A disaster emergency alarm method and alarm device based on artificial intelligence

Abstract

The present application provides a disaster emergency alarm method and alarm device based on artificial intelligence, which monitors the target area for fire, automatically collects temperature monitoring data, and divides the temperature monitoring data into multiple temperature data samples; determines the fire risk associated information of the time node based on all temperature data samples, and determines the dominant temperature based on different fire risk associated information and combustion load; automatically collects flame monitoring data, extracts multiple flame hotspots from the flame monitoring data, determines the flame front area based on all flame hotspots, and obtains diffusion characteristics based on the flame front area extraction; determines the thermal responsiveness based on the dominant temperature and diffusion characteristics, and generates a fire risk signal based on the thermal responsiveness; and finally performs a fire disaster emergency alarm based on the fire risk signal. The above scheme is based on the fire risk signal, and can realize the identification of the fire situation in the target area under multi-dimensional interference factors, thereby improving the accuracy of disaster emergency alarms.

Description

Technical Field

[0001] The present application relates to the field of intelligent disaster alarm technology, and more specifically, to a disaster emergency alarm method and alarm device based on artificial intelligence. Background Art

[0002] In modern society, rapid detection and response to disasters such as fires are crucial. Traditional disaster alarms rely on simple sensors, which have problems such as response delays and high false alarm rates. With the advancement and development of science and technology, disaster alarm technology based on artificial intelligence has significantly improved the accuracy and response speed of disaster detection. At the same time, artificial intelligence algorithms can distinguish between real disaster signals and false alarms through big data analysis and pattern recognition. For example, combined with historical disaster data and environmental parameters, the alarm device can adaptively adjust the alarm threshold to reduce the false alarm rate. Through real-time data processing and prediction, artificial intelligence disaster alarm technology can not only detect disasters in the early stages, but also provide predictive information on the spread of disasters to help formulate effective emergency response strategies. Artificial intelligence disaster alarm technology uses advanced data analysis and machine learning technologies to achieve efficient and accurate disaster detection and response, significantly improving the level of public safety.

[0003] In the prior art, intelligent disaster alarms usually use a variety of technologies to achieve efficient disaster detection and early warning. First, environmental data is collected in real time using sensor networks. These data are transmitted to the central processing unit through the Internet of Things technology. Then, artificial intelligence algorithms, especially machine learning models, are used to analyze these data to identify abnormal patterns or early signs of disasters. For example, in fire disaster monitoring, artificial intelligence disaster alarm technology integrates a variety of sensor data, including temperature, smoke, pressure, audio and video, to monitor environmental changes in real time, and identifies potential fire signals and spread conditions with the help of machine learning and deep learning algorithms; however, in existing fire disaster alarms, there are usually multi-dimensional interference factors at the fire scene, such as inaccurate fire burning characteristics recognized by sensors due to direct sunlight and complex fire spread in the target area under the influence of factors such as wind speed, wind direction, type of combustibles, and humidity. These multi-dimensional interference factors will make it difficult to accurately identify the fire situation at the fire scene, resulting in delays or even errors in alarm information, resulting in fire out of control and property losses. Therefore, how to identify the fire situation in the target area under multi-dimensional interference factors, thereby improving the accuracy of disaster emergency alarms, has become a difficult problem faced by the industry. Summary of the invention

[0004] The present application provides a disaster emergency alarm method and alarm device based on artificial intelligence, which can identify the fire situation in the target area under multi-dimensional interference factors, thereby improving the accuracy of disaster emergency alarms.

[0005] In a first aspect, the present application provides a disaster emergency alarm method based on artificial intelligence, comprising the following steps:

[0006] Conduct fire monitoring in the target area and automatically collect temperature monitoring data in the target area;

[0007] The collected temperature monitoring data is divided into multiple temperature data samples, and the fire risk association information at each time node when a fire occurs in the target area is determined based on all the temperature data samples, and then the dominant temperature when a fire occurs in the target area is determined according to the fire risk association information at different time nodes and the combustion load when the target area is burning;

[0008] Automatically collect flame monitoring data of the target area, extract multiple flame hot spots of the target area when burning from the flame monitoring data, and then determine the flame front area when the fire is burning based on all the flame hot spots, and extract the diffusion characteristics of the flame front area when burning;

[0009] Determining the thermal responsiveness of the fire during the combustion and diffusion according to the dominant temperature and the diffusion characteristics, and then generating a fire risk signal of the target area according to the thermal responsiveness;

[0010] Performing a fire disaster emergency alarm on the target area based on the fire risk signal;

[0011] Among them, the dominant temperature when a fire occurs in the target area is determined based on the fire risk association information at different time nodes and the combustion load when the target area is burning, which specifically includes:

[0012] Obtain the combustion load when the target area is burning;

[0013] Construct a fire risk correlation entropy matrix based on the fire risk correlation information at different time nodes;

[0014] Determine all fire load factors and all fire load response vectors of the target area when it is subjected to fire combustion based on the fire risk correlation entropy matrix;

[0015] Determine the combustion response characteristics of the target area when it is subjected to fire combustion according to all fire load factors and all fire load response vectors;

[0016] A dominant temperature when a fire occurs in a target area is determined based on the combustion response characteristics and the combustion load.

[0017] In some embodiments, dividing the collected temperature monitoring data into a plurality of temperature data samples specifically includes:

[0018] Get all temperature measurement paths of the temperature measurement equipment;

[0019] For each temperature measurement path, the temperature data corresponding to the temperature measurement path in the temperature monitoring data is divided into a temperature data sample, so as to obtain the temperature data sample of each temperature measurement path.

[0020] In some embodiments, determining the fire risk association information at each time point when a fire occurs in the target area based on all temperature data samples specifically includes:

[0021] Obtain temperature data samples corresponding to each temperature measurement path;

[0022] Extract multiple temperature correlation values ​​at each time node from all temperature data samples, and then obtain the temperature correlation sequence corresponding to each time node;

[0023] For each time node, the fire risk association information at the time node is determined according to the temperature association sequence corresponding to the time node and the temperature association sequences corresponding to the other time nodes, and then the fire risk association information at each time node when a fire occurs in the target area is obtained.

