Fire-fighting monitoring method and system based on field environment information acquisition

Abstract

The invention is suitable for the technical field of fire-fighting monitoring, and provides a fire-fighting monitoring method and system based on field environment information acquisition, and the method comprises the following steps: acquiring field environment information based on a sensor and a camera, the field environment information comprising temperature information, humidity information, smoke concentration information and a monitoring video, each piece of field environment information is marked with a timestamp and a place stamp; performing preliminary analysis on the field environment information to determine a field environment area needing to be monitored, and obtaining fire-fighting monitoring information; security personnel transfer information is generated based on the fire-fighting monitoring information, the security personnel transfer information comprises a plurality of personnel numbers, and each personnel number corresponds to a responsible area and a responsible project; and carrying out deep analysis on the fire-fighting monitoring information, generating fire-fighting early warning information when potential safety hazards exist, and sending the fire-fighting early warning information to a security personnel account. Therefore, the information processing efficiency is higher, automation and intelligence of personnel scheduling are realized, and the emergency response efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of fire monitoring, and in particular to a fire monitoring method and system based on obtaining on-site environmental information. Background Art

[0002] In the field of fire safety monitoring, traditional monitoring methods mainly rely on manual patrols and fixed smoke detectors. However, these methods have obvious limitations. With the rapid development of sensor technology and computer vision technology, real-time collection and analysis of on-site environmental information has become possible. Sensors can continuously monitor key indicators such as temperature, humidity and smoke concentration, while cameras can capture video information in the monitoring area, providing intuitive environmental images and personnel behavior dynamics. Integrating this information can provide more comprehensive, accurate and real-time data support for fire safety monitoring.

[0003] However, there is no more comprehensive and in-depth analysis of sensor information, and when there are safety hazards, how to quickly and effectively mobilize security personnel for emergency response is a problem that needs to be solved. Therefore, it is necessary to provide a fire monitoring method and system based on on-site environmental information acquisition to solve the above problems. Summary of the invention

[0004] In view of the deficiencies in the prior art, the purpose of the present invention is to provide a fire monitoring method and system based on on-site environmental information acquisition to solve the problems existing in the above-mentioned background technology.

[0005] The present invention is implemented as follows: a fire monitoring method based on obtaining on-site environmental information, the method comprising the following steps: Based on the sensors and cameras, the on-site environment information is obtained, the on-site environment information includes temperature information, humidity information, smoke concentration information and monitoring video, and each on-site environment information is marked with a time stamp and a location stamp; Conduct a preliminary analysis of the on-site environmental information to determine the on-site environmental areas that need to be monitored and obtain fire monitoring information; Generate security personnel mobilization information based on fire monitoring information. The security personnel mobilization information includes several personnel numbers, each of which corresponds to a responsible area and a responsible project; Conduct in-depth analysis of fire monitoring information to determine whether there are any safety hazards. When there are safety hazards, generate fire warning information and send it to the corresponding security personnel account.

[0006] Another object of the present invention is to provide a fire monitoring system based on on-site environmental information acquisition, the system comprising: An environmental information collection module is used to collect on-site environmental information based on sensors and cameras. The on-site environmental information includes temperature information, humidity information, smoke concentration information, and monitoring video. Each on-site environmental information is marked with a timestamp and a location stamp. The monitoring information determination module is used to perform preliminary analysis on the on-site environmental information to determine the on-site environmental area that needs to be monitored and obtain fire monitoring information; The security personnel mobilization module is used to generate security personnel mobilization information based on fire monitoring information. The security personnel mobilization information includes several personnel numbers, each of which corresponds to a responsible area and a responsible project; The fire protection in-depth analysis module is used to conduct in-depth analysis of fire protection monitoring information to determine whether there are any safety hazards. When there are safety hazards, fire protection warning information is generated and sent to the corresponding security personnel account.

[0007] Compared with the prior art, the present invention has the following beneficial effects: The present invention determines the on-site environmental area that needs to be monitored by performing a preliminary analysis of the on-site environmental information, obtains fire monitoring information, and then deeply analyzes the fire monitoring information to determine whether there are safety hazards and generate fire warning information. In this way, the information processing efficiency is higher and the fire data monitoring effect is better. In addition, security personnel mobilization information is generated based on the fire monitoring information, realizing the automation and intelligence of personnel dispatch and improving the efficiency of emergency response. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 The present invention is a flow chart of a fire monitoring method based on obtaining on-site environmental information.

[0009] Figure 2 The present invention is a flow chart for obtaining fire monitoring information in a fire monitoring method based on obtaining on-site environmental information.

[0010] Figure 3 The present invention is a flowchart for feature analysis in a fire monitoring method based on on-site environmental information acquisition.

[0011] Figure 4 The present invention is a flow chart for generating security personnel mobilization information in a fire monitoring method based on obtaining on-site environmental information.

[0012] Figure 5 The present invention is a flow chart for determining safety hazards in a fire monitoring method based on obtaining on-site environmental information.

