Fire monitoring method and fire monitoring device

Abstract

The invention discloses a fire monitoring method and a fire monitoring device, and belongs to the technical field of fire monitoring. The fire monitoring method comprises the steps that at least one type of environment information of a target area at the current collection moment is acquired, and the types of the environment information comprise the type corresponding to meteorological information, the type corresponding to satellite remote sensing information and the type corresponding to unmanned aerial vehicle monitoring information; carrying out weighted prediction on various types of environment information to obtain a comprehensive fire danger index; and performing fire early warning on the target area based on the comprehensive fire danger index. According to the fire monitoring method, the fire danger monitoring cost can be reduced, the accuracy of the monitoring result can be improved, missing report and false report can be reduced, an area where a fire may occur or an area where a fire has occurred can be concerned in time, and huge loss caused by fire spreading due to the fact that the fire is not found is reduced.

Description

Technical Field

[0001] The present application belongs to the technical field of fire monitoring, and in particular, relates to a fire monitoring method and a fire monitoring device. Background Art

[0002] With global climate change and frequent natural disasters, fire has become a serious problem that endangers the ecological environment and human safety. In related technologies, manual inspections, thermal imaging cameras, and remote monitoring cameras are mainly used to inspect areas that need to be monitored around the clock to detect fires and issue early warnings in a timely manner. However, the above-mentioned fire monitoring and early warning methods have high monitoring costs, and the monitoring results may be false alarms or missed alarms, which can easily cause large losses. Summary of the invention

[0003] The present application aims to solve at least one of the technical problems existing in the related art. To this end, the present application proposes a fire monitoring method and a fire monitoring device, which can reduce the cost of fire risk monitoring, improve the accuracy of monitoring results, reduce the occurrence of missed reports and false reports, and can timely pay attention to areas where fires may occur or areas where fires have occurred, reducing the huge losses caused by the spread of fires due to undetected fires.

[0004] In a first aspect, the present application provides a fire monitoring method, the method comprising:

[0005] Acquire at least one type of environmental information of the target area at the current collection time, the types of environmental information including: types corresponding to meteorological information, types corresponding to satellite remote sensing information, and types corresponding to drone monitoring information;

[0006] Make weighted predictions on various environmental information to obtain a comprehensive fire risk index;

[0007] Based on the comprehensive fire risk index, fire warnings are issued for target areas.

[0008] According to the fire monitoring method of the present application, by acquiring meteorological information, satellite remote sensing information and drone monitoring information of the target area, the comprehensive fire risk index corresponding to the multiple types of environmental information at the current collection time is effectively determined, so that based on the actual value of the comprehensive fire risk index, the target area is warned, the cost of fire risk monitoring is reduced, the accuracy of monitoring results is improved, the probability of missed reports and false alarms is reduced, and timely attention can be paid to areas where fires may occur or areas where fires have occurred, reducing the huge losses caused by the spread of fires due to undetected fires.

[0009] According to the fire monitoring method of the present application, various types of environmental information are weighted and predicted to obtain a comprehensive fire risk index, including:

[0010] Obtain the fire risk index corresponding to each type of environmental information and the first weight corresponding to each type of environmental information;

[0011] The fire risk index corresponding to each type of environmental information is processed based on each first weight to obtain a comprehensive fire risk index.

[0012] According to the fire monitoring method of the present application, the fire risk index corresponding to various types of environmental information is obtained, including:

[0013] Classifying target environmental information among various types of environmental information to obtain at least one type of sub-environmental information;

[0014] Obtain the fire risk index corresponding to each type of sub-environment information and the second weight corresponding to each type of sub-environment information,

[0015] The fire risk index corresponding to each type of sub-environmental information is processed based on each second weight to obtain the fire risk index corresponding to the target type of environmental information.

[0016] According to the fire monitoring method of the present application, various types of environmental information are weighted and predicted to obtain a comprehensive fire risk index, including:

[0017] Various types of environmental information are input into the target model to obtain the comprehensive fire risk index output by the target model.

