A disaster emergency response system based on big data

Abstract

The present invention discloses a disaster emergency response system based on big data, comprising a data collection module, a fire warning module and a fire emergency module, characterized in that: the data collection module is used to collect visual data and comprehensive environmental data in the monitored target area in real time, and input the comprehensive information of all personnel entering the building into the system; the fire warning module is used to detect whether a fire occurs in the target area, further analyze the spread trend and danger level of the fire, and adjust the operation mode of the sprinkler system according to the analysis result; the fire emergency module is used to obtain the position of trapped personnel in the building, divert the trapped personnel, and plan an evacuation route for each diversion group; the data collection module comprises a data entry module, a visual module and a sensor module; the data entry module is used to input the comprehensive information of personnel entering the building into the system; the present invention has the characteristics of improving evacuation efficiency and accuracy.

Description

Technical Field

[0001] The present invention relates to the technical field of fire emergency management, and in particular to a disaster emergency response system based on big data. Background Art

[0002] With the rapid development of social economy, the construction process of urbanization in my country is accelerating, and large-scale comprehensive buildings are constantly emerging, such as large supermarkets, airports, and high-speed rail stations. Fires in these large buildings often occur. These places have a large flow of people, and it takes a certain amount of time to wait for fire rescue. During this period, how to evacuate trapped people to a safer place to wait for rescue is of great significance. However, since most people do not understand the structure of large buildings, the existing technology manually guides the trapped people, which is prone to negligence and omissions, resulting in low efficiency of personnel evacuation. Moreover, the existing technology senses the environment through temperature control or smoke sensing devices, and then uses covering spraying to prevent the spread of fire. Since the temperature and smoke concentration cannot instantly reach the threshold of the system when a fire occurs, the existing technology is prone to misjudgment and missed judgment, resulting in poor accuracy of the sprinkler system. Some items that will not be spread by the fire are sprayed, which aggravates the fire loss. Therefore, it is very necessary to design a disaster emergency system based on big data to improve evacuation efficiency and accuracy. Summary of the invention

[0003] The purpose of the present invention is to provide a disaster emergency response system based on big data to solve the problems raised in the above background technology.

[0004] In order to solve the above technical problems, the present invention provides the following technical solutions: a disaster emergency response system based on big data, comprising a data collection module, a fire warning module and a fire emergency module, characterized in that: the data collection module is used to collect visual data and comprehensive environmental data in the monitored target area in real time, and enter the comprehensive information of all personnel entering the building into the system; the fire warning module is used to detect whether a fire occurs in the target area, further analyze the spread trend and danger level of the fire, and adjust the operation mode of the sprinkler system according to the analysis results; the fire emergency module is used to obtain the position of trapped personnel in the building, divert the trapped personnel, and plan an evacuation route for each diversion group;

[0005] The environment detection module includes a visual analysis submodule, an air analysis submodule and a combustible material analysis submodule. The visual analysis submodule is used to analyze whether there is a flame or smoke in the target detection area and control the sprinkler system according to the analysis result. The air analysis submodule is used to analyze the smoke concentration in the environment and adjust the sprinkler system in combination with the visual analysis result. The combustible material analysis submodule is used to analyze the combustible materials around the fire point.

[0006] The sprinkler management module includes a spread trend analysis submodule, a danger level analysis submodule and a fire sprinkler control submodule. The spread trend analysis submodule is used to analyze the spread trend of the fire and adjust the sprinkler system according to the spread trend. The danger level analysis submodule is used to analyze the danger level of the fire and adjust the evacuation strategy of trapped personnel according to the analysis results. The fire sprinkler control submodule is used to control the spraying mode of the sprinkler equipment according to the fire spread trend and the danger level of the fire.

[0007] According to the above technical solution, the data collection module includes a data entry module, a visual module and a sensor module. The data entry module is used to enter the comprehensive information of people entering the building into the system, the visual module is used to collect visual images in the target detection area in real time, and the sensor module is used to collect the temperature and smoke concentration of the surrounding environment in real time.