[0013] Figure 6 The schematic diagram is a structural diagram of a fire monitoring system based on obtaining on-site environmental information. DETAILED DESCRIPTION

[0014] In order to make the purpose, technical solution and advantages of the present invention clearer, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0015] The specific implementation of the present invention is described in detail below in conjunction with specific embodiments.

[0016] like Figure 1 As shown, an embodiment of the present invention provides a fire monitoring method based on obtaining on-site environmental information, and the method includes the following steps: S100, obtaining on-site environmental information based on sensors and cameras, wherein the on-site environmental information includes temperature information, humidity information, smoke concentration information, and surveillance video, and each on-site environmental information is marked with a timestamp and a location stamp; S200, performing a preliminary analysis on the on-site environmental information to determine the on-site environmental area that needs to be monitored, and obtaining fire monitoring information; S300, generating security personnel mobilization information based on fire monitoring information, the security personnel mobilization information including a number of personnel numbers, each personnel number corresponding to a responsible area and a responsible project; S400, deeply analyze the fire monitoring information to determine whether there are safety hazards. When there are safety hazards, generate fire warning information and send the fire warning information to the corresponding security personnel account.

[0017] It should be noted that with the rapid development of sensor technology and computer vision technology, real-time collection and analysis of on-site environmental information has become possible. Sensors can continuously monitor key indicators such as temperature, humidity and smoke concentration, while cameras can capture video information in the monitoring area, providing intuitive environmental images and personnel behavior dynamics. Integrating this information can provide more comprehensive, accurate and real-time data support for fire safety monitoring. However, there is currently no more comprehensive and in-depth analysis of sensor information, and when there are safety hazards, how to quickly and effectively mobilize security personnel for emergency response is a key issue.

[0018] In the embodiment of the present invention, sensors and cameras are first installed in each environmental area, and then on-site environmental information is collected based on the sensors and cameras. The on-site environmental information specifically includes temperature information, humidity information, smoke concentration information and monitoring video, and each on-site environmental information is marked with a timestamp and a location stamp. Then, the temperature information, humidity information, smoke concentration information and monitoring video are preliminarily analyzed to determine the on-site environmental area that needs to be monitored, and the fire monitoring information is obtained. The fire monitoring information is the focus of attention, and the security personnel mobilization information is automatically generated according to the fire monitoring information. The security personnel mobilization information includes several personnel numbers, and each personnel number corresponds to a responsible area and a responsible project, which realizes the automation and intelligence of personnel dispatch and improves the efficiency of emergency response. Finally, the fire monitoring information is deeply analyzed to determine whether there are safety hazards. When there are safety hazards, fire warning information is generated and sent to the corresponding security personnel account, which improves the accuracy and practicality of the warning.

[0019] like Figure 2 As shown, as a preferred embodiment of the present invention, the step of performing preliminary analysis on the on-site environmental information to determine the on-site environmental area that needs to be monitored and obtaining fire monitoring information specifically includes: S201, comparing the temperature information, humidity information and smoke concentration information in each on-site environment information with the corresponding temperature threshold, humidity threshold and smoke concentration threshold, and determining the corresponding location stamp as the on-site environment area to be monitored; S202, performing feature analysis on the surveillance video in each on-site environment information to determine whether there are flame features, crowd gathering features, or abnormal behavior features, and if so, determining the corresponding location stamp as the on-site environment area to be monitored; S203, summarizing the on-site environmental information of the on-site environmental area that needs to be monitored to obtain fire monitoring information.

[0020] In the embodiment of the present invention, in order to preliminarily and quickly identify the on-site environment area that needs to be monitored, the temperature information, humidity information and smoke concentration information in each on-site environment information will be compared with the corresponding temperature threshold, humidity threshold and smoke concentration threshold. The temperature threshold, humidity threshold and smoke concentration threshold are all set in advance. When it is higher than the threshold, the corresponding location stamp is determined to be the on-site environment area that needs to be monitored. In addition, the monitoring video in each on-site environment information will be analyzed for features to determine whether there are flame features, personnel gathering features or abnormal behavior features (such as running, falling, climbing). If so, the corresponding location stamp is determined to be the on-site environment area that needs to be monitored. Finally, the on-site environment information of all on-site environment areas that need to be monitored is summarized to obtain fire monitoring information.

[0021] As a preferred embodiment of the present invention, the step of comparing the temperature information, humidity information and smoke concentration information in each on-site environment information with the corresponding temperature threshold, humidity threshold and smoke concentration threshold to determine that the corresponding location stamp is the on-site environment area to be monitored specifically includes: Obtain the temperature information, humidity information and smoke concentration information of each on-site environment information, According to the change trend of temperature information, humidity information and smoke concentration information within a preset time period, the change rate of temperature, humidity and smoke concentration is obtained; According to the changing rates of temperature, humidity and smoke concentration, the corresponding temperature dynamic weight, humidity dynamic weight and smoke dynamic weight are given; Get temperature threshold, humidity threshold and smoke concentration threshold; The ratio between the temperature information and the temperature threshold is used as the temperature contribution factor, the ratio between the smoke concentration information and the smoke concentration threshold is used as the smoke contribution factor, and the inverse of the ratio between the humidity information and the humidity threshold is used as the humidity sharing factor; The temperature contribution factor, smoke contribution factor and humidity sharing factor are fused by multiplication to capture the interaction effect between each variable and obtain a comprehensive interaction factor. The temperature dynamic weight, humidity dynamic weight and smoke dynamic weight are combined in an additive manner to obtain a comprehensive dynamic weight, and then the comprehensive dynamic weight is added to the comprehensive interaction factor to calculate a preliminary score; The preliminary score was amplified by applying an exponential function to obtain a comprehensive risk score; The fuzzy logic method is used to divide the hazard level, and the comprehensive hazard score is compared with the hazard level to determine whether the current area is a field environment area that needs to be monitored.

