the adaptability and comprehensiveness of emergency prevention and control strategies.

[0040] S7. Calculate the overall superiority value of each candidate prevention and control plan and select the optimal response strategy to drive the on-site prevention and control equipment to perform actions and responses.

[0041] Furthermore, obtain the estimated fire extinguishing efficiency coefficient, response rate factor, estimated energy consumption cost, and equipment response delay time for each candidate prevention and control scheme; A comprehensive excellence evaluation model is constructed, using the product of the estimated fire extinguishing efficiency coefficient and the response rate factor as the numerator, and the weighted sum of the estimated energy consumption cost and equipment response delay time as the denominator, to calculate the comprehensive excellence value of each candidate prevention and control scheme. The formula for calculating the comprehensive excellence value is as follows: ; in, The overall excellence score, To estimate the fire extinguishing efficiency coefficient, For the response rate factor, To estimate energy costs, For device response delay time, This is the energy consumption balance adjustment constant. This is the time balance adjustment constant; The candidate prevention and control scheme with the highest comprehensive excellence value is selected as the optimal response strategy, and control commands are generated to drive the on-site prevention and control equipment to perform actions and responses.

[0042] It should be noted that by constructing a multi-dimensional excellence evaluation model that includes efficiency, rate, energy consumption and delay, the optimal strategy with the highest cost performance and fastest response can be selected from many candidate solutions within milliseconds, realizing the leap from "passive response" to "intelligent optimization" in fire response.

[0043] This embodiment also provides an intelligent fire extinguishing risk assessment and control system, including: a data acquisition module, an image processing module, a complexity calculation module, a dynamic assessment module, a risk simulation module, a scheme generation module, and a strategy execution module; The data acquisition module is used to acquire raw visible light images and environmental sensor data of the monitored area; The image processing module is used to perform reflection suppression and signal enhancement processing on the original visible light image and output a clean fire feature image; The complexity calculation module is used to calculate the environmental complexity coefficient based on the clean fire feature image and environmental sensor data; The dynamic evaluation module assigns dynamic weights to clean fire feature images and environmental sensor data based on the environmental complexity coefficient, and generates a comprehensive score for regional fire conditions. The risk simulation module is used to acquire historical fire sample data and historical environmental status data, construct a risk level cloud model based on the historical fire sample data and historical environmental status data, and convert the regional fire comprehensive score into a feature vector and input it into the risk level cloud model to output a qualitative risk level. The scheme generation module is used to establish a matter-element model when the qualitative risk level triggers the control threshold, and use extension transformation to generate several candidate control schemes. The strategy execution module is used to calculate the comprehensive superiority value of each candidate prevention and control plan and select the optimal response strategy to drive the on-site prevention and control equipment to perform actions and responses.

[0044] This embodiment also provides a computer device applicable to the intelligent prevention and control method for fire extinguishing risk assessment, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to realize the intelligent prevention and control method for fire extinguishing risk assessment as proposed in the above embodiment.

[0045] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0046] This embodiment also provides a storage medium storing a computer program, which, when executed by a processor, implements the intelligent prevention and control method for fire extinguishing risk assessment as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0047] In summary, this invention effectively filters out specular reflection interference under complex lighting conditions and enhances weak fire characteristics by performing targeted reflection suppression and signal enhancement processing on the original visible light image. It solves the problem of key visual feature distortion under adverse imaging conditions. Furthermore, it deeply integrates the texture information of the image with the fluctuation characteristics of real-time environmental data to construct an environmental complexity coefficient that quantitatively represents the degree of interference in the current scene. This transforms abstract environmental interference factors into calculable numerical indicators, giving the system the ability to objectively perceive the level of environmental interference.