[0018] According to the fire monitoring method of the present application, the target model is trained based on the following steps:

[0019] Obtain multiple types of historical environmental information and the fire risk levels corresponding to each type of historical environmental information;

[0020] Based on multiple categories of historical environmental information and the fire risk levels corresponding to each category of historical environmental information, the target model is trained with the goal of outputting a comprehensive fire risk index, a first weight corresponding to each category of environmental information, and a second weight corresponding to at least one category of sub-environmental information included in each category of environmental information.

[0021] According to the fire monitoring method of the present application, the fire early warning for the target area based on the comprehensive fire risk index includes:

[0022] Based on the comprehensive fire risk index, determining the fire risk level corresponding to the target area;

[0023] Based on the fire risk level corresponding to the target area, a fire warning is issued to the target area.

[0024] According to the fire monitoring method of the present application, the fire early warning for the target area based on the fire risk level corresponding to the target area includes:

[0025] When the fire risk level corresponding to the target area is the target fire risk level, collecting image information of the target area;

[0026] Processing the image information through an image recognition model to obtain flame information in the image information;

[0027] Based on the flame information, a fire warning is issued to the target area.

[0028] According to the fire monitoring method of the present application, the step of processing the image information through an image recognition model to obtain flame information in the image information includes:

[0029] Performing at least one of Gaussian low-pass filtering, hue-saturation-brightness block homomorphic filtering color image enhancement, and image segmentation based on a K-means model in a brightness-color opposition dimension color space on the image information to obtain first image information;

[0030] The first image information is processed by an image recognition model to obtain the flame information in the first image information.

[0031] In a second aspect, the present application provides a fire monitoring device, the device comprising:

[0032] The first processing module is used to obtain at least one type of environmental information of the target area at the current collection time, and the types of environmental information include: types corresponding to meteorological information, types corresponding to satellite remote sensing information, and types corresponding to drone monitoring information;

[0033] The second processing module is used to perform weighted prediction on various environmental information to obtain a comprehensive fire risk index;

[0034] The third processing module is used to issue a fire warning to the target area based on the comprehensive fire risk index.

[0035] According to the fire monitoring device of the present application, by acquiring the meteorological information, satellite remote sensing information and drone monitoring information of the target area, the comprehensive fire risk index corresponding to the multiple types of environmental information at the current collection time can be effectively determined, so as to issue an early warning to the target area based on the actual value of the comprehensive fire risk index, thereby reducing the cost of fire risk monitoring, improving the accuracy of monitoring results, reducing the probability of missed reports and false alarms, and being able to pay timely attention to areas where fire may occur or where fire has already occurred, thereby reducing the huge losses caused by the spread of fire due to undetected fire.

[0036] In a third aspect, the present application provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the fire monitoring method as described in the first aspect above when executing the computer program.

[0037] In a fourth aspect, the present application provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the fire monitoring method as described in the first aspect above.

[0038] In a fifth aspect, the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the fire monitoring method as described in the first aspect above.

[0039] The above one or more technical solutions in the embodiments of the present application have at least one of the following technical effects:

[0040] By acquiring meteorological information, satellite remote sensing information and drone monitoring information in the target area, the comprehensive fire risk index corresponding to the multiple types of environmental information at the current collection time can be effectively determined. Based on the actual value of the comprehensive fire risk index, the target area can be warned, reducing the cost of fire risk monitoring, improving the accuracy of monitoring results, and reducing the probability of missed reports and false alarms. It is possible to pay timely attention to areas where fires may occur or where fires have already occurred, reducing the huge losses caused by the spread of fires due to undetected fires.

[0041] Furthermore, by obtaining different categories of environmental information and obtaining first weights that characterize the degree of influence of different categories of environmental information on the calculation of the comprehensive fire risk index, each category of environmental information is processed by the first weights corresponding to each category of environmental information. On the basis of considering the degree of influence of each category of environmental information on the calculation of the comprehensive fire risk index, different weights are assigned to each category of environmental information to improve the accuracy of the determined comprehensive fire risk index.