[0022] In the embodiment of the present invention, the combined effects of high temperature, low humidity and high smoke concentration are considered, and the coupling effect between the three is reflected in the form of product to accurately quantify the potential danger. The rate of change (speed) of environmental variables and the change of the rate of change (acceleration) are also included in the evaluation. Environmental variables with high rates of change and high acceleration of change are usually accompanied by greater uncertainty, so they are given higher weights to assess the urgency of the risk. Finally, the exponential function is introduced in the comprehensive score to enhance the ability to identify critical scenarios.

[0023] like Figure 3 As shown, as a preferred embodiment of the present invention, the step of performing feature analysis on the surveillance video in each on-site environment information specifically includes: S2021, converting the surveillance video from RGB color space to YUV color space, applying color threshold segmentation, extracting potential flame areas, and applying morphological operations to the flame areas to determine flame features; S2022, using a background subtraction algorithm to separate foreground objects from the video frame, identifying people through a human body detection algorithm, and obtaining a human gathering feature; S2023, extracting the movement trajectory and behavior characteristics of the person based on the optical flow method, comparing the behavior characteristics with the normal pattern, calculating the behavior similarity, and obtaining the abnormal behavior characteristics.

[0024] In the embodiment of the present invention, the monitoring video is first converted from the RGB color space to the YUV color space. The flame usually has obvious features in the U (chroma) and V (saturation) channels of the YUV space. Then, the color threshold segmentation is applied to extract the potential flame area. The morphological operation (such as expansion, corrosion, opening operation, closing operation) is applied to the flame area to determine whether there is a flame feature. The dynamic changes of the flame area, such as flickering and diffusion, are analyzed to further confirm the existence of the flame. Then, the background subtraction algorithm (such as Gaussian mixture model GMM) is used to separate the foreground object from the video frame, and the person is identified by the human body detection algorithm (such as HOG+SVM, OpenPose) to determine whether there is a person gathering feature. Specifically, the density estimation algorithm (such as density map estimation based on convolutional neural network) is applied to the detected foreground area to calculate the number of people in the unit area. Finally, the motion trajectory and behavior characteristics of the person, such as speed, direction, acceleration, etc., are extracted based on the optical flow method. The behavior characteristics are compared with the normal mode and the behavior similarity is calculated. If the behavior similarity is too low, there is an abnormal behavior feature.

[0025] As a preferred embodiment of the present invention, the steps of extracting the motion trajectory and behavior characteristics of a person based on the optical flow method, comparing the behavior characteristics with the normal mode, calculating the behavior similarity, and obtaining the abnormal behavior characteristics specifically include: Obtain historical normal behavior data, and map the historical normal behavior data to a low-dimensional space using principal component analysis, calculate the average eigenvalue of the historical normal behavior and the covariance matrix of the characteristic distribution, thereby determining the distribution area of ​​the historical normal behavior in the embedding space, and constructing a characteristic distribution model of normal behavior in the low-dimensional space; Analyze moving objects in video frame sequences through optical flow method and extract motion trajectory data Based on the motion trajectory data, calculate the speed, acceleration, trajectory curvature, spatiotemporal consistency and energy consumption value at each moment; The speed, acceleration, trajectory curvature, spatiotemporal consistency and energy consumption value at each moment are combined into a high-dimensional vector, and the high-dimensional vector is mapped to a low-dimensional space using principal component analysis to obtain several behavioral feature points; Calculate the distance between the behavior feature point and the center of the region of the feature distribution model of normal behavior to obtain the embedding space distance; Calculate the probability density value of the behavior feature point in the feature distribution model of normal behavior to determine the abnormal probability; According to the importance between the embedding space distance, anomaly probability, speed, spatiotemporal consistency and energy consumption value, the corresponding weights are assigned; According to the dynamics of historical data, use genetic algorithms to optimize each weight and find the optimal weight combination; The embedding space distance, anomaly probability, standard deviation of speed, spatiotemporal consistency and energy consumption value are weighted and summed with the optimal weight combination to obtain a comprehensive anomaly score; Based on the comprehensive abnormality score, behavioral abnormality characteristics were determined.