[0048] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A smart prevention and control method for fire extinguishing risk assessment, characterized in that: include: Collect raw visible light images and environmental sensor data of the monitored area; The original visible light image is subjected to reflection suppression and signal enhancement processing to output a clean fire feature image; The environmental complexity coefficient is calculated based on the clean fire feature image and environmental sensor data. Based on the environmental complexity coefficient, dynamic weights are assigned to the clean fire feature image and environmental sensor data, and a comprehensive score for regional fire conditions is generated. Historical fire sample data and historical environmental status data are acquired. A risk level cloud model is constructed based on the historical fire sample data and historical environmental status data. The comprehensive score of the regional fire is converted into a feature vector and input into the risk level cloud model to output a qualitative risk level. When the qualitative risk level triggers the control threshold, a matter-element model is established, and several candidate control schemes are generated by extension transformation. Calculate the overall superiority value of each candidate prevention and control scheme and select the optimal response strategy to drive the on-site prevention and control equipment to perform action response.

2. The intelligent prevention and control method for fire extinguishing risk assessment as described in claim 1, characterized in that: The specific steps for performing reflection suppression and signal enhancement processing on the original visible light image to output a clean fire feature image are as follows: The RGB three-channel components of the original visible light image are obtained, the transmittance estimate of each pixel is calculated using the dark channel prior algorithm, and an atmospheric light value matrix is ​​constructed. The original visible light image is convolved using a multi-scale retinal cortex enhancement algorithm to separate the illumination component and the reflection component. The illumination component is then suppressed using the transmittance estimate to obtain a preliminary dehazed image. The preliminary dehazed image is subjected to histogram equalization to enhance high-frequency edge information in the image, eliminate artifact interference caused by ambient light reflection, and output the clean fire feature image.

3. The intelligent prevention and control method for fire extinguishing risk assessment as described in claim 2, characterized in that: The specific steps for calculating the environmental complexity coefficient based on the clean fire feature image and environmental sensor data are as follows: The clean fire feature image is converted to grayscale, the average grayscale value of the image pixels is calculated, and the texture complexity entropy value of the clean fire feature image is calculated using the Shannon entropy formula. The texture complexity entropy value characterizes the degree of clutter in the image background. Obtain the temperature sampling sequence per unit time from the environmental sensing data, and calculate the standard deviation of the temperature sampling sequence as a parameter of temperature fluctuation intensity. The real-time smoke concentration value from the environmental sensing data is obtained, an environmental complexity calculation model is constructed, and the environmental complexity coefficient is calculated. The formula for calculating the environmental complexity coefficient is as follows: ; in, This represents the environmental complexity coefficient. The texture complexity entropy value is... This refers to the parameter indicating the severity of temperature fluctuations. The grayscale mean value is... The real-time smoke concentration value is... It is a natural constant. It is the natural logarithm function.

4. The intelligent prevention and control method for fire extinguishing risk assessment as described in claim 3, characterized in that: The specific steps for assigning dynamic weights to the clean fire feature image and environmental sensor data based on the environmental complexity coefficient, and generating a comprehensive regional fire score, are as follows: Based on the environmental complexity coefficient, a visual weight function and an environmental weight function are constructed. When the environmental complexity coefficient increases, the output value of the visual weight function is reduced and the output value of the environmental weight function is increased. The square root of the mean gray level and the entropy of the texture complexity of the pure fire feature image is extracted as a visual feature factor, and the sum of the real-time smoke concentration and the real-time ambient temperature in the environmental sensing data is extracted as an environmental feature factor. The visual feature factors are weighted using the visual weighting function, and the environmental feature factors are weighted using the environmental weighting function. The summation yields the comprehensive regional fire score. The formula for calculating the comprehensive regional fire score is as follows: ; in, The overall score for regional fire situation, The grayscale mean value is... The environmental complexity coefficient is... The texture complexity entropy value is... The real-time smoke concentration value is... The real-time ambient temperature value. It is a natural constant.