[0042] Furthermore, by acquiring multiple categories of historical environmental information and the historical fire risk levels corresponding to each category of historical environmental information, the target model is trained so that the target model can learn the correlation between the historical environmental information and the historical fire risk levels corresponding to each category of historical environmental information, thereby enabling the target model to output an accurate comprehensive fire risk index, the first weights corresponding to each category of environmental information and the second weights corresponding to each category of sub-environmental information, and the determined first weights corresponding to each category of environmental information and the second weights corresponding to each category of sub-environmental information are scientific and reasonable.

[0043] Furthermore, by comparing the correlation between the fire risk level of the target area and the given target fire risk level, the safety level of the target area can be effectively determined. When the fire risk of the target area is high, the inspection frequency of the area is increased, and image information of the target area is collected to increase attention to the target area, so as to timely discover and process the acquisition of the target area, reduce the speed of fire spread, and reduce unnecessary losses.

[0044] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which:

[0046] Figure 1 This is one of the flow charts of the fire monitoring method provided in the embodiment of the present application;

[0047] Figure 2 This is the second flow chart of the fire monitoring method provided in the embodiment of the present application;

[0048] Figure 3 is a schematic diagram of an interface of a fire monitoring method provided in an embodiment of the present application;

[0049] Figure 4 This is the third flow chart of the fire monitoring method provided in the embodiment of the present application;

[0050] Figure 5 This is the fourth flow chart of the fire monitoring method provided in the embodiment of the present application;

[0051] Figure 6 is a structural schematic diagram of a fire monitoring device provided in an embodiment of the present application;

[0052] Figure 7 It is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0053] The following will be combined with the drawings in the embodiments of the present application to clearly describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments in the present application belong to the scope of protection of this application.

[0054] The terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described here, and the objects distinguished by "first", "second", etc. are generally of one type, and the number of objects is not limited. For example, the first object can be one or more. In addition, "and / or" in the specification and claims represents at least one of the connected objects, and the character " / " generally indicates that the objects associated with each other are in an "or" relationship.

[0055] The fire monitoring method, fire monitoring device, electronic device and readable storage medium provided in the embodiments of the present application are described in detail below with reference to the accompanying drawings through specific embodiments and their application scenarios.

[0056] The fire monitoring method may be applied to a terminal, and may be specifically executed by hardware or software in the terminal.

[0057] The terminal includes but is not limited to portable communication devices such as mobile phones or tablet computers. It should also be understood that in some embodiments, the terminal may not be a portable communication device, but a desktop computer.

[0058] The terminal may include one or more other physical user interface devices such as a physical keyboard, mouse, and joystick.

[0059] The fire monitoring method provided in the embodiment of the present application may be executed by an electronic device or a functional module or functional entity in the electronic device that can implement the fire monitoring method. The electronic devices mentioned in the embodiment of the present application include but are not limited to mobile phones, tablet computers, computers, cameras, and wearable devices. The fire monitoring method provided in the embodiment of the present application is described below using an electronic device as an example of an execution subject.

[0060] like Figure 1 As shown, the fire monitoring method includes: step 110, step 120 and step 130.

[0061] Step 110: Obtain at least one type of environmental information of the target area at the current collection time;

[0062] In this step, the target area is the area to be monitored for fire.

[0063] Environmental information is the environment-related information required to judge the fire situation in the target area.

[0064] In the present application, the acquired environmental information may be divided into different categories based on different acquisition methods.

[0065] The categories of environmental information include: categories corresponding to meteorological information, categories corresponding to satellite remote sensing information, and categories corresponding to drone monitoring information.

[0066] Meteorological information refers to the meteorological data measured by meteorological observation stations.

[0067] Satellite remote sensing information is information obtained by long-distance detection and data collection of target areas through sensors carried by satellites.

[0068] Drone monitoring information is information obtained through real-time monitoring and data collection of multiple target areas using various sensors and equipment carried by drones.

[0069] Environmental information can be obtained in a number of different ways.

[0070] In the actual implementation process, meteorological data can be obtained from the meteorological station in the monitoring target area.

[0071] Obtain satellite remote sensing information of the target area from the satellite platform corresponding to the target area.