[0026] In the embodiment of the present invention, a variety of behavioral characteristics are combined, including multiple dimensions such as motion speed, acceleration, trajectory curvature, spatiotemporal consistency and energy consumption, which fully reflects the dynamic changes and complexity of the behavior and covers the static and dynamic characteristics of the behavior. Some more subtle behavioral anomalies can be detected. For example, a person appears to be walking normally, but due to the inconsistency between the direction and the crowd trend, it may be marked as potential abnormal behavior. In addition, by using the embedded learning method, high-dimensional features can be mapped to a simpler embedding space while retaining key behavioral characteristic information. Embedded learning improves the model's ability to distinguish complex and nonlinear behaviors, making the detection of abnormal behaviors more accurate. Finally, a genetic algorithm is introduced to optimize the scoring weights, which can dynamically adjust the impact of different features on the abnormal scoring results. This optimization mechanism ensures that the model can maintain good adaptability and accuracy in different scenarios and different types of anomalies.

[0027] like Figure 4 As shown, as a preferred embodiment of the present invention, the step of generating security personnel mobilization information based on fire monitoring information specifically includes: S301, determining the hazardous items and hazard levels of each on-site environmental area according to the threshold exceeding ratio, the degree of personnel gathering, the abnormal behavior type and the flame characteristics; S302, determining the number of personnel required for each on-site environment area based on the dangerous items and danger levels, and retrieving the security personnel's duty information; S303, generating security personnel mobilization information according to the security personnel duty information, the required number of personnel, dangerous items and danger levels.

[0028] In the embodiment of the present invention, the dangerous items and danger levels of each on-site environment area are determined based on the threshold exceeding ratio, the degree of personnel gathering, the type of abnormal behavior and the flame characteristics. It is easy to understand that the higher the threshold exceeding ratio and the higher the degree of personnel gathering, the higher the corresponding danger level. When the flame characteristics exist, the highest danger level is set by default. Then, the required number of personnel for each on-site environment area is determined based on the dangerous items and danger levels. The higher the danger level, the more personnel are required for the corresponding area, and the security personnel duty information is retrieved. The security personnel duty information includes the geographical location, availability and professional expertise of the security personnel. Finally, the security personnel mobilization information is generated based on the security personnel duty information, the required number of personnel, dangerous items and danger levels.

[0029] As a preferred embodiment of the present invention, the step of determining the required number of personnel for each on-site environment area based on the dangerous items and the dangerous levels, and retrieving the duty information of the security personnel specifically includes: Each hazardous project is divided into three levels: low risk, medium risk and high risk according to the segment range. The detection value of each dangerous item is obtained, and according to the segment range of the detection value of each dangerous item, the segment scoring rule is applied to perform dynamic adjustment to obtain the adjusted score value, wherein the segment scoring rule specifically includes: If the test value is in the low-risk range, the score is zero; If the test value is in the medium risk range, the score is calculated according to a linear growth rule; If the detection value is in the high-risk range, an additional offset is introduced as a penalty based on the medium-risk scoring rule; The actual area value of each on-site environment area is obtained, and the area size of the on-site environment area is converted into an area size weight in a smooth growth manner; Calculate the time variation trend of the threshold exceeding ratio within the time interval, perform fuzzy processing on the time variation trend, and obtain the time variation weight; According to the size of the area weight and the time change weight, fuzzy logic is used to combine them to obtain the fuzzy logic weight; Then the adjusted score value is adjusted using the logical fuzzy weight to obtain the score value of each item; The score values ​​of each project are summarized, and the summary result is used as the required number of personnel in the on-site environment area, and the security personnel’s duty information is retrieved based on the required number of personnel in the on-site environment area.

[0030] Among them, according to the size of the area weight and the time change weight, the fuzzy logic method is used to combine them as follows: If the area is small and the temporal variation is low, the fuzzy logic weight should be low; If the area is small but the temporal variation is high, the fuzzy logic weight should be medium; If the area is large and the temporal variation is low, the fuzzy logic weight should be medium; If the area is large and the temporal variation is high, the fuzzy logic weight should be high; In the embodiments of the present invention, multiple factors such as the diversity of dangerous projects, danger levels, regional areas, and time changes are fully considered. By combining static and dynamic factors through fuzzy logic weights and dynamic changes over time, the complex situation on site can be more realistically reflected. The larger the area and the more significant the cumulative changes in risk over time, the greater the demand for deployment personnel. This dynamic adjustment method is very consistent with the actual scene characteristics in security work.

[0031] like Figure 5 As shown, as a preferred embodiment of the present invention, the step of deeply analyzing the fire monitoring information to determine whether there is a potential safety hazard specifically includes: S401, drawing a change trend graph for each area based on fire monitoring information, wherein a change trend graph includes a temperature change curve, a humidity change curve, a smoke concentration change curve, and a personnel gathering change curve; S402, slicing the change trend graph to obtain a plurality of trend graph segments; S403, input each trend graph segment into the potential safety hazard curve library for matching, and when the matching value is higher than a preset similarity, it is determined that there is a potential safety hazard.