[0008] According to the above technical solution, the fire warning module includes an environmental detection module, which is used to analyze the temperature, smoke concentration and whether there are flames in the target area, determine whether there is a fire based on the analysis results, and then anchor the fire location for spraying.

[0009] According to the above technical solution, the fire warning module also includes a sprinkler management module, which is used to analyze the spread trend of the fire and the danger level of the fire, and adjust the operation mode of the sprinkler system according to the analysis results.

[0010] According to the above technical solution, the fire emergency module includes a communication module, a route planning module and a personnel evacuation module. The communication module is used to comfort and guide the evacuation of trapped persons, the route planning module is used to divert trapped persons according to their positions and plan evacuation routes, and the personnel evacuation module is used to select an evacuation strategy for trapped persons according to the degree of danger of the fire.

[0011] According to the above technical solution, the operation method of the disaster emergency system includes the following steps:

[0012] Step S1: Through the data entry module, the comprehensive information of all people entering the building and the annotated fire image data are entered into the system, the visual image of the target area is collected in real time through the visual module, and the comprehensive data of the surrounding environment is collected in real time through the sensor module;

[0013] Step S2: After the data is entered into the system, the system starts the environmental detection module, starts to analyze whether smoke or flames appear in the target area, analyzes the smoke concentration in the target area, and controls the start of the sprinkler system according to the analysis results;

[0014] Step S3: After the sprinkler system is started, the system begins to analyze the flammable materials around the fire point, the fire spread trend and the corresponding fire danger level, and further adjusts the operation mode of the sprinkler system according to the analysis results;

[0015] Step S4: When a fire is discovered, the system starts the fire emergency module, begins real-time monitoring of the fire, adjusts the personnel evacuation strategy according to the development trend of the fire, and plans evacuation routes to evacuate personnel.

[0016] According to the above technical solution, step S2 further includes the following steps:

[0017] Step S21: Obtain the annotated fire image data, identify the applicable scene identifier in the image, cluster the fire image data according to the applicable scene identifier, and call the corresponding model algorithm in the database, use the fire image data in each cluster as sample data, train the fireworks recognition model applicable to each scene according to the model algorithm corresponding to each applicable scene, fuse the trained fireworks recognition models, obtain the visual image of the target detection area, identify the scene features of the target detection area, and adjust the operation mode of the fireworks recognition model according to the scene features;

[0018] Step S22: Obtain the recognition result of the target detection area. When a flame is identified in the target area, anchor the flame position in the image, establish a coordinate system, identify the flame position coordinates and the sprinkler nozzle coordinates, and calculate the distance r between the flame position and the sprinkler nozzle by the distance formula. If the distance between the flame position and the sprinkler nozzle is greater than the minimum threshold and less than the maximum threshold, calculate the opening angle of the sprinkler system nozzle by the formula In the formula, θ indicates the opening angle of the sprinkler nozzle, H indicates the width of the flame in the image, and α indicates the influence coefficient of the flame width on the opening angle of the sprinkler nozzle. If the distance between the flame position and the sprinkler nozzle is less than the minimum threshold, the sprinkler nozzle will be fully opened, otherwise the sprinkler nozzle will not be started.

[0019] Step S23: When smoke is identified in the target area, the infrared temperature detection data of the target area is obtained to identify the temperature change of the target area. If the temperature change value of the target area is greater than the system set threshold, the target area is marked as a fire and the sprinkler system is started. Otherwise, the target area is marked as not a fire and the sprinkler system is not started.