[0072] Based on the specific category of the required data, you can choose a drone that can collect the required category of environmental information, or deploy a sensor that can collect the required category of environmental information on the drone, and deploy the drone at different locations in the target area so that the drone monitors the target area in real time to obtain the drone monitoring information of the target area.

[0073] In some embodiments, at least one type of environmental information of the target area at the current collection moment can be collected through GPRS and a high-precision long-distance sensor system.

[0074] like Figure 2 As shown, multiple monitoring nodes can be set up in the target area, and various environmental information of the target area can be collected through modern high-speed and high-precision sensors, and connected to the sending module through a serial interface to send the collected various environmental information to the monitoring center server.

[0075] In the actual implementation process, the primary processing of various environmental information is realized inside the single-chip microcomputer, and then the single-chip microcomputer drives the GPRS module SIM300 to use its embedded GPRSDTU function to connect to the public Internet through the GPRS wireless network, and finally establish a data path with the fire monitoring center computer on the Internet, and use TCP / IP technology to realize network communication, thereby realizing fast and efficient wireless data communication between each monitoring node and the monitoring center server.

[0076] In some embodiments, solar panels may be provided to provide power to each monitoring node and monitoring center server and other equipment.

[0077] In some embodiments, each monitoring node and the monitoring central server can communicate with each other through a wireless data communication module.

[0078] In this embodiment, the wireless data communication module can select components with General Packet Radio Service (GPRS) and Global System for Mobile Communications (GSM) functions to perform GPRS transmission and send and receive text messages.

[0079] In other embodiments, the wireless data communication module may also use a component with a short-range data information receiving and sending function.

[0080] In this embodiment, by providing a component with a short-range data information receiving and sending function, the problem that there are sub-areas in the target area without GPRS signals and data cannot be effectively received can be considered, thereby improving the comprehensiveness of data acquisition.

[0081] According to the fire monitoring system provided in the embodiment of the present application, various environmental information of the target area can be obtained in real time through GPRS and high-precision long-distance sensors, and the acquired data can be quickly uploaded to the monitoring center server, effectively providing uninterrupted monitoring services for the target area, and improving the stability of the monitoring process and the comprehensiveness of monitoring.

[0082] In some embodiments, the required environmental information can be obtained through the front end through image sensors, drones, and weather stations.

[0083] In other embodiments, different data acquisition devices may be arranged in the target area, such as a temperature sensor, a humidity sensor, and a smoke sensor, etc., to collect various types of environmental information.

[0084] Of course, multiple types of environmental information of the target area at the current collection time can also be obtained in any feasible way, and this application is not limited to this.

[0085] In some embodiments, after obtaining at least one type of environmental information, data preprocessing may be performed on the at least one type of environmental information.

[0086] In this embodiment, data preprocessing includes but is not limited to: data cleaning, data correction, and data standardization.

[0087] Among them, data cleaning can include: missing value processing and outlier processing.

[0088] In the actual implementation process, missing data values ​​can be filled by using mean interpolation or nearest neighbor interpolation.

[0089] Outliers can be identified and handled using the interquartile range or standard deviation method.

[0090] In the actual execution process, the environmental information at the current collection moment can be corrected by using the environmental information at multiple historical moments before the current collection moment to eliminate errors caused by changes in sensors or the environment.

[0091] Data normalization is to normalize all data so that the data falls into the range of [0-1].

[0092] According to the fire monitoring method provided in the embodiment of the present application, data quality and data accuracy are effectively improved by performing data preprocessing on various types of environmental data obtained.

[0093] Step 120, performing weighted prediction on various types of environmental information to obtain a comprehensive fire risk index;

[0094] In this step, the comprehensive fire risk index is a numerical value used to determine the probability of fire in the target area.

[0095] It is understandable that the degree to which environmental information affects the determination of the fire risk situation in the target area may vary depending on the type of environmental information.

[0096] In the actual implementation process, the different degrees of influence of multiple types of environmental information on determining the fire risk situation in the target area can be considered, and weighted predictions can be made on multiple types of environmental data to obtain a comprehensive fire risk index for the target area.