[0032] In an embodiment of the present invention, in order to conduct a more in-depth and detailed analysis, a change trend graph of each area is drawn according to the fire monitoring information. A change trend graph includes a temperature change curve, a humidity change curve, a smoke concentration change curve and a personnel gathering change curve, each curve has a corresponding color, and then the change trend graph is sliced ​​to obtain a number of trend graph segments. The length of each trend graph segment is the same as the length of the hazard curve in the safety hazard curve library. The safety hazard curve library contains a large number of hazard curves. Each trend graph segment is input into the safety hazard curve library for matching, and the similarity between the trend graph segment and each hazard curve is calculated, and the highest similarity is determined as the matching value. When the matching value is higher than the preset similarity, it means that it is very similar to the corresponding hazard curve, and it is determined that there is a safety hazard.

[0033] As a preferred embodiment of the present invention, the step of inputting each trend graph segment into the potential safety hazard curve library for matching, and determining that a potential safety hazard exists when the matching value is higher than a preset similarity, specifically includes: S4031, input each trend graph segment into the safety hazard curve library; S4032, decomposing the current trend graph segment and the hidden danger curve into multiple dimensions; S4033, interpolating the trend graph segments and hidden danger curves of multiple dimensions on the time axis to align the time points; S4034, using the DTW algorithm to calculate the trend pattern similarity between the trend image segment and the hidden danger curve in the corresponding dimension, obtaining similarity values ​​of different dimensions, and performing a normalization operation to obtain a normalized similarity value; S4035, calculating the distribution characteristics of each dimension of the trend graph segment and the hidden danger curve; S4036, calculating the difference in distribution characteristics of each dimension of the trend graph segment and the hidden danger curve to obtain a statistical distribution difference, and then converting the statistical distribution difference into a penalty factor; S4037, multiplying the normalized similarity value of the current dimension by the corresponding penalty factor to obtain a final matching score of the current dimension; S4038, assigning dimension weights according to the importance of each dimension, performing weighted summation on the final matching score of each dimension and the corresponding dimension weight, and then normalizing the weighted summation result by the sum of the dimension weights to obtain the final matching value of the current segment; S4039, repeat the operations from S4032 to S4037 for all trend image segments to obtain final matching values ​​of all trend images, take the highest value among the final matching values ​​of all trend images and match it with the preset similarity, and confirm whether there is a safety hazard based on the matching result.

[0034] Among them, the preset similarity of this embodiment is 1. If the final matching value is close to 1, it means that the pattern and distribution of the trend graph segment and the hidden danger curve are highly similar, and it can be determined that there are potential safety hazards in the area.

[0035] If the final matching value is low, it means that their similarity is poor, and it can be determined that there are no obvious hidden dangers in the area for the time being. In an embodiment of the present invention, the current trend graph segment and the hidden danger curve are decomposed into multiple dimensions, and data of multiple dimensions such as temperature, humidity, smoke concentration, and personnel gathering are considered at the same time, which can more comprehensively reflect the complex situation on the scene. Among them, DTW can dynamically adjust the different time scales of the trend graph segment and the hidden danger curve, so that the system can adapt to the scene with more prominent time changes, and filter out some non-hidden danger area data that are similar in appearance but different in essence through the penalty mechanism of statistical distribution. By introducing a dual matching judgment mechanism, the matching judgment is made by using the trend pattern similarity and statistical distribution difference at the same time, so that it can not only pay attention to the pattern similarity of the trend graph segment and the hidden danger curve, but also consider the overall statistical distribution characteristics of the two, which can effectively reduce misjudgment.

[0036] like Figure 6As shown, an embodiment of the present invention further provides a fire monitoring system based on obtaining on-site environmental information, the system comprising: The environmental information acquisition module 100 is used to acquire on-site environmental information based on sensors and cameras, wherein the on-site environmental information includes temperature information, humidity information, smoke concentration information, and surveillance video, and each on-site environmental information is marked with a timestamp and a location stamp; The monitoring information determination module 200 is used to perform a preliminary analysis on the on-site environment information to determine the on-site environment area that needs to be monitored, and obtain fire monitoring information; The security personnel mobilization module 300 is used to generate security personnel mobilization information based on fire monitoring information. The security personnel mobilization information includes a number of personnel numbers, each of which corresponds to a responsible area and a responsible project; The fire protection in-depth analysis module 400 is used to deeply analyze the fire protection monitoring information to determine whether there are any safety hazards. When there are safety hazards, fire protection warning information is generated and sent to the corresponding security personnel account.

[0037] As a preferred embodiment of the present invention, the monitoring information determination module 200 includes: The temperature and humidity concentration comparison unit is used to compare the temperature information, humidity information and smoke concentration information in each on-site environment information with the corresponding temperature threshold, humidity threshold and smoke concentration threshold, and determine the corresponding location stamp as the on-site environment area that needs to be monitored; A monitoring video analysis unit is used to perform feature analysis on the monitoring video in each on-site environment information to determine whether there are flame features, personnel gathering features or abnormal behavior features. If so, the corresponding location stamp is determined as the on-site environment area that needs to be monitored; The environmental information aggregation unit is used to aggregate the on-site environmental information of the on-site environmental area that needs to be monitored to obtain fire monitoring information.