[0020] According to the above technical solution, step S3 further includes the following steps:

[0021] Step S31: obtaining visual images of objects around the fire point, establishing a coordinate system, identifying the coordinates of the fire point and surrounding objects, calculating the distance L between the fire point and surrounding objects by a distance formula, and when the distance between the fire point and surrounding objects is less than a first threshold, starting the spray nozzle to spray the target object;

[0022] Step S32: When the distance between the fire point and the surrounding objects is greater than the first threshold and less than the second threshold, the visual image features of the surrounding objects are identified, and the corresponding cargo information in the database is retrieved according to the features of the objects, and the flammability coefficient μ and the corresponding value coefficient κ of the objects are identified, and the vulnerability of the target objects is calculated by the formula In the formula, Q represents the vulnerability of the target object. If the vulnerability of the target object is less than the minimum threshold set by the system, the system retrieves the corresponding sprinkler influence coefficient β from the database according to the vulnerability of the target object, and calculates the distance L between the drop point of the extinguishing agent sprayed by the sprinkler system and the sprinkler head through the formula 1 =β·L, where L 1 Indicates the distance between the fire extinguishing agent drop point and the nozzle of the sprinkler system. According to the distance between the fire extinguishing agent drop point and the nozzle of the sprinkler system, the corresponding water pressure and water pump power in the database are retrieved. Otherwise, the system starts the sprinkler system to spray at all angles.

[0023] Step S33: When the distance between the fire point and the surrounding objects is greater than the second threshold, the environmental data of the fire point is obtained, the ventilation environment of the fire point is further analyzed, and the spread trend of the fire and the degree of danger of the fire are predicted based on the analysis results.

[0024] According to the above technical solution, step S33 further includes the following steps:

[0025] Step S331: retrieve the flame image marked by the fireworks recognition model, extract the flame image according to the set frame rate cycle, overlap and compare the extracted flame images, mark the parts that cannot overlap, identify the pixel position of the marked part in each flame image, extract the pixel position features, fuse the position features, and obtain the spread trend of the fire. The system retrieves the vulnerability of the target object in the spread trend direction, and adjusts the sprinkler system according to the vulnerability;

[0026] Step S332: retrieve the wind direction and wind speed of the environment where the fire is located, and retrieve the corresponding influence coefficient η on the fire risk coefficient in the database according to the wind speed of the environment where the fire is located;

[0027] Step S333: Identify the wind direction of the environment where the fire point is located, retrieve the building model where the fire point is located, establish a coordinate system, identify the switch status of doors and windows in the building model, anchor the positions of the opened doors and windows in the building model, identify and mark the coordinates of the opened doors and windows, connect the coordinates of two adjacent opened doors and windows, scan the intersections of the connecting line segments in the building model and the building model, identify the number of intersections, and retrieve the corresponding influence coefficient λ of the hazard factor in the database according to the number of intersections;

[0028] Step S334: Obtain comprehensive information on flammable materials around the target detection area, retrieve the corresponding danger level T in the database based on the comprehensive information on the flammable materials, and calculate the danger coefficient of the target detection area by the formula P=η·λ·T, where P represents the danger coefficient of the target detection area.

[0029] According to the above technical solution, in step S4, the position information of the people who have not escaped from the building is obtained, all routes from the position of the trapped person to the exit are retrieved, and the risk factor in each route is identified. If there is an area in the route with a risk factor greater than a threshold, the route is eliminated. Otherwise, the system continues to detect, anchors the intersections in each route, and retrieves the corresponding number of passers-by in the database according to the position of the intersection. If the current number of people is greater than the number of passers-by, all trapped people passing through the current intersection are retrieved, and the system's alternative routes are identified. The route with fewer passers-by is selected by comparing the alternative routes, the position of the trapped person in the building model is obtained in real time, and the trapped person is commanded to evacuate through the communication module.