[0024] In some embodiments, extracting multiple flame hot spots of the target area during combustion from the flame monitoring data specifically includes:

[0025] Acquire a flame monitoring image at a current moment from the flame monitoring data;

[0026] Performing image enhancement on the flame monitoring image to obtain a flame monitoring enhanced image;

[0027] Extracting flame characteristic values ​​from the flame monitoring enhanced image;

[0028] Combustion heat analysis is performed on all pixel points in the flame monitoring enhanced image based on the flame characteristic value, thereby obtaining a plurality of flame hot spots in the target area when burning.

[0029] In some embodiments, extracting the diffusion characteristics of the flame front region during combustion specifically includes:

[0030] determining combustion geometry information of the flame front region;

[0031] The diffusion characteristics of the flame front region during combustion are extracted from the combustion geometry information.

[0032] In some embodiments, the flame monitoring data includes flame monitoring images of the fire scene taken at different time points.

[0033] In a second aspect, the present application provides a disaster emergency alarm device based on artificial intelligence, which includes:

[0034] An acquisition module is used to monitor the target area for fire and automatically collect temperature monitoring data of the target area;

[0035] A processing module, for dividing the collected temperature monitoring data into a plurality of temperature data samples, determining the fire risk association information at each time node when a fire occurs in the target area based on all the temperature data samples, and then determining the dominant temperature when a fire occurs in the target area according to the fire risk association information at different time nodes and the combustion load when the target area burns;

[0036] The processing module is also used to automatically collect flame monitoring data of the target area, extract multiple flame hotspots of the target area when burning from the flame monitoring data, and then determine the flame front area when the fire is burning according to all the flame hotspots, and extract the diffusion characteristics of the flame front area when burning;

[0037] The processing module is further used to determine the thermal responsiveness of the fire during combustion and diffusion according to the dominant temperature and the diffusion characteristics, and then generate a fire risk signal of the target area according to the thermal responsiveness;

[0038] The execution module is used to perform a fire disaster emergency alarm on the target area based on the fire risk signal.

[0039] In a third aspect, the present application provides a computer device, comprising a memory and a processor, wherein the memory stores code, and the processor is configured to obtain the code and execute the above-mentioned artificial intelligence-based disaster emergency alarm method.

[0040] In a fourth aspect, the present application provides a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, it implements the above-mentioned artificial intelligence-based disaster emergency alarm method.

[0041] The technical solution provided by the embodiments disclosed in this application has the following beneficial effects:

[0042] In the artificial intelligence-based disaster emergency alarm method and alarm device provided by the present application, first, fire monitoring is performed on the target area, and temperature monitoring data of the target area is automatically collected; further, the collected temperature monitoring data is divided into multiple temperature data samples, and the fire risk association information at each time node when a fire occurs in the target area is determined based on all the temperature data samples, and then the dominant temperature when a fire occurs in the target area is determined based on the fire risk association information at different time nodes and the combustion load when the target area is burning; then, flame monitoring data of the target area is automatically collected, and multiple flame hotspots of the target area when burning are extracted from the flame monitoring data, and then the flame front area when the fire is burning is determined based on all the flame hotspots, and the diffusion characteristics of the flame front area when burning are extracted; secondly, the thermal responsiveness of the fire when burning and spreading is determined based on the dominant temperature and the diffusion characteristics, and then a fire risk signal of the target area is generated based on the thermal responsiveness; finally, a fire disaster emergency alarm is performed on the target area based on the fire risk signal.

[0043] It can be seen that the present application can realize the identification of fire conditions in the target area under multi-dimensional interference factors, thereby improving the accuracy of disaster emergency alarms; first, the collected temperature monitoring data is divided into multiple temperature data samples to evaluate the temperature changes in the target area and identify abnormal combustion conditions during fire burning, and then adjust the acquisition accuracy of the acquisition sensor according to the abnormal combustion conditions, thereby avoiding the problem of sensors identifying inaccurate fire combustion characteristics due to direct sunlight; secondly, the target area is analyzed based on the fire risk association information determined by the temperature data samples to obtain the dominant temperature when a fire occurs in the target area, so as to effectively analyze the spread of the fire and understand the combustion risk of the target area, thereby formulating an effective fire fighting strategy. , thereby avoiding the monitoring error caused by the complex fire spread; further, the fire spread analysis is carried out on the fire scene according to the flame front area determined by the most vigorous burning point in the target area, and the diffusion characteristics of the target area are obtained, and then the risk characteristics of the flame spreading from the target area to the surrounding areas are identified, so as to formulate effective emergency response and fire extinguishing strategies; then, the thermal responsiveness determined based on the dominant temperature and diffusion characteristics is used to characterize the spread of the fire in the target area; finally, the fire disaster emergency alarm is carried out in the target area according to the fire risk signal determined by the thermal responsiveness; in summary, the technical solution provided by the present application can realize the identification of the fire in the target area under multi-dimensional interference factors, thereby improving the accuracy of disaster emergency alarms. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 is an exemplary flow chart of a disaster emergency warning method based on artificial intelligence according to some embodiments of the present application;

[0045] Figure 2is an exemplary flow chart of determining a dominant temperature according to some embodiments of the present application;

[0046] Figure 3 is an exemplary flow chart of determining multiple flame hot spots according to some embodiments of the present application;

[0047] Figure 4 is a schematic diagram of exemplary hardware and / or software of an artificial intelligence-based disaster emergency alarm device according to some embodiments of the present application;

[0048] Figure 5 It is a structural diagram of a computer device for implementing an artificial intelligence-based disaster emergency alarm method according to some embodiments of the present application. DETAILED DESCRIPTION

[0049] The core of this application is to first monitor the target area for fire and automatically collect temperature monitoring data of the target area; further, divide the collected temperature monitoring data into multiple temperature data samples, determine the fire risk association information at each time node when a fire occurs in the target area based on all temperature data samples, and then determine the dominant temperature when a fire occurs in the target area based on the fire risk association information at different time nodes and the combustion load when the target area burns; then, automatically collect the flame monitoring data of the target area, extract multiple flame hotspots of the target area when burning from the flame monitoring data, and then determine the flame front area when the fire burns based on all flame hotspots, and extract the diffusion characteristics of the flame front area when burning; secondly, determine the thermal responsiveness of the fire when burning and spreading based on the dominant temperature and the diffusion characteristics, and then generate a fire risk signal for the target area based on the thermal responsiveness; finally, perform a fire disaster emergency alarm on the target area based on the fire risk signal. The above scheme is based on the fire risk signal, which can realize the identification of the fire situation in the target area under multi-dimensional interference factors, thereby improving the accuracy of disaster emergency alarm.