5. The intelligent prevention and control method for fire extinguishing risk assessment as described in claim 4, characterized in that: The steps for acquiring historical fire sample data and historical environmental status data, constructing a risk level cloud model based on the historical fire sample data and historical environmental status data, converting the regional fire comprehensive score into a feature vector and inputting it into the risk level cloud model, and outputting a qualitative risk level are as follows: Statistical analysis was performed on the historical fire sample data and historical environmental state data to calculate the expected value, entropy value and hyperentropy value corresponding to each preset risk level, and to construct a positive cloud generator model. The comprehensive fire score of the region is used as a cloud droplet input to the positive cloud generator model to generate a temporary entropy value that follows a normal distribution. The temporary entropy value introduces a random perturbation factor to simulate the uncertainty of the environment. The degree of certainty of the comprehensive fire score of the area relative to each preset risk level is calculated, and the preset risk level corresponding to the maximum degree of certainty is selected as the qualitative risk level. The formula for calculating the degree of certainty is as follows: ; in, For certainty, The overall score for the fire situation in the area is calculated as follows: The expected value, The temporary entropy value is... It is an exponential function.

6. The intelligent prevention and control method for fire extinguishing risk assessment as described in claim 5, characterized in that: When the qualitative risk level triggers the control threshold, a matter-element model is established, and several candidate control schemes are generated using extension transformation. The specific steps are as follows: When the qualitative risk level exceeds the preset control threshold, a matter-element model is constructed that includes the control target, fire-fighting equipment, and motion values. The matter-element model is subjected to permutation and addition / deletion transformations, and the corresponding fire-fighting equipment combination is matched according to the current fire type. The fire-fighting equipment combination includes a sprinkler system, a smoke exhaust fan, and a fireproof roller shutter. The fire-fighting equipment combination is logically arranged to generate several candidate prevention and control schemes, including single equipment actions and multi-equipment linkage actions.

7. The intelligent prevention and control method for fire extinguishing risk assessment as described in claim 6, characterized in that: The steps for calculating the comprehensive superiority value of each candidate prevention and control scheme and selecting the optimal response strategy to drive the on-site prevention and control equipment to perform action response are as follows: Obtain the estimated fire extinguishing efficiency coefficient, response rate factor, estimated energy consumption cost, and equipment response delay time for each of the candidate prevention and control schemes. A comprehensive excellence evaluation model is constructed, using the product of the estimated fire extinguishing efficiency coefficient and the response rate factor as the numerator, and the weighted sum of the estimated energy consumption cost and the equipment response delay time as the denominator, to calculate the comprehensive excellence value of each candidate prevention and control scheme. The formula for calculating the comprehensive excellence value is as follows: ; in, The overall excellence score, The estimated fire extinguishing efficiency coefficient is... The response rate factor, For the estimated energy consumption cost, The device response delay time. This is the energy consumption balance adjustment constant. This is the time balance adjustment constant; The candidate prevention and control scheme with the highest comprehensive excellence value is selected as the optimal handling strategy, and control commands are generated to drive the on-site prevention and control equipment to perform actions and responses.

8. A fire extinguishing risk assessment and intelligent control system, based on the fire extinguishing risk assessment and intelligent control method according to any one of claims 1 to 7, characterized in that: It includes a data acquisition module, an image processing module, a complexity calculation module, a dynamic evaluation module, a risk simulation module, a solution generation module, and a strategy execution module; The data acquisition module is used to acquire raw visible light images and environmental sensor data of the monitored area; The image processing module is used to perform reflection suppression and signal enhancement processing on the original visible light image and output a clean fire feature image; The complexity calculation module is used to calculate the environmental complexity coefficient based on the clean fire feature image and environmental sensor data. The dynamic evaluation module assigns dynamic weights to the clean fire feature image and environmental sensor data based on the environmental complexity coefficient, and generates a comprehensive score for the regional fire situation. The risk simulation module is used to acquire historical fire sample data and historical environmental status data, construct a risk level cloud model based on the historical fire sample data and historical environmental status data, convert the regional fire comprehensive score into a feature vector and input it into the risk level cloud model, and output a qualitative risk level. The scheme generation module is used to establish a matter-element model when the qualitative risk level triggers the control threshold, and to generate several candidate control schemes by using extension transformation. The strategy execution module is used to calculate the comprehensive superiority value of each candidate prevention and control scheme and select the optimal handling strategy to drive the on-site prevention and control equipment to perform action response.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the intelligent prevention and control method for fire extinguishing risk assessment as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the intelligent prevention and control method for fire extinguishing risk assessment as described in any one of claims 1 to 7.

Images