[0030] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: the present invention can accurately lock the fire area by calculating the distance between the nozzle and the flame and adjusting the opening angle of the nozzle according to the distance, and then accurately spray the fire extinguishing agent, thereby greatly improving the accuracy of the system; by verifying the temperature change of the target area when the system recognizes smoke, it can avoid the sprinkler system being affected by the smoke and the smoke temperature, resulting in poor spraying accuracy, thereby greatly improving the accuracy of the system; at the same time, it can avoid damage to some items, resulting in increased losses, and further reduce losses; by analyzing the flammability coefficient and value coefficient of the target item, the falling point of the fire extinguishing agent is calculated according to the distance between the target item and the fire point, and then the water pressure and the power of the water pump in the sprinkler system can be accurately controlled, greatly improving the accuracy of the system; by predicting the fire spread trend, analyzing the fragility of the items in the spread direction, and then adjusting the sprinkler equipment, the operating power of the sprinkler equipment can be accurately adjusted, further improving the accuracy of the system; by real-time positioning the position of the trapped personnel, it can command evacuation according to the position of the trapped personnel, avoiding the trapped personnel not understanding the building structure, resulting in low evacuation efficiency, thereby greatly improving the evacuation efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0032] Figure 1 It is a schematic diagram of the system module composition of the present invention. DETAILED DESCRIPTION

[0033] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0034] See also Figure 1 , the present invention provides a technical solution: a disaster emergency response system based on big data, including a data collection module, a fire warning module and a fire emergency module, characterized in that: the data collection module is used to collect visual data and environmental comprehensive data in the monitored target area in real time, and enter the comprehensive information of all people entering the building into the system; the fire warning module is used to detect whether a fire occurs in the target area, further analyze the spread trend and danger level of the fire, and adjust the operation mode of the sprinkler system according to the analysis results; the fire emergency module is used to obtain the position of trapped people in the building, divert the trapped people, and plan an evacuation route for each diversion group;

[0035] The environment detection module includes a visual analysis submodule, an air analysis submodule and a flammable material analysis submodule. The visual analysis submodule is used to analyze whether there is flame or smoke in the target detection area and control the sprinkler system according to the analysis results. The air analysis submodule is used to analyze the smoke concentration in the environment and adjust the sprinkler system based on the visual analysis results. The flammable material analysis submodule is used to analyze the flammable materials around the fire point.

[0036] The sprinkler management module includes a spread trend analysis submodule, a danger level analysis submodule and a fire sprinkler control submodule. The spread trend analysis submodule is used to analyze the spread trend of the fire and adjust the sprinkler system according to the spread trend. The danger level analysis submodule is used to analyze the danger level of the fire and adjust the evacuation strategy of trapped people according to the analysis results. The fire sprinkler control submodule is used to control the spraying mode of the sprinkler equipment according to the fire spread trend and the danger level of the fire.

[0037] The data collection module includes a data entry module, a vision module and a sensor module. The data entry module is used to enter the comprehensive information of people entering the building into the system. The vision module is used to collect visual images in the target detection area in real time. The sensor module is used to collect the temperature and smoke concentration of the surrounding environment in real time.

[0038] The fire warning module includes an environmental detection module, which is used to analyze the temperature, smoke concentration and whether there are flames in the target area, determine whether there is a fire based on the analysis results, and then anchor the fire location for spraying.

[0039] The fire warning module also includes a sprinkler management module, which is used to analyze the spread trend of the fire and the danger level of the fire, and adjust the operation mode of the sprinkler system according to the analysis results.

[0040] The fire emergency module includes a communication module, a route planning module and a personnel evacuation module. The communication module is used to comfort and guide the evacuation of trapped personnel. The route planning module is used to divert trapped personnel according to their locations and plan evacuation routes. The personnel evacuation module is used to select an evacuation strategy for trapped personnel based on the degree of fire danger.

[0041] The method of operation of the disaster emergency response system includes the following steps:

[0042] Step S1: Through the data entry module, the comprehensive information of all people entering the building and the annotated fire image data are entered into the system, the visual image of the target area is collected in real time through the visual module, and the comprehensive data of the surrounding environment is collected in real time through the sensor module;

[0043] Step S2: After the data is entered into the system, the system starts the environmental detection module, starts to analyze whether smoke or flames appear in the target area, analyzes the smoke concentration in the target area, and controls the start of the sprinkler system according to the analysis results;

[0044] Step S3: After the sprinkler system is started, the system begins to analyze the flammable materials around the fire point, the fire spread trend and the corresponding fire danger level, and further adjusts the operation mode of the sprinkler system according to the analysis results;

[0045] Step S4: When a fire is discovered, the system starts the fire emergency module, begins real-time monitoring of the fire, adjusts the personnel evacuation strategy according to the development trend of the fire, and plans evacuation routes to evacuate personnel.