[0050] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods. Figure 1 , which is an exemplary flow chart of a disaster emergency alarm method based on artificial intelligence according to some embodiments of the present application. The disaster emergency alarm method 100 based on artificial intelligence mainly includes the following steps:

[0051] In step 101, fire monitoring is performed on a target area, and temperature monitoring data of the target area is automatically collected.

[0052] In specific implementation, the target area can be monitored for fire through an infrared thermometer to automatically collect temperature monitoring data of the target area, wherein the temperature monitoring data includes temperature data of the target area under different temperature measurement paths, wherein the temperature data under each temperature measurement path is collected by the infrared thermometer, that is, the temperature data includes multiple temperature values ​​collected by the infrared thermometer under the corresponding temperature measurement path, and the same number of temperature values ​​are collected in an equal time interval collection manner under each temperature measurement path, wherein the temperature measurement path represents the path that the infrared thermometer passes from its installation position to the temperature measurement point. By automatically collecting temperature monitoring data, the occurrence and development of fire can be effectively monitored, thereby timely discovering and responding to fire risks.

[0053] In step 102, the collected temperature monitoring data is divided into multiple temperature data samples, and the fire risk association information at each time node when a fire occurs in the target area is determined based on all the temperature data samples, and then the dominant temperature when a fire occurs in the target area is determined based on the fire risk association information at different time nodes and the combustion load when the target area burns.

[0054] In some embodiments, the collected temperature monitoring data may be divided into a plurality of temperature data samples in the following manner, namely:

[0055] Get all temperature measurement paths of the temperature measurement equipment;

[0056] For each temperature measurement path, the temperature data corresponding to the temperature measurement path in the temperature monitoring data is divided into a temperature data sample, so as to obtain the temperature data sample of each temperature measurement path.

[0057] In specific implementation, all temperature measurement paths of the temperature measurement equipment can be obtained through the fire monitoring database. In addition, in other embodiments, other methods can be used to obtain the temperature measurement paths, which are not limited here.

[0058] It should be noted that the temperature data sample in the present application represents a set of all temperature values ​​measured by the infrared thermometer according to the corresponding temperature measurement path. The sample number of each temperature data sample is consistent. By determining the temperature data sample, the temperature changes in different monitoring areas can be effectively evaluated and combustion anomalies during combustion can be identified.

[0059] It should also be noted that, in this embodiment, all temperature values ​​collected at the same time node under each temperature measurement path are used as the temperature value corresponding to the time node, that is, one time node corresponds to multiple temperature values.

[0060] In some embodiments, the fire risk association information at each time point when a fire occurs in the target area can be determined based on all temperature data samples in the following manner, namely:

[0061] Obtain temperature data samples corresponding to each temperature measurement path;

[0062] Extract multiple temperature correlation values ​​at each time node from all temperature data samples, and then obtain the temperature correlation sequence corresponding to each time node;

[0063] For each time node, the fire risk association information at the time node is determined according to the temperature association sequence corresponding to the time node and the temperature association sequences corresponding to the other time nodes, and then the fire risk association information at each time node when a fire occurs in the target area is obtained.

[0064] In some embodiments, multiple temperature-related values ​​at each time node are extracted from all temperature data samples, and then the temperature-related sequences corresponding to each time node are obtained, which can be specifically obtained in the following manner, namely:

[0065] Align all temperature data samples;

[0066] Extract multiple temperature values ​​at each time node from all aligned temperature data samples in the order of time nodes. For each time node, perform correlation calculation on all temperature values ​​at the time node to obtain all temperature correlation values.

[0067] All temperature correlation values ​​are combined in the order of correlation calculation to obtain the temperature correlation sequence corresponding to the time node, and then the temperature correlation sequence corresponding to each time node is obtained.

[0068] In specific implementation, all temperature values ​​under the time node are associated with each other to obtain all temperature associated values. The association calculation can be processed by sliding calculation, that is, starting from the second temperature value, the current temperature value is calculated with the previous temperature value for difference, until the last temperature value is traversed, and the result of each difference calculation is used as the temperature associated value to obtain all temperature associated values, which will not be repeated here. In addition, in other embodiments, other calculation methods can be used to calculate the temperature associated values, which are not limited here.

[0069] It should be noted that, in this embodiment, the temperature correlation sequence includes multiple temperature correlation values, and the temperature correlation values ​​represent the degree of correlation between temperatures under different temperature measurement paths. By determining the temperature correlation sequence, the temperature change trend and correlation characteristics along the temperature measurement path can be effectively characterized, thereby improving the accuracy of subsequent analysis of the fire situation.

[0070] In some embodiments, the fire risk association information at a time node may be determined according to the temperature association sequence corresponding to the time node and the temperature association sequences corresponding to other time nodes in the following manner, namely:

[0071] Get the total number of temperature-related values ​​of the temperature-related sequence corresponding to the time node;

[0072] The inner product calculation is performed on the temperature association sequence corresponding to the time node and the temperature association sequence corresponding to the other time nodes in the order of the time nodes, and the ratio calculation is performed on each inner product calculation result and the total number of the temperature association values, and the result of each ratio calculation is used as the fire risk association entropy, and all the fire risk association entropies are combined into the fire risk association information under the time node.