[0038] As a preferred embodiment of the present invention, the monitoring video analysis unit includes: The flame feature determination subunit is used to convert the monitoring video from the RGB color space to the YUV color space, apply color threshold segmentation, extract potential flame areas, and apply morphological operations to the flame areas to determine the flame features; The clustering feature determination subunit is used to separate the foreground object from the video frame using the background subtraction algorithm, identify the person through the human body detection algorithm, and obtain the person clustering feature; The behavior abnormality determination subunit is used to extract the movement trajectory and behavior characteristics of the personnel based on the optical flow method, compare the behavior characteristics with the normal mode, calculate the behavior similarity, and obtain the behavior abnormality characteristics.

[0039] As a preferred embodiment of the present invention, the security personnel mobilization module 300 includes: Dangerous item level unit, used to determine the dangerous items and danger levels of each on-site environmental area based on the threshold exceeding ratio, the degree of personnel gathering, the type of abnormal behavior and the characteristics of flames; Security personnel analysis unit, used to determine the number of personnel required for each on-site environment area based on dangerous items and danger levels, and retrieve security personnel duty information; The personnel transfer information unit is used to generate security personnel transfer information based on security personnel duty information, required number of personnel, dangerous items and danger levels.

[0040] As a preferred embodiment of the present invention, the fire protection depth analysis module 400 includes: A change trend graph drawing unit is used to draw a change trend graph of each area based on fire monitoring information. A change trend graph includes a temperature change curve, a humidity change curve, a smoke concentration change curve, and a personnel gathering change curve; A slicing processing unit, used for slicing the change trend graph to obtain a plurality of trend graph segments; The potential safety hazard matching unit is used to input each trend graph segment into the potential safety hazard curve library for matching. When the matching value is higher than a preset similarity, it is determined that a potential safety hazard exists.

[0041] The above only describes in detail the preferred embodiments of the present invention, which is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

[0042] It should be understood that, although each step in the flow chart of each embodiment of the present invention is shown in sequence according to the indication of the arrow, these steps are not necessarily performed in sequence according to the order indicated by the arrow. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be performed in other orders. Moreover, at least a portion of the steps in each embodiment may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of other steps or sub-steps or stages of other steps.

[0043] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0044] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the disclosure in the specification and examples. This application is intended to cover any variations, uses or adaptations of the present disclosure, which follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present disclosure are indicated by the claims.

Claims

1. A fire monitoring method based on obtaining on-site environmental information, characterized in that: The method comprises the following steps: Based on the sensors and cameras, the on-site environment information is obtained, the on-site environment information includes temperature information, humidity information, smoke concentration information and monitoring video, and each on-site environment information is marked with a time stamp and a location stamp; Conduct a preliminary analysis of the on-site environmental information to determine the on-site environmental areas that need to be monitored and obtain fire monitoring information; Generate security personnel mobilization information based on fire monitoring information. The security personnel mobilization information includes several personnel numbers, each of which corresponds to a responsible area and a responsible project; In-depth analysis of fire monitoring information to determine whether there are safety hazards. If there are safety hazards, fire warning information is generated and sent to the corresponding security personnel account; The step of performing a preliminary analysis on the on-site environmental information to determine the on-site environmental area that needs to be monitored and obtaining fire monitoring information specifically includes: Compare the temperature information, humidity information and smoke concentration information in each on-site environment information with the corresponding temperature threshold, humidity threshold and smoke concentration threshold, and determine the corresponding location stamp as the on-site environment area that needs to be monitored; Perform feature analysis on the surveillance video in each on-site environment information to determine whether there are flame features, crowd gathering features, or abnormal behavior features. If so, determine the corresponding location stamp as the on-site environment area that needs to be monitored; The on-site environmental information of the on-site environmental area that needs to be monitored is summarized to obtain fire monitoring information.

2. The fire monitoring method based on on-site environmental information acquisition according to claim 1 is characterized in that: The step of comparing the temperature information, humidity information and smoke concentration information in each on-site environment information with the corresponding temperature threshold, humidity threshold and smoke concentration threshold to determine that the corresponding location stamp is the on-site environment area to be monitored specifically includes the following sub-steps: Obtain the temperature information, humidity information and smoke concentration information of each on-site environment information, According to the change trend of temperature information, humidity information and smoke concentration information within a preset time period, the change rate of temperature, humidity and smoke concentration is obtained; According to the changing rates of temperature, humidity and smoke concentration, the corresponding temperature dynamic weight, humidity dynamic weight and smoke dynamic weight are given; Get temperature threshold, humidity threshold and smoke concentration threshold; The ratio between the temperature information and the temperature threshold is used as the temperature contribution factor, the ratio between the smoke concentration information and the smoke concentration threshold is used as the smoke contribution factor, and the inverse of the ratio between the humidity information and the humidity threshold is used as the humidity sharing factor; The temperature contribution factor, smoke contribution factor and humidity sharing factor are fused by multiplication to capture the interaction effect between each variable and obtain a comprehensive interaction factor. The temperature dynamic weight, humidity dynamic weight and smoke dynamic weight are combined in an additive manner to obtain a comprehensive dynamic weight, and then the comprehensive dynamic weight is added to the comprehensive interaction factor to calculate a preliminary score; The preliminary score was amplified by applying an exponential function to obtain a comprehensive risk score; The fuzzy logic method is used to divide the hazard level, and the comprehensive hazard score is compared with the hazard level to determine whether the current area is a field environment area that needs to be monitored.