[0046] Step S2 further comprises the following steps:

[0047] Step S21: Obtain the annotated fire image data, identify the applicable scene identifier in the image, cluster the fire image data according to the applicable scene identifier, and call the corresponding model algorithm in the database, use the fire image data in each cluster as sample data, train the fireworks recognition model applicable to each scene according to the model algorithm corresponding to each applicable scene, fuse the trained fireworks recognition models, obtain the visual image of the target detection area, identify the scene features of the target detection area, and adjust the operation mode of the fireworks recognition model according to the scene features, so that the operation mode of the fireworks recognition model can be more suitable for the application scene, thereby greatly increasing the accuracy of system detection;

[0048] Step S22: Obtain the recognition result of the target detection area. When a flame is identified in the target area, anchor the flame position in the image, establish a coordinate system, identify the flame position coordinates and the sprinkler nozzle coordinates, and calculate the distance r between the flame position and the sprinkler nozzle by the distance formula. If the distance between the flame position and the sprinkler nozzle is greater than the minimum threshold and less than the maximum threshold, calculate the opening angle of the sprinkler system nozzle by the formula In the formula, θ indicates the opening angle of the sprinkler nozzle, H indicates the width of the flame in the image, and α indicates the influence coefficient of the flame width on the opening angle of the sprinkler nozzle. If the distance between the flame position and the sprinkler nozzle is less than the minimum threshold, the sprinkler nozzle is fully opened, otherwise the sprinkler nozzle is not started. By calculating the distance between the sprinkler nozzle and the flame and adjusting the opening angle of the sprinkler nozzle according to the distance, the fire area can be accurately locked, and the fire extinguishing agent can be accurately sprayed, thereby greatly improving the accuracy of the system;

[0049] Step S23: When smoke is identified in the target area, the infrared temperature detection data of the target area is obtained, and the temperature change of the target area is identified. If the temperature change value of the target area is greater than the system-set threshold, the target area is marked as a fire and the sprinkler system is started. Otherwise, the target area is marked as not a fire and the sprinkler system is not started. By verifying the temperature change of the target area when the system identifies smoke, it is possible to avoid the sprinkler system being affected by the smoke and the smoke temperature, resulting in poor sprinkler accuracy, thereby greatly improving the accuracy of the system. At the same time, it is possible to avoid damage to some items, resulting in increased losses, and further reduce losses.

[0050] Step S3 further comprises the following steps:

[0051] Step S31: obtaining visual images of objects around the fire point, establishing a coordinate system, identifying the coordinates of the fire point and surrounding objects, calculating the distance L between the fire point and surrounding objects by a distance formula, and when the distance between the fire point and surrounding objects is less than a first threshold, starting the spray nozzle to spray the target object;

[0052] Step S32: When the distance between the fire point and the surrounding objects is greater than the first threshold and less than the second threshold, the visual image features of the surrounding objects are identified, and the corresponding cargo information in the database is retrieved according to the features of the objects, and the flammability coefficient μ and the corresponding value coefficient κ of the objects are identified, and the vulnerability of the target objects is calculated by the formula In the formula, Q represents the vulnerability of the target object. If the vulnerability of the target object is less than the minimum threshold set by the system, the system retrieves the corresponding sprinkler influence coefficient β from the database according to the vulnerability of the target object, and calculates the distance L between the drop point of the extinguishing agent sprayed by the sprinkler system and the sprinkler head through the formula 1 =β·L, where L 1 Indicates the distance between the fire extinguishing agent landing point and the nozzle of the sprinkler system. According to the distance between the fire extinguishing agent landing point and the nozzle of the sprinkler system, the corresponding water pressure and water pump power in the database are retrieved. Otherwise, the system starts the sprinkler system to spray at all angles. By analyzing the flammability coefficient and value coefficient of the target object, the landing point of the fire extinguishing agent is calculated according to the distance between the target object and the fire point, and then the water pressure and water pump power in the sprinkler system can be accurately controlled, which greatly improves the accuracy of the system;