[0073] It should be noted that the fire risk correlation information includes multiple fire risk correlation entropies. In the present application, the fire risk correlation entropy is used to measure the degree of confusion of the fire risk changes between two time nodes. The larger the fire risk correlation entropy, the greater the degree of confusion of the fire risk changes between the two time nodes. The smaller the fire risk correlation entropy, the smaller the degree of confusion of the fire risk changes between the two time nodes. By determining the fire risk correlation entropy, the correlation between temperature changes at the same time nodes under different temperature measurement paths in the target area can be effectively identified and evaluated, thereby more accurately predicting the fire risk.

[0074] It should also be noted that in this application, all fire risk correlation entropies are sorted according to the order of the inner product calculation corresponding to the time nodes to obtain the fire risk correlation information under the time nodes.

[0075] In some embodiments, reference Figure 2 As shown, this figure is an exemplary flow chart for determining the dominant temperature according to some embodiments of the present application. In this embodiment, the dominant temperature when a fire occurs in the target area can be determined according to the fire risk association information at different time nodes and the combustion load when the target area is burning. The following steps can be used to achieve:

[0076] In step 1021, the combustion load of the target area during combustion is obtained;

[0077] Secondly, in step 1022, a fire risk correlation entropy matrix is ​​constructed according to the fire risk correlation information at different time nodes;

[0078] Then in step 1023, all fire load factors and all fire load response vectors of the target area when it is subjected to fire combustion are determined based on the fire risk correlation entropy matrix;

[0079] Then, in step 1024, the combustion response characteristics of the target area when subjected to fire combustion are determined according to all fire load factors and all fire load response vectors;

[0080] Finally, in step 1025, the dominant temperature of the target area when a fire occurs is determined based on the combustion response characteristics and the combustion load.

[0081] In specific implementation, the combustion load of the target area when it is burning can be obtained from the fire monitoring database. The combustion load represents the total amount of combustible materials in the target area and the heat that can be released by its combustion. By determining the combustion load, a scientific basis can be provided for fire risk alarm. In addition, in other embodiments, other methods can also be used to obtain the combustion load, which is not limited here.

[0082] In some embodiments, the fire risk correlation entropy matrix can be constructed according to the fire risk correlation information at different time nodes in the following manner, namely:

[0083] Get the temperature correlation sequence corresponding to each time node;

[0084] Determine the autocorrelation entropy of each temperature correlation series under autocorrelation;

[0085] The fire risk correlation entropy matrix is ​​determined based on all fire risk correlation information and all autocorrelation entropies.

[0086] In the specific implementation, the autocorrelation entropy of each temperature correlation sequence under autocorrelation is determined, that is, a temperature correlation sequence is selected as the selected temperature correlation sequence, the variance of all temperature correlation values ​​in the selected temperature correlation sequence is used as the autocorrelation entropy of the selected temperature correlation sequence under autocorrelation, and the autocorrelation entropy of the remaining temperature correlation sequences under autocorrelation is continued to be determined, so as to obtain the autocorrelation entropy of each temperature correlation sequence under autocorrelation.

[0087] It should be noted that the autocorrelation entropy in this embodiment represents the degree of disorder of all temperature correlation values ​​in the temperature correlation sequence. By determining the autocorrelation entropy, data that has a significant impact on the fire development process in fire monitoring can be identified, thereby improving the accuracy of fire analysis.

[0088] In the specific implementation, the fire risk correlation entropy matrix is ​​determined based on all the fire risk correlation information and all the autocorrelation entropies, that is: first, all the autocorrelation entropies are arranged as the main diagonal elements of the fire risk correlation entropy matrix in the order of time nodes; then, each fire risk correlation entropy in the corresponding fire risk correlation information is filled in according to the order of time nodes, and then the fire risk correlation entropy matrix is ​​obtained.

[0089] It should be noted that the fire risk correlation entropy matrix in this embodiment represents the relationship between the fire risk factors of different temperature measurement paths. By determining the fire risk correlation entropy matrix, the occurrence and spread trend of the fire can be effectively identified and understood, thereby improving the efficiency of fire monitoring and disaster emergency alarm.

[0090] In a specific implementation, all fire load factors and all fire load response vectors of the target area when it is subjected to fire combustion are determined based on the fire risk associated entropy matrix, that is: the fire risk associated entropy matrix is ​​subjected to eigendecomposition, and each eigenvalue obtained by the decomposition is used as the fire load factor of the target area when it is subjected to fire combustion, and the eigenvector corresponding to each eigenvalue is used as the fire load response vector of the target area when it is subjected to fire combustion, thereby obtaining all fire load factors and all fire load response vectors of the target area when it is subjected to fire combustion.

[0091] It should be noted that the fire load response vector in this embodiment represents the weighted combination of the combustion response when the fire is burning on different temperature measurement paths under the target area. The fire load factor in this embodiment represents the combustion load index of the target area under the current fire burning conditions. The larger the fire load factor, the greater the combustion load of the target area under the current fire burning conditions, indicating that the target area is more intensely burned. The smaller the fire load factor, the smaller the combustion load of the target area under the current fire burning conditions, indicating that the target area is more moderately burned. By determining the fire load factor and the fire load response vector, the main temperature change patterns can be identified in fire monitoring, thereby improving the efficiency of fire monitoring.

[0092] In specific implementation, the combustion response characteristics of the target area when it is subjected to fire combustion are determined based on all fire load factors and all fire load response vectors, that is, the fire load response vector corresponding to the maximum fire load factor is extracted, and the extracted fire load response vector is used as the combustion response characteristic of the target area when it is subjected to fire combustion.

[0093] It should be noted that the combustion response characteristics in this application represent the combustion response weight combination when the fire burns on different temperature measurement paths under the target area. By determining the combustion response characteristics, the efficiency and accuracy of subsequent fire analysis can be improved, so as to make more reasonable response measures.