3. The fire monitoring method based on on-site environmental information acquisition according to claim 2 is characterized in that: The step of performing feature analysis on the surveillance video in each on-site environment information specifically includes: Convert the surveillance video from RGB color space to YUV color space, apply color threshold segmentation to extract potential flame areas, and apply morphological operations to the flame areas to determine the flame features; Use the background subtraction algorithm to separate the foreground object from the video frame, and use the human detection algorithm to identify people and obtain the human aggregation features; The motion trajectory and behavior characteristics of the personnel are extracted based on the optical flow method, the behavior characteristics are compared with the normal mode, the behavior similarity is calculated, and the abnormal behavior characteristics are obtained.

4. The fire monitoring method based on on-site environmental information acquisition according to claim 3 is characterized in that: The steps of extracting the motion trajectory and behavior characteristics of the person based on the optical flow method, comparing the behavior characteristics with the normal mode, calculating the behavior similarity, and obtaining the abnormal behavior characteristics specifically include: Obtain historical normal behavior data, and map the historical normal behavior data to a low-dimensional space using principal component analysis, calculate the average eigenvalue of the historical normal behavior and the covariance matrix of the characteristic distribution, thereby determining the distribution area of ​​the historical normal behavior in the embedding space, and constructing a characteristic distribution model of normal behavior in the low-dimensional space; Analyze the moving objects in the video frame sequence by optical flow method and extract the motion trajectory data; Based on the motion trajectory data, calculate the speed, acceleration, trajectory curvature, spatiotemporal consistency and energy consumption value at each moment; The speed, acceleration, trajectory curvature, spatiotemporal consistency and energy consumption value at each moment are combined into a high-dimensional vector, and the high-dimensional vector is mapped to a low-dimensional space using principal component analysis to obtain several behavioral feature points; Calculate the distance between the behavior feature point and the center of the region of the feature distribution model of normal behavior to obtain the embedding space distance; Calculate the probability density value of the behavior feature point in the feature distribution model of normal behavior to determine the abnormal probability; According to the importance between the embedding space distance, anomaly probability, speed, spatiotemporal consistency and energy consumption value, the corresponding weights are assigned; According to the dynamics of historical data, use genetic algorithms to optimize each weight and find the optimal weight combination; The embedding space distance, anomaly probability, standard deviation of speed, spatiotemporal consistency and energy consumption value are weighted and summed with the optimal weight combination to obtain a comprehensive anomaly score; Based on the comprehensive abnormality score, behavioral abnormality characteristics were determined.

5. The fire monitoring method based on on-site environmental information acquisition according to claim 4 is characterized in that: The step of generating security personnel mobilization information based on fire monitoring information specifically includes: Determine the hazardous items and hazard levels of each on-site environmental area based on the threshold exceeding ratio, the degree of personnel gathering, the type of abnormal behavior and the characteristics of the flame; Determine the number of personnel required for each on-site environment area based on the hazardous items and hazard levels, and retrieve the security personnel's duty information; Generate security personnel mobilization information based on security personnel duty information, required number of personnel, dangerous items and danger levels.

6. The fire monitoring method based on on-site environmental information acquisition according to claim 5 is characterized in that: The step of determining the number of personnel required for each on-site environment area based on the dangerous items and the dangerous levels and retrieving the security personnel's duty information specifically includes the following sub-steps: Divide each hazardous project into three levels: low risk, medium risk and high risk according to the segment range; The detection value of each dangerous item is obtained, and according to the segment range of the detection value of each dangerous item, the segment scoring rule is applied to perform dynamic adjustment to obtain the adjusted score value, wherein the segment scoring rule specifically includes: If the test value is in the low-risk range, the score is zero; If the test value is in the medium risk range, the score is calculated according to a linear growth rule; If the detection value is in the high-risk range, an additional offset is introduced as a penalty based on the medium-risk scoring rule; The actual area value of each on-site environment area is obtained, and the area size of the on-site environment area is converted into an area size weight in a smooth growth manner; Calculate the time variation trend of the threshold exceeding ratio within the time interval, perform fuzzy processing on the time variation trend, and obtain the time variation weight; According to the size of the area weight and the time change weight, fuzzy logic is used to combine them to obtain the fuzzy logic weight; Then the adjusted score value is adjusted using the logical fuzzy weight to obtain the score value of each item; The score values ​​of each project are summarized, and the summary result is used as the required number of personnel in the on-site environment area, and the security personnel’s duty information is retrieved based on the required number of personnel in the on-site environment area.