[0053] Step S33: When the distance between the fire point and the surrounding objects is greater than the second threshold, the environmental data of the fire point is obtained, the ventilation environment of the fire point is further analyzed, and the spread trend of the fire and the degree of danger of the fire are predicted based on the analysis results.

[0054] Step S33 further includes the following steps:

[0055] Step S331: retrieve the flame image marked by the fireworks recognition model, extract the flame image according to the set frame rate cycle, overlap and compare the extracted flame images, mark the parts that cannot overlap, identify the pixel position of the marked part in each flame image, extract the pixel position features, fuse the position features, and obtain the spread trend of the fire. The system retrieves the vulnerability of the target object in the spread trend direction, adjusts the sprinkler system according to the vulnerability, predicts the fire spread trend, analyzes the vulnerability of the objects in the spread direction, and then adjusts the sprinkler equipment, which can accurately adjust the operating power of the sprinkler equipment and further improve the accuracy of the system.

[0056] Step S332: retrieve the wind direction and wind speed of the environment where the fire is located, and retrieve the corresponding influence coefficient η on the fire risk coefficient in the database according to the wind speed of the environment where the fire is located;

[0057] Step S333: Identify the wind direction of the environment where the fire point is located, retrieve the building model where the fire point is located, establish a coordinate system, identify the switch status of doors and windows in the building model, anchor the positions of the opened doors and windows in the building model, identify and mark the coordinates of the opened doors and windows, connect the coordinates of two adjacent opened doors and windows, scan the intersections of the connecting line segments in the building model and the building model, identify the number of intersections, and retrieve the corresponding influence coefficient λ of the hazard factor in the database according to the number of intersections;

[0058] Step S334: Obtain comprehensive information on flammable materials around the target detection area, retrieve the corresponding danger level T in the database based on the comprehensive information on the flammable materials, and calculate the danger coefficient of the target detection area by the formula P=η·λ·T, where P represents the danger coefficient of the target detection area.

[0059] In step S4, the position information of the people who have not escaped from the building is obtained, all routes from the position of the trapped person to the exit are retrieved, and the risk factor in each route is identified. If there is an area in the route with a risk factor greater than a threshold, the route is eliminated. Otherwise, the system continues to detect and anchor the intersections in each route. The corresponding number of passers-by in the database is retrieved according to the position of the intersection. If the current number of people is greater than the number of passers-by, all trapped people passing through the current intersection are retrieved, and the system's alternative routes are identified. Compared with the alternative routes, a route with a smaller number of passers-by is selected, and the position of the trapped person in the building model is obtained in real time. The trapped person is commanded to evacuate through the communication module. By locating the position of the trapped person in real time, the evacuation can be commanded according to the position of the trapped person, thereby avoiding the trapped person not understanding the building structure, resulting in low evacuation efficiency, thereby greatly improving the evacuation efficiency.