[0097] In some embodiments, step 120 may further include:

[0098] Obtain the fire risk index corresponding to each type of environmental information and the first weight corresponding to each type of environmental information;

[0099] The fire risk index corresponding to each type of environmental information is processed based on each first weight to obtain a comprehensive fire risk index.

[0100] In this embodiment, the fire risk index corresponding to the environmental information is a quantitative value of the fire risk level determined based on the actual situation of various types of environmental information.

[0101] It is understandable that different types of environmental information may correspond to different fire risk indices.

[0102] The first weight corresponding to the environmental information is the degree of influence of the environmental information on the calculation of the comprehensive fire risk index.

[0103] In the actual implementation process, the comprehensive fire risk index can be calculated by the following formula:

[0104] I fusion =α·I opt +β·I R (R)+γ·I D (D)

[0105] Among them, I fusion is the comprehensive fire risk index; α is the first weight corresponding to the meteorological information; I opt is the fire risk index corresponding to meteorological information; β is the first weight corresponding to satellite remote sensing information; I R is the fire risk index corresponding to satellite remote sensing information; R is the rainfall; γ is the first weight corresponding to the UAV monitoring information; I Dis the fire risk index corresponding to the drone monitoring information; D is the number of days the drought lasts.

[0106] According to the fire detection method provided in the embodiment of the present application, different categories of environmental information are obtained, and first weights that characterize the degree of influence of different categories of environmental information on the calculation of the comprehensive fire risk index are obtained. Various types of environmental information are processed according to the first weights corresponding to various types of environmental information. On the basis of considering the degree of influence of various types of environmental information on the calculation of the comprehensive fire risk index, different weights are assigned to various types of environmental information to improve the accuracy of the determined comprehensive fire risk index.

[0107] In some embodiments, obtaining the fire risk index corresponding to each type of environmental information may also include:

[0108] Classifying target environmental information among various types of environmental information to obtain at least one type of sub-environmental information;

[0109] Obtaining fire risk indexes corresponding to various sub-environment information and second weights corresponding to various sub-environment information;

[0110] The fire risk index corresponding to each type of sub-environmental information is processed based on each second weight to obtain the fire risk index corresponding to the target type of environmental information.

[0111] In this embodiment, the target type of environmental information is any type of environmental information in at least one type of environmental information.

[0112] The target category environment information may include multiple subcategories.

[0113] When the target environmental information is meteorological information, the subcategories included in the target environmental information may include: temperature, precipitation, wind speed, relative humidity, and consecutive rainless days.

[0114] The temperature is the temperature data at the target time of each day, in degrees Celsius (°C), for example, the temperature data at 14:00 p.m. each day.

[0115] The precipitation is the daily precipitation in millimeters (mm), for example, the precipitation at 14:00 every afternoon.

[0116] The wind speed is the wind speed at the target time of the day, in meters per second (m / s), for example, the wind speed at 14:00 p.m. every day.

[0117] The relative humidity is the relative humidity at the target time of the day, in percentage (%), for example, the relative humidity at 14:00 p.m. every day.

[0118] The number of consecutive rainless days is the number of consecutive days without rainfall recorded.

[0119] The specific value of the target time can be customized by the user or determined based on actual conditions, and this application does not limit it.

[0120] When the target environmental information is satellite remote sensing information, the subcategories of the target environmental information may include: surface temperature, vegetation coverage, humidity data, etc.

[0121] Among them, the surface temperature is the surface temperature data obtained through satellite remote sensing, which can supplement the deficiency of ground meteorological data.

[0122] Vegetation cover reflects the density of surface vegetation cover and can be used to assess fire susceptibility.

[0123] Humidity data is obtained based on remote sensing technology.

[0124] When the target environmental information is drone monitoring information, the subcategories of the target environmental information may include: real-time high-resolution images and environmental temperature data.

[0125] Among them, the real-time high-resolution image is an image obtained by real-time monitoring of the fire risk situation in the target area.

[0126] The ambient temperature data refers to the surface temperature data collected by the sensors carried by the drone.