[0094] In some embodiments, the dominant temperature of a target area when a fire occurs may be determined based on the combustion response characteristics and the combustion load in the following manner, namely:

[0095] Acquire all temperature data samples, and construct a fire temperature array when a fire occurs in the target area from all temperature data samples;

[0096] Performing data projection on the fire temperature array according to the combustion response characteristics to obtain temperature projection characteristics;

[0097] The dominant temperature of a target area when a fire occurs is determined by the combustion load and the temperature projection characteristics.

[0098] In a specific implementation, a fire temperature array when a fire occurs in a target area is constructed from all temperature data samples, that is, all temperature values ​​in each temperature data sample are used as row elements of the fire temperature array in the order of the temperature measurement path, and then the fire temperature array when a fire occurs in the target area is obtained, and the fire temperature array represents a matrix composed of all temperature data samples.

[0099] In a specific implementation, data projection is performed on the fire temperature array according to the combustion response characteristic to obtain a temperature projection characteristic, that is, the fire temperature array is multiplied by the combustion response characteristic and the multiplication result is used as the temperature projection characteristic.

[0100] It should be noted that, in this embodiment, data projection represents the process of multiplying two sets of data, and the temperature projection feature in this embodiment represents the responsiveness evaluation of each temperature data sample in the target area to the combustion reflection during the fire. Each element value in the temperature projection feature represents the responsiveness of the corresponding temperature data sample to the combustion reflection, that is, the first element value corresponds to the first temperature data sample, and so on. By determining the temperature projection feature, the thermodynamic behavior of the fire can be effectively analyzed, thereby improving the safety and effectiveness of fire prevention and control.

[0101] In some embodiments, the dominant temperature when a fire occurs in the target area is determined by the combustion load and the temperature projection feature, that is:

[0102] Get all temperature data samples;

[0103] Extracting the maximum element value from the temperature projection feature, and using the temperature data sample corresponding to the element value as the dominant temperature data sample;

[0104] A dominant temperature when a fire occurs in a target area is determined based on the dominant temperature data samples and the combustion load.

[0105] In a specific implementation, the dominant temperature when a fire occurs in the target area is determined based on the dominant temperature data sample and the combustion load, that is, the combustion load is multiplied by each temperature value in the dominant temperature data sample, and each multiplication result is recombined in the order of the dominant temperature data sample to obtain the dominant temperature when a fire occurs in the target area.

[0106] It should be noted that the dominant temperature in this application refers to the temperature combination that plays a decisive role in the change of combustion temperature in the fire in the target area. By determining the dominant temperature, the spread of subsequent fires can be analyzed and the combustion intensity in the target area can be understood, so as to formulate effective fire fighting strategies to improve the efficiency of fire fighting response and the accuracy of fire emergency alarms.

[0107] In step 103, flame monitoring data of the target area is automatically collected, and multiple flame hotspots of the target area when burning are extracted from the flame monitoring data. Then, the flame front area during the fire burning is determined based on all the flame hotspots, and the diffusion characteristics of the flame front area during combustion are extracted.

[0108] It should be noted that the flame monitoring data in the present application includes multiple flame monitoring images of the fire scene automatically taken by an artificial intelligence-based intelligent monitoring camera at different time points. The flame monitoring images record the key combustion characteristics of the target area, such as flames, smoke, etc. Automatically collecting flame monitoring data is conducive to quickly identifying the key combustion characteristics of the fire and evaluating the fire situation through artificial intelligence. In addition, in other embodiments, other shooting equipment can also be used to collect flame monitoring data of the target area, which is not limited here.

[0109] It should also be noted that the smart monitoring camera in this application refers to a monitoring device with artificial intelligence as the background, integrating advanced image processing, data analysis and network communication functions, which will not be repeated here.

[0110] In some embodiments, reference Figure 3 As shown in FIG. 1 , this figure is an exemplary flow chart of determining multiple flame hotspots according to some embodiments of the present application. In this embodiment, extracting multiple flame hotspots of the target area during combustion from the flame monitoring data can be achieved by using the following steps:

[0111] In step 1031, a flame monitoring image at the current moment is obtained from the flame monitoring data;

[0112] Then in step 1032, the flame monitoring image is enhanced to obtain a flame monitoring enhanced image;

[0113] Then in step 1033, the flame characteristic value in the flame monitoring enhanced image is extracted;

[0114] Finally, in step 1034, combustion heat analysis is performed on all pixel points in the flame monitoring enhanced image based on the flame characteristic value, thereby obtaining multiple flame hot spots in the target area during combustion.

[0115] In specific implementation, the flame monitoring image can be enhanced by using a method for improving contrast to obtain a flame monitoring enhanced image. Specifically, the flame monitoring image can be input into the image processing software Open CV to improve the image contrast. In addition, in other embodiments, other methods can be used to enhance the flame monitoring image, such as image sharpening, etc., which are not limited here.

[0116] It should be noted that the flame monitoring enhanced image in this embodiment represents an enhanced image obtained by improving the color difference and image brightness in the flame monitoring image. By determining the flame monitoring enhanced image, the burning characteristics of the fire can be made clearer, thereby improving the accuracy of the fire situation analysis.

[0117] In some embodiments, the flame characteristic value in the flame monitoring enhanced image may be extracted in the following manner, namely:

[0118] Performing grayscale conversion on the flame detection enhanced image, and extracting all pixel points whose grayscale values ​​are greater than a preset grayscale threshold;

[0119] The flame characteristic value in the flame monitoring enhanced image is determined according to the grayscale values ​​corresponding to all the extracted pixel points.

[0120] In specific implementation, the image processing software Open CV can be used to perform grayscale conversion on the flame detection enhancement image, and all pixels whose grayscale values ​​are greater than a preset grayscale threshold are extracted. No further details are given here. The preset grayscale threshold represents a grayscale value set in advance. The specific value can be set according to actual application requirements and is not limited here.

[0121] In specific implementation, the flame characteristic value in the flame monitoring enhanced image is determined according to the grayscale values ​​corresponding to all the extracted pixel points, that is, the grayscale values ​​corresponding to all the extracted pixel points are averaged, and the result of the average calculation is used as the flame characteristic value in the flame monitoring enhanced image. In this embodiment, the flame characteristic value represents a combustion characteristic parameter describing the flame during the combustion process.