[0060] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0061] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A disaster emergency response system based on big data, comprising a data collection module, a fire warning module and a fire emergency module, characterized in that: The data collection module is used to collect visual data and comprehensive environmental data in the monitored target area in real time, and enter the comprehensive information of all people entering the building into the system. The fire warning module is used to detect whether there is a fire in the target area, further analyze the spread trend and danger level of the fire, and adjust the operation mode of the sprinkler system according to the analysis results. The fire emergency module is used to obtain the location of trapped people in the building, divert the trapped people, and plan an evacuation route for each diversion group; The fire warning module includes an environment detection module, which is used to analyze the temperature, smoke concentration and whether there is a flame in the target area, determine whether there is a fire according to the analysis results, and then anchor the fire location for spraying; The fire warning module also includes a sprinkler management module, which is used to analyze the spread trend of the fire and the danger level of the fire, and adjust the operation mode of the sprinkler system according to the analysis results; The environment detection module includes a visual analysis submodule, an air analysis submodule and a combustible material analysis submodule. The visual analysis submodule is used to analyze whether there is a flame or smoke in the target detection area and control the sprinkler system according to the analysis result. The air analysis submodule is used to analyze the smoke concentration in the environment and adjust the sprinkler system in combination with the visual analysis result. The combustible material analysis submodule is used to analyze the combustible materials around the fire point. The sprinkler management module includes a spread trend analysis submodule, a danger level analysis submodule and a fire sprinkler control submodule. The spread trend analysis submodule is used to analyze the spread trend of the fire and adjust the sprinkler system according to the spread trend. The danger level analysis submodule is used to analyze the danger level of the fire and adjust the evacuation strategy of the trapped personnel according to the analysis results. The fire sprinkler control submodule is used to control the spraying mode of the sprinkler equipment according to the fire spread trend and the danger level of the fire. The data collection module includes a data entry module, a visual module and a sensor module. The data entry module is used to enter the comprehensive information of people entering the building into the system, the visual module is used to collect visual images in the target detection area in real time, and the sensor module is used to collect the temperature and smoke concentration of the surrounding environment in real time; The fire emergency module includes a communication module, a route planning module and a personnel evacuation module. The communication module is used to comfort and guide the evacuation of trapped personnel. The route planning module is used to divert trapped personnel according to their positions and plan evacuation routes. The personnel evacuation module is used to select an evacuation strategy for trapped personnel according to the degree of fire danger. The operation method of the disaster emergency response system comprises the following steps: Step S1: Through the data entry module, the comprehensive information of all people entering the building and the annotated fire image data are entered into the system, the visual image of the target area is collected in real time through the visual module, and the comprehensive data of the surrounding environment is collected in real time through the sensor module; Step S2: After the data is entered into the system, the system starts the environmental detection module, starts to analyze whether smoke or flames appear in the target area, analyzes the smoke concentration in the target area, and controls the start of the sprinkler system according to the analysis results; Step S3: After the sprinkler system is started, the system begins to analyze the flammable materials around the fire point, the fire spread trend and the corresponding fire danger level, and further adjusts the operation mode of the sprinkler system according to the analysis results; Step S4: When a fire is discovered, the system starts the fire emergency module, starts real-time monitoring of the fire, adjusts the personnel evacuation strategy according to the development trend of the fire, and plans evacuation routes to evacuate personnel; The step S2 further comprises the following steps: Step S21: Obtain the annotated fire image data, identify the applicable scene identifier in the image, cluster the fire image data according to the applicable scene identifier, and call the corresponding model algorithm in the database, use the fire image data in each cluster as sample data, train the fireworks recognition model applicable to each scene according to the model algorithm corresponding to each applicable scene, fuse the trained fireworks recognition models, obtain the visual image of the target detection area, identify the scene features of the target detection area, and adjust the operation mode of the fireworks recognition model according to the scene features; Step S22: Obtain the identification result of the target detection area. When a flame is identified in the target area, anchor the flame position in the image, establish a coordinate system, identify the flame position coordinates and the sprinkler nozzle coordinates, and calculate the distance between the flame position and the sprinkler nozzle using the distance formula. If the distance between the flame position and the sprinkler nozzle is greater than the minimum threshold and less than the maximum threshold, the opening angle of the sprinkler system nozzle is calculated by the formula , where Indicates the opening angle of the sprinkler nozzle. Indicates the width of the flame in the image, Indicates the influence coefficient of flame width on the opening angle of the sprinkler nozzle. If the distance between the flame position and the sprinkler nozzle is less than the minimum threshold, the sprinkler nozzle will be fully opened, otherwise the sprinkler nozzle will not be started; Step S23: when smoke is identified in the target area, infrared temperature detection data of the target area is obtained to identify the temperature change of the target area. If the temperature change value of the target area is greater than the system set threshold, the target area is marked as a fire and the sprinkler system is activated. Otherwise, the target area is marked as not a fire and the sprinkler system is not activated. The step S3 further comprises the following steps: Step S31: Obtain visual images of objects around the fire point, establish a coordinate system, identify the coordinates of the fire point and surrounding objects, and calculate the distance between the fire point and surrounding objects using a distance formula , when the distance between the fire point and the surrounding objects is less than the first threshold, the spray nozzle is started to spray the target object; Step S32: When the distance between the fire point and the surrounding objects is greater than the first threshold and less than the second threshold, the visual image features of the surrounding objects are identified, and the corresponding cargo information in the database is retrieved according to the object features to identify the flammability coefficient of the objects. And the corresponding value coefficient , the vulnerability of the target item is calculated by the formula , where Identify the vulnerability of the target item. If the vulnerability of the target item is less than the minimum threshold set by the system, the system will retrieve the corresponding spray impact coefficient from the database based on the vulnerability of the target item. , the distance between the fire extinguishing agent drop point and the sprinkler nozzle is calculated by the formula , where Indicates the distance between the fire extinguishing agent drop point and the sprinkler nozzle of the sprinkler system. According to the distance between the fire extinguishing agent drop point and the sprinkler nozzle of the sprinkler system, the corresponding water pressure and water pump power in the database are retrieved. Otherwise, the system starts the sprinkler system to spray at all angles. Step S33: When the distance between the fire point and the surrounding objects is greater than the second threshold, the environmental data of the fire point is obtained, the ventilation environment of the fire point is further analyzed, and the spread trend of the fire and the degree of danger of the fire are predicted based on the analysis results.