7. The fire monitoring method based on on-site environmental information acquisition according to claim 6 is characterized in that: The steps of deeply analyzing the fire monitoring information to determine whether there are potential safety hazards specifically include: Draw a change trend graph for each area based on fire monitoring information. A change trend graph includes temperature change curve, humidity change curve, smoke concentration change curve and personnel gathering change curve; Slice the change trend graph to obtain several trend graph segments; Each trend graph segment is input into the safety hazard curve library for matching. When the matching value is higher than the preset similarity, it is determined that there is a safety hazard.

8. The fire monitoring method based on obtaining on-site environmental information according to claim 7 is characterized in that: The step of inputting each trend graph segment into the potential safety hazard curve library for matching, and determining that a potential safety hazard exists when the matching value is higher than a preset similarity, specifically includes the following sub-steps: S4031, input each trend graph segment into the safety hazard curve library; S4032, decomposing the current trend graph segment and the hidden danger curve into multiple dimensions; S4033, interpolating the trend graph segments and hidden danger curves of multiple dimensions on the time axis to align the time points; S4034, using the DTW algorithm to calculate the trend pattern similarity between the trend image segment and the hidden danger curve in the corresponding dimension, obtaining similarity values ​​of different dimensions, and performing a normalization operation to obtain a normalized similarity value; S4035, calculating the distribution characteristics of each dimension of the trend graph segment and the hidden danger curve; S4036, calculating the difference in distribution characteristics of each dimension of the trend graph segment and the hidden danger curve to obtain a statistical distribution difference, and then converting the statistical distribution difference into a penalty factor; S4037, multiplying the normalized similarity value of the current dimension by the corresponding penalty factor to obtain a final matching score of the current dimension; S4038, assigning dimension weights according to the importance of each dimension, performing weighted summation on the final matching score of each dimension and the corresponding dimension weight, and then normalizing the weighted summation result by the sum of the dimension weights to obtain the final matching value of the current segment; S4039, repeat the operations from S4032 to S4037 for all trend image segments to obtain final matching values ​​of all trend images, take the highest value among the final matching values ​​of all trend images and match it with the preset similarity, and confirm whether there is a safety hazard based on the matching result.

9. A fire monitoring system based on on-site environmental information acquisition, the system being applied to the fire monitoring method based on on-site environmental information acquisition according to any one of claims 1 to 8, characterized in that: The system comprises: An environmental information collection module is used to collect on-site environmental information based on sensors and cameras. The on-site environmental information includes temperature information, humidity information, smoke concentration information, and surveillance video. Each on-site environmental information is marked with a timestamp and a location stamp. The monitoring information determination module is used to perform preliminary analysis on the on-site environmental information to determine the on-site environmental area that needs to be monitored and obtain fire monitoring information; The security personnel mobilization module is used to generate security personnel mobilization information based on fire monitoring information. The security personnel mobilization information includes several personnel numbers, each of which corresponds to a responsible area and a responsible project; The fire protection in-depth analysis module is used to conduct in-depth analysis of fire protection monitoring information to determine whether there are any safety hazards. When there are safety hazards, fire protection warning information is generated and sent to the corresponding security personnel account.

10. The fire monitoring system based on on-site environmental information acquisition according to claim 9, characterized in that: The monitoring information determination module includes: The temperature and humidity concentration comparison unit is used to compare the temperature information, humidity information and smoke concentration information in each on-site environment information with the corresponding temperature threshold, humidity threshold and smoke concentration threshold, and determine the corresponding location stamp as the on-site environment area that needs to be monitored; A monitoring video analysis unit is used to perform feature analysis on the monitoring video in each on-site environment information to determine whether there are flame features, personnel gathering features or abnormal behavior features. If so, the corresponding location stamp is determined as the on-site environment area that needs to be monitored; An environmental information aggregation unit is used to aggregate the on-site environmental information of the on-site environmental area that needs to be monitored to obtain fire monitoring information; The monitoring video analysis unit comprises: The flame feature determination subunit is used to convert the monitoring video from the RGB color space to the YUV color space, apply color threshold segmentation, extract the potential flame area, and apply morphological operations to the flame area to determine the flame features; The clustering feature determination subunit is used to separate the foreground object from the video frame using the background subtraction algorithm, identify the person through the human body detection algorithm, and obtain the person clustering feature; The behavior abnormality determination subunit is used to extract the movement trajectory and behavior characteristics of the person based on the optical flow method, compare the behavior characteristics with the normal mode, calculate the behavior similarity, and obtain the behavior abnormality characteristics; The security personnel mobilization module includes: Dangerous item level unit, used to determine the dangerous items and danger levels of each on-site environmental area based on the threshold exceeding ratio, the degree of personnel gathering, the type of abnormal behavior and the characteristics of flames; Security personnel analysis unit, used to determine the number of personnel required for each on-site environment area based on dangerous items and danger levels, and retrieve security personnel duty information; The personnel transfer information unit is used to generate security personnel transfer information based on security personnel duty information, required number of personnel, dangerous items and danger levels.

Images