[0122] In some embodiments, the combustion heat analysis is performed on all pixels in the flame monitoring enhanced image based on the flame characteristic value, and then a plurality of flame hot spots in the target area during combustion are obtained in the following manner, namely:

[0123] Comparing the flame characteristic value with the grayscale value of each pixel in the flame monitoring enhanced image, extracting all grayscale values ​​greater than the flame characteristic value, and not processing all grayscale values ​​less than or equal to the flame characteristic value;

[0124] The pixel points corresponding to each extracted gray value are used as the flame hotspots of the target area when it is burning.

[0125] It should be noted that the flame hotspots in the present application represent the points where burning is most vigorous in the target area. These points often correspond to the places where the flames burn most vigorously in the fire. By determining the flame hotspots, the location of the fire and the fire intensity assessment can be achieved, thereby improving the accuracy and real-time performance of fire monitoring.

[0126] It should be noted that, in this embodiment, the combustion heat analysis represents the process of obtaining the point with the most vigorous combustion in the target area, wherein the combustion heat analysis is performed on all the pixel points in the flame monitoring and enhanced image based on the flame characteristic value, that is: the flame characteristic value is compared with the grayscale value of each pixel point in the flame monitoring and enhanced image, and all grayscale values ​​greater than the flame characteristic value are extracted, and all grayscale values ​​less than or equal to the flame characteristic value are not processed; the pixel points corresponding to the extracted grayscale values ​​are used as the flame hotspots of the target area when burning, that is, the combustion heat analysis of all the pixel points in the flame monitoring and enhanced image is completed.

[0127] In some embodiments, the flame front area during fire burning is determined based on all flame hotspots, that is, all flame hotspots are connected to obtain the flame front area during fire burning. Specifically, the convex hull algorithm in image processing can be used to connect all flame hotspots, and a closed area formed by the connection is used as the flame front area during fire burning. It will not be repeated here. In addition, in other embodiments, other connection algorithms can also be used to connect all flame hotspots, which are not limited here.

[0128] It should be noted that the flame front area in the present application refers to the front area of ​​the flame during the fire combustion process. This area is the front line of flame propagation, marking the boundary where the flame spreads from one place to another. This area is usually where the fire combustion reaction is most intense. By determining the flame front area, we can effectively formulate fire extinguishing strategies and evaluate the diffusivity of the fire, thereby providing accurate early warning of fire disasters.

[0129] In some embodiments, the diffusion characteristics of the flame front region during combustion may be extracted in the following manner, namely:

[0130] determining combustion geometry information of the flame front region;

[0131] The diffusion characteristics of the flame front region during combustion are extracted from the combustion geometry information.

[0132] In specific implementation, a contour detection algorithm can be used to perform geometric analysis on the flame front area to obtain combustion geometry information of the flame front area. Specifically, first, the contour information of the flame front area is extracted using the FindContours function in the contour detection algorithm; then, the contour information is traversed through Open CV to obtain combustion geometry information. The specific implementation process will not be repeated here.

[0133] It should be noted that, in this embodiment, the combustion geometry information includes the front zone area and the front zone boundary length of the flame front area. By determining the combustion geometry information, the diffusion mode and diffusion characteristics of the fire can be effectively identified.

[0134] In a specific implementation, the diffusion characteristics of the flame front area during combustion are extracted from the combustion geometric information, that is, the front area and the front area boundary length are extracted from the combustion geometric information, and the ratio of the product of the front area and the preset geometric coefficient to the square value of the front area boundary length is used as the diffusion characteristic to characterize the trend of fire spread. In addition, in other embodiments, other calculation methods can also be used to calculate the diffusion characteristics, which are not limited here.

[0135] It should be noted that the preset geometric coefficient in this embodiment can be used to standardize the results of the above-mentioned geometric analysis, so as to ensure the geometric rationality of the results. Specifically, the preset geometric coefficient is usually taken as 4π. In addition, in other embodiments, it can be set according to actual needs.

[0136] In addition, it should be noted that the diffusion characteristics in the present application represent the risk characteristics of the flame spreading from the target area to the surrounding areas. The diffusion characteristics reflect the risk of the flame spreading from the target area to the surrounding areas. In the target area, under the influence of environmental factors such as air flow, the fire will show a trend of gradually spreading outward. The larger the diffusion characteristics, the more obvious the risk characteristics of the flame spreading from the target area to the surrounding areas. When the diffusion characteristics are smaller, the weaker the risk characteristics of the flame spreading from the target area to the surrounding areas. Therefore, by determining the diffusion characteristics, it is helpful to understand the spread dynamics of the fire and evaluate the fire risk, so as to formulate effective emergency response and fire extinguishing strategies.

[0137] In step 104, the thermal responsiveness of the fire during combustion and diffusion is determined according to the dominant temperature and the diffusion characteristics, and then a fire risk signal of the target area is generated according to the thermal responsiveness.

[0138] In some embodiments, the thermal responsiveness of the fire during the combustion and diffusion process may be determined according to the dominant temperature and the diffusion characteristics in the following manner, namely:

[0139] determining a load factor for the prevailing temperature;

[0140] The thermal responsiveness of the fire during combustion diffusion is determined based on the load factor and the diffusion characteristic.

[0141] In specific implementation, the load coefficient of the dominant temperature is determined, that is, the mean value of the dominant temperature is used as the load coefficient of the dominant temperature. The load coefficient represents the influence weight of the dominant temperature on the fire development process in the target area. By determining the load coefficient, the dominant influence of the dominant temperature on the temperature distribution of the entire fire development process can be reflected. In addition, in other embodiments, other methods can be used to determine the load coefficient, which is not limited here.