2. A disaster emergency response system based on big data according to claim 1, characterized in that: The step S33 further comprises the following steps: Step S331: retrieve the flame image marked by the fireworks recognition model, extract the flame image according to the set frame rate cycle, overlap and compare the extracted flame images, mark the parts that cannot overlap, identify the pixel position of the marked part in each flame image, extract the pixel position features, fuse the position features, and obtain the spread trend of the fire. The system retrieves the vulnerability of the target object in the spread trend direction, and adjusts the sprinkler system according to the vulnerability; Step S332: retrieve the wind direction and wind speed of the environment where the fire is located, and retrieve the corresponding influence coefficient on the fire risk coefficient in the database according to the wind speed of the environment where the fire is located. ; Step S333: Identify the wind direction of the environment where the fire point is located, retrieve the building model where the fire point is located, establish a coordinate system, identify the switch status of doors and windows in the building model, anchor the positions of open doors and windows in the building model, identify and mark the coordinates of open doors and windows, connect the coordinates of two adjacent open doors and windows, scan the intersections of the connecting line segments in the building model and the building model, identify the number of intersections, and retrieve the corresponding risk factor influence coefficient in the database according to the number of intersections ; Step S334: Obtain comprehensive information about flammable objects around the target detection area, and retrieve the corresponding danger level in the database based on the comprehensive information about the flammable objects. , the risk factor of the target detection area is calculated by the formula , where Indicates the risk factor of the target detection area.

3. The disaster emergency response system based on big data according to claim 1, characterized in that: In the step S4, the position information of the persons who have not escaped from the building is obtained, all routes from the position of the trapped persons to the exit are retrieved, and the risk factor in each route is identified. If there is an area in the route with a risk factor greater than a threshold, the route is eliminated. Otherwise, the system continues to detect, anchors the intersections in each route, and retrieves the corresponding number of passers-by in the database according to the position of the intersection. If the current number of people is greater than the number of passers-by, all trapped persons passing through the current intersection are retrieved, and the system's alternative routes are identified. The route with fewer passers-by is selected by comparing the alternative routes, the position of the trapped persons in the building model is obtained in real time, and the trapped persons are commanded to evacuate through the communication module.

Images