[0127] Sub-environmental information is environmental information obtained by further dividing various types of environmental information.

[0128] The fire risk index corresponding to the sub-environment information is a quantitative value of the fire risk level determined based on the actual situation of the sub-environment information.

[0129] During the actual implementation process, a table describing the association between various types of sub-environment information and the fire risk index corresponding to the various types of sub-environment information may be pre-constructed.

[0130] In case that the fire risk index corresponding to the sub-environment information is used, the fire risk index corresponding to the required sub-environment information can be obtained by searching a pre-built table.

[0131] For example, find the fire risk index corresponding to temperature and the fire risk index corresponding to precipitation, etc.

[0132] The second weight corresponding to the sub-environment information is the degree of influence of the sub-environment information on the calculation of the comprehensive fire risk index.

[0133] It is understandable that different types of environmental information may include different sub-environmental information.

[0134] The fire risk index corresponding to the target category environmental information is a fire risk index calculated based on environmental information of each subcategory included in the target category environmental information.

[0135] For example, the target environmental information is meteorological information, and the subcategories of meteorological information include temperature, precipitation, wind speed, and relative humidity.

[0136] In the actual implementation process, the following formula can be used:

[0137] I opt =w T (T)·I T (T)+w r (r)·I r (r)+w v (v) l v (v)+w u (u)·I u (u)

[0138] Calculate the fire risk index corresponding to the meteorological information.

[0139] Among them, w T (T) is the second weight corresponding to the temperature; I T (T) is the fire risk index corresponding to the temperature; w r (r) is the second weight corresponding to the precipitation; r (r) is the fire risk index corresponding to the precipitation; w v (v) is the second weight corresponding to the wind speed; v (v) is the fire risk index corresponding to the temperature; w u (u) is the second weight corresponding to the relative humidity; u (u) is the fire danger index corresponding to relative humidity.

[0140] In the actual implementation process, each type of environmental information includes multiple subcategories. The fire risk index corresponding to the target type of environmental information can be calculated based on the fire risk index of each type of sub-environmental information. After obtaining the fire risk index corresponding to all categories of environmental information, the fire risk index corresponding to each category of environmental information can be integrated to obtain a comprehensive fire risk index.

[0141] Taking the meteorological information obtained at the current collection time as follows: temperature T = 25. C, precipitation r = 0 mm, wind speed v = 5 m / s and relative humidity u = 40%, the second weight corresponding to the temperature is 0.4, the second weight corresponding to the precipitation is 0.3, the second weight corresponding to the wind speed is 0.2, and the second weight corresponding to the relative humidity is 0.1 as an example, the following formula can be used:

[0142] I opt =0.4·I T (25)+0.3·I r (0)+0.2·I v (5)+0.1·Iu (40)

[0143] Calculate the fire risk index corresponding to the meteorological information.

[0144] Among them, I opt It is the fire danger index corresponding to the meteorological information.

[0145] Similarly, the fire risk index corresponding to the satellite remote sensing information and the fire risk index corresponding to the drone monitoring information can also be calculated in the same way, which will not be elaborated here.

[0146] After obtaining the fire risk index corresponding to the meteorological information, the fire risk index corresponding to the satellite remote sensing information, and the fire risk index corresponding to the drone monitoring information, the comprehensive fire risk index can be further calculated.

[0147] Taking the first weight corresponding to meteorological information as 0.6, the first weight corresponding to satellite remote sensing information as 0.3, and the first weight corresponding to drone monitoring information as 0.1 as an example, the following formula can be used:

[0148] I fusion =0.6·I opt +0.3·I R (20)+0.1·I D (30)

[0149] The comprehensive fire risk index is calculated.

[0150] Among them, I fusion is the comprehensive fire risk index, I opt is the fire risk index corresponding to the meteorological information; I R is the fire risk index corresponding to satellite remote sensing information; I D It is the fire risk index corresponding to the drone monitoring information.