[0142] In specific implementation, the thermal responsiveness of the fire during combustion and diffusion is determined based on the load coefficient and the diffusion characteristic, that is, the load coefficient and the diffusion characteristic are weightedly summed, and the result of the weighted summation is used as the thermal responsiveness of the fire during combustion. The size of the specific weight in the calculation process can be set according to the actual application requirements. In addition, in other embodiments, other calculation methods can also be used to calculate the thermal responsiveness, which are not limited here.

[0143] It should be noted that the thermal responsiveness in the present application indicates the sensitivity of the environment of the target area to heat input and fire spread at the current moment. The greater the thermal responsiveness, the more sensitive the environment of the target area to heat input and fire spread at the current moment. The smaller the thermal responsiveness, the less sensitive the environment of the target area to heat input and fire spread at the current moment. The thermal responsiveness can help the fire protection and emergency management departments to better assess the speed of fire spread and the impact of the released heat on the environment of the target area, so as to provide timely warnings and take corresponding prevention and control measures.

[0144] In specific implementation, a fire risk signal of the target area is generated based on the thermal responsiveness, that is, the thermal responsiveness is compared with a preset response index. When the thermal responsiveness is greater than the preset response index, the fire disaster emergency alarm device is triggered to automatically generate a fire risk signal when the fire is burning.

[0145] It should be noted that the preset response index in this embodiment represents a preset thermal response threshold value, and the specific value can be set according to actual application requirements and is not limited here.

[0146] It should also be noted that the fire risk signal in this application is a signal instruction used to indicate that the fire in the target area under the current monitoring state has the risk of burning and spreading. Burning and spreading in a fire are manifested as a rapid rise in temperature, a rapid spread of flames, and a sudden increase in smoke concentration. Burning and spreading can quickly expand the scale of the fire and greatly increase the difficulty of fire fighting. Therefore, generating a fire risk signal is crucial for fire prevention and emergency response.

[0147] In step 105, a fire disaster emergency alarm is performed on the target area based on the fire risk signal.

[0148] In specific implementation, a fire disaster emergency alarm is performed in the target area based on the fire risk signal, that is: when a fire risk signal is generated, the fire disaster emergency alarm device will use wireless communication technology to send the fire risk signal to each alarm center for a fire disaster emergency alarm. When each alarm center receives the fire risk signal, it will start the fire fighting execution process to execute the relevant fire emergency plan.

[0149] In addition, in another aspect of the present application, in some embodiments, the present application provides a disaster emergency alarm device based on artificial intelligence, referring to Figure 4 , which is a schematic diagram of exemplary hardware and / or software of an artificial intelligence-based disaster emergency alarm device according to some embodiments of the present application. The artificial intelligence-based disaster emergency alarm device 200 includes: an acquisition module 201, a processing module 202 and an execution module 203, which are described as follows:

[0150] Acquisition module 201, in this application, acquisition module 201 is mainly used to monitor the fire in the target area and automatically collect temperature monitoring data of the target area;

[0151] Processing module 202, in the present application, the processing module 202 is mainly used to divide the collected temperature monitoring data into multiple temperature data samples, determine the fire risk association information at each time node when a fire occurs in the target area based on all the temperature data samples, and then determine the dominant temperature when a fire occurs in the target area according to the fire risk association information at different time nodes and the combustion load when the target area burns;

[0152] The processing module 202 is also used to automatically collect flame monitoring data of the target area, extract multiple flame hot spots in the target area when burning from the flame monitoring data, and then determine the flame front area when the fire is burning according to all the flame hot spots, and extract the diffusion characteristics of the flame front area when burning;

[0153] The processing module 202 is further used to determine the thermal responsiveness of the fire during combustion and diffusion according to the dominant temperature and the diffusion characteristics, and then generate a fire risk signal of the target area according to the thermal responsiveness;

[0154] Execution module 203, in the present application, the execution module 203 is mainly used to perform fire disaster emergency alarm on the target area based on the fire risk signal.

[0155] In addition, the present application also provides a computer device, which includes a memory and a processor, the memory stores code, and the processor is configured to obtain the code and execute the above-mentioned artificial intelligence-based disaster emergency alarm method.

[0156] In some embodiments, reference Figure 5, which is a schematic diagram of the structure of a computer device for implementing an artificial intelligence-based disaster emergency alarm method according to some embodiments of the present application. The artificial intelligence-based disaster emergency alarm method in the above embodiment can be implemented by Figure 5 The computer device 300 shown in the figure is implemented, and the computer device 300 includes at least one processor 301, a communication bus 302, a memory 303 and at least one communication interface 304.

[0157] Processor 301 can be a general-purpose central processing unit (CPU), or an application-specific integrated circuit (ASIC) or one or more for controlling the execution of the artificial intelligence-based disaster emergency alarm method in the present application.

[0158] The communication bus 302 may be used to transmit information between the above-mentioned components.

[0159] The memory 303 may be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, an optical disc storage (including a compressed optical disc, a laser disc, an optical disc, a digital versatile disc, a Blu-ray disc, etc.), a magnetic disk or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of an instruction or data structure and can be accessed by a computer, but is not limited thereto. The memory 303 may exist independently and be connected to the processor 301 via the communication bus 302. The memory 303 may also be integrated with the processor 301.

[0160] The memory 303 is used to store the program code for executing the solution of the present application, and the execution is controlled by the processor 301. The processor 301 is used to execute the program code stored in the memory 303. The program code may include one or more software modules. The determination of the disaster emergency alarm method based on artificial intelligence in the above embodiment can be implemented by the processor 301 and one or more software modules in the program code in the memory 303.

[0161] The communication interface 304 uses any transceiver or other device for communicating with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc.

[0162] In a specific implementation, as an embodiment, a computer device may include multiple processors, each of which may be a single-CPU processor or a multi-CPU processor. The processor here may refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions).

[0163] The above-mentioned computer device may be a general-purpose computer device or a special-purpose computer device. In a specific implementation, the computer device may be a desktop computer, a portable computer, a network server, a personal digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device or an embedded device. The embodiment of the present application does not limit the type of computer device.

[0164] In addition, the present application also provides a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, it implements the above-mentioned disaster emergency alarm method based on artificial intelligence.