[0151] In the process of calculating the comprehensive fire risk index, satellite remote sensing information can further consider rainfall, such as rainfall of 20, and drone monitoring information can further consider the number of days of drought, such as 30 consecutive days of drought, to obtain the final comprehensive fire risk index.

[0152] According to the fire risk monitoring method provided in the embodiment of the present application, various types of environmental information are further subdivided to obtain sub-environmental information included in each type of environmental information, and the fire risk index corresponding to each type of environmental information is calculated through each type of sub-environmental information and the second weights corresponding to each type of sub-environmental information, thereby improving the accuracy of the fire risk index corresponding to each type of environmental information, thereby improving the accuracy of the comprehensive fire risk index calculated based on the fire risk index corresponding to each type of environmental information.

[0153] Step 130: Based on the comprehensive fire risk index, a fire warning is issued to the target area.

[0154] In this step, it can be understood that the probability of fire occurring in the target area varies with the size of the comprehensive fire risk index.

[0155] In the actual implementation process, it can be determined whether to issue an early warning to the target area based on the actual value of the comprehensive fire risk index.

[0156] In some embodiments, when the comprehensive fire risk index is greater than a target threshold, a fire warning may be issued to the target area.

[0157] In this embodiment, the target threshold is a preset value that determines a high probability of fire in the target area.

[0158] The specific value of the target threshold can be customized by the user or determined based on actual conditions, and this application does not limit it.

[0159] In the actual implementation process, taking the comprehensive fire risk index as 10 and the target threshold as 20 as an example, the comprehensive fire risk index is less than the target threshold. At this time, the probability of fire in the target area is small, and there is no need for fire warning. The target area can continue to be monitored.

[0160] Taking the comprehensive fire risk index as 25 and the target threshold as 20 as an example, the comprehensive fire risk index is greater than the target threshold. At this time, a fire warning can be issued for the target area.

[0161] In some embodiments, providing a fire warning to a target area may include:

[0162] The output is used to instruct the start of early warning inspection of the target area.

[0163] In this embodiment, after determining to issue a fire warning, an administrator of the target area may be arranged to patrol the target area, and / or a drone may be dispatched to patrol the target area, and the target area may be monitored in real time through an image sensor.

[0164] During the research and development process, the inventors discovered that fire has become a serious problem that endangers the ecological environment and human safety; in related technologies, manual inspections, thermal imaging cameras, and remote monitoring cameras are mainly used to patrol the areas that need to be monitored around the clock to detect fires and issue early warnings in a timely manner; however, the above-mentioned fire monitoring and early warning methods have high monitoring costs, and the monitoring results may be false alarms or missed alarms, which can easily cause large losses.

[0165] This application can obtain the meteorological information, satellite remote sensing information and drone monitoring information of the target area in real time, so as to have a more comprehensive understanding of the actual situation of the target area, provide detailed data support for subsequent fire warning, and make weighted predictions on various environmental information, consider the different effects of various environmental information on the calculation of the comprehensive fire risk index, and improve the accuracy of the calculated comprehensive fire risk index, so as to effectively judge the fire risk situation represented by the environmental information at the current collection time, so as to give a fire warning to the target area based on the comprehensive fire risk index, and effectively improve the accuracy of the monitoring results;

[0166] In addition, by acquiring multiple types of environmental information in real time and automatically calculating the fire risk index of the target area for fire risk warning, the target area can be monitored in real time without manual round-the-clock inspections, and effective fire monitoring can reduce costs.

[0167] According to the fire monitoring method provided in the embodiment of the present application, by acquiring meteorological information, satellite remote sensing information and drone monitoring information of the target area, the comprehensive fire risk index corresponding to the multiple types of environmental information at the current collection time is effectively determined, so that based on the actual value of the comprehensive fire risk index, the target area is warned, the cost of fire risk monitoring is reduced, the accuracy of monitoring results is improved, the probability of missed reports and false alarms is reduced, and timely attention can be paid to areas where fire may occur or areas where fire has occurred, reducing the huge losses caused by the spread of fire due to undetected fire.

[0168] In some embodiments, step 120 may further include:

[0169] Various types of environmental information are input into the target model to obtain the comprehensive fire risk index output by the target model.

[0170] In this embodiment, the target model is a model for calculating a comprehensive fire risk index of a target area.

[0171] In some embodiments, the target model can be trained based on the following steps:

[0172] Obtain multiple types of historical environmental information and the fire risk levels corresponding to each type of historical environmental information;

[0173] Based on multiple categories of historical environmental information and the fire risk levels corresponding to each category of historical environmental information, the target model is trained with the goal of outputting a comprehensive fire risk index, a first weight corresponding to each category of environmental information, and a second weight corresponding to at least one category of sub-environmental information included in each category of environmental information.

[0174] In this embodiment, the historical environment information includes historical environment information of the target area, historical environment information of multiple areas with similar terrain and geographical structure to the target area, and historical environment information of areas surrounding the target area.

[0175] The distance between the target area and the surrounding area can be determined based on user customization or actual conditions, and this application does not limit it.

[0176] Historical environmental information refers to various environmental information before the current collection time.

[0177] Historical environmental information can be obtained through the meteorological databases of regional meteorological units, meteorological data service centers, and the website of the World Meteorological Organization.

[0178] Of course, historical environmental information may also be obtained through any other feasible method, which is not limited in this application.

[0179] The fire risk level corresponding to the historical environmental information is the actual fire risk situation of the area corresponding to the historical environmental information at the time of collection of the historical environmental information.

[0180] In the actual implementation process, the various types of historical environmental information obtained can be divided into meteorological information, satellite remote sensing information and drone monitoring information.

[0181] For example, when multiple categories of historical environmental information are divided into meteorological information, satellite remote sensing information, and drone monitoring information, historical meteorological information, historical satellite remote sensing information, and historical drone monitoring information are obtained respectively.

[0182] In some embodiments, historical meteorological information, historical satellite remote sensing information, and historical drone monitoring information at the same collection time are grouped together, and the historical fire risk level corresponding to the group of data is obtained.

[0183] In some embodiments, the target training model may include a target first sub-model and a target second sub-model.

[0184] In some embodiments, based on multiple types of historical environmental information and fire risk levels corresponding to each type of historical environmental information, with the goal of outputting a comprehensive fire risk index, a first weight corresponding to each type of environmental information, and a second weight corresponding to at least one type of sub-environmental information included in each type of environmental information, training a target model may include:

[0185] Based on multiple types of historical environmental information and the fire risk levels corresponding to each type of historical environmental information, the first target sub-model is trained with the goal of outputting a comprehensive fire risk index and the first weights corresponding to each type of environmental information;

[0186] Based on multiple types of historical environmental information and fire risk levels corresponding to each type of historical environmental information, a target second sub-model is trained with the goal of outputting a fire risk index corresponding to each type of historical environmental information and a second weight corresponding to at least one type of sub-environmental information included in each type of environmental information;

[0187] A target model is obtained based on the target first sub-model and the target second sub-model.

[0188] In this embodiment, the target first sub-model and the target second sub-model may be neural network models or machine learning models, etc.

[0189] The following is an example in which the target first sub-model is a multiple regression model.

[0190] Based on at least one type of historical environmental information and the historical fire risk levels corresponding to each type of historical environmental information, a multivariate regression model is trained to determine the first weights corresponding to each type of environmental information.

[0191] In this embodiment, the first weight corresponding to each type of environmental information can be obtained by training the target first sub-model through the following formula:

[0192] w f =(Z T Z) -1 Z T y

[0193] Among them, w f is the first weight corresponding to each type of environmental information; Z is the matrix composed of various types of historical environmental information, Z = [X, R, D], X is the historical meteorological information, R is the historical satellite remote sensing information, and D is the historical UAV monitoring information; y is the historical fire risk level corresponding to various types of historical environmental information.

[0194] In some embodiments, historical environmental information and various types of environmental information may also be used to form sample data of the target first sub-model.

[0195] In the actual implementation process, the historical environmental information can be processed through the above formula to obtain the first weight corresponding to each type of environmental information.

[0196] In some embodiments, based on multiple types of historical environmental