[0165] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0166] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A disaster emergency alarm method based on artificial intelligence, characterized in that: The steps include: Conduct fire monitoring in the target area and automatically collect temperature monitoring data in the target area; The collected temperature monitoring data is divided into multiple temperature data samples, and the fire risk association information at each time node when a fire occurs in the target area is determined based on all the temperature data samples, and then the dominant temperature when a fire occurs in the target area is determined according to the fire risk association information at different time nodes and the combustion load when the target area is burning; Automatically collect flame monitoring data of the target area, extract multiple flame hot spots of the target area when burning from the flame monitoring data, and then determine the flame front area when the fire is burning based on all the flame hot spots, and extract the diffusion characteristics of the flame front area when burning; Determining the thermal response of the fire burning and spreading according to the dominant temperature and the diffusion characteristics, and then generating a fire risk signal of the target area according to the thermal response; Performing a fire disaster emergency alarm on the target area based on the fire risk signal; Among them, the dominant temperature when a fire occurs in the target area is determined based on the fire risk association information at different time nodes and the combustion load when the target area is burning, which specifically includes: Obtain the combustion load when the target area is burning; Construct a fire risk correlation entropy matrix based on the fire risk correlation information at different time nodes; Determine all fire load factors and all fire load response vectors of the target area when it is subjected to fire combustion based on the fire risk correlation entropy matrix; Determine the combustion response characteristics of the target area when it is subjected to fire combustion according to all fire load factors and all fire load response vectors; Determining a dominant temperature when a fire occurs in a target area based on the combustion response characteristics and the combustion load; Wherein, the dominant temperature represents the temperature combination that plays a decisive role in the change of the combustion temperature in the fire of the target area; Among them, the fire risk correlation entropy matrix is ​​constructed according to the fire risk correlation information at different time nodes in the following way: Get the temperature correlation sequence corresponding to each time node; Determine the autocorrelation entropy of each temperature correlation sequence under autocorrelation, wherein the autocorrelation entropy represents the degree of disorder of all temperature correlation values ​​in the temperature correlation sequence; The fire risk correlation entropy matrix is ​​determined based on all fire risk correlation information and all autocorrelation entropies.

2. The method according to claim 1, characterized in that Dividing the collected temperature monitoring data into multiple temperature data samples specifically includes: Get all temperature measurement paths of the temperature measurement equipment; For each temperature measurement path, the temperature data corresponding to the temperature measurement path in the temperature monitoring data is divided into a temperature data sample, so as to obtain the temperature data sample of each temperature measurement path.

3. The method according to claim 1, characterized in that The fire risk related information at each time node when a fire occurs in the target area based on all temperature data samples specifically includes: Obtain temperature data samples corresponding to each temperature measurement path; Extract multiple temperature correlation values ​​at each time node from all temperature data samples, and then obtain the temperature correlation sequence corresponding to each time node; For each time node, the fire risk association information at the time node is determined according to the temperature association sequence corresponding to the time node and the temperature association sequences corresponding to the other time nodes, and then the fire risk association information at each time node when a fire occurs in the target area is obtained.

4. The method according to claim 1, characterized in that Extracting multiple flame hot spots of the target area during combustion from the flame monitoring data specifically includes: Acquire a flame monitoring image at a current moment from the flame monitoring data; Performing image enhancement on the flame monitoring image to obtain a flame monitoring enhanced image; Extracting flame characteristic values ​​from the flame monitoring enhanced image; Combustion heat analysis is performed on all pixel points in the flame monitoring enhanced image based on the flame characteristic value, thereby obtaining a plurality of flame hot spots in the target area when burning.

5. The method according to claim 1, characterized in that Extracting the diffusion characteristics of the flame front region during combustion specifically includes: determining combustion geometry information of the flame front region; The diffusion characteristics of the flame front region during combustion are extracted from the combustion geometry information.

6. The method according to claim 1, characterized in that The flame monitoring data includes flame monitoring images of the fire scene taken at different time points.

7. A disaster emergency alarm device based on artificial intelligence, which uses the method described in any one of claims 1 to 6 to perform disaster emergency alarm, characterized in that: The disaster emergency alarm device comprises: An acquisition module is used to monitor the target area for fire and automatically collect temperature monitoring data of the target area; The processing module is used to divide the collected temperature monitoring data into multiple temperature data samples, determine the fire risk association information at each time node when a fire occurs in the target area based on all the temperature data samples, and then determine the dominant temperature when a fire occurs in the target area according to the fire risk association information at different time nodes and the combustion load when the target area burns, wherein determining the dominant temperature when a fire occurs in the target area according to the fire risk association information at different time nodes and the combustion load when the target area burns specifically includes: Obtain the combustion load when the target area is burning; Construct a fire risk correlation entropy matrix based on the fire risk correlation information at different time nodes; Determine all fire load factors and all fire load response vectors of the target area when it is subjected to fire combustion based on the fire risk correlation entropy matrix; Determine the combustion response characteristics of the target area when it is subjected to fire combustion according to all fire load factors and all fire load response vectors; Determining a dominant temperature when a fire occurs in a target area based on the combustion response characteristics and the combustion load; The processing module is also used to automatically collect flame monitoring data of the target area, extract multiple flame hot spots in the target area when burning from the flame monitoring data, and then determine the flame front area when the fire is burning based on all the flame hot spots, and extract the diffusion characteristics of the flame front area when burning; The processing module is further used to determine the thermal responsiveness of the fire during combustion and diffusion according to the dominant temperature and the diffusion characteristics, and then generate a fire risk signal of the target area according to the thermal responsiveness; The execution module is used to perform a fire disaster emergency alarm on the target area based on the fire risk signal.

8. A computer device, characterized in that: The computer device includes a memory and a processor, the memory stores codes, and the processor is configured to obtain the codes and execute the artificial intelligence-based disaster emergency alarm method as described in any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the artificial intelligence-based disaster emergency alarm method according to any one of claims 1 to 6 is implemented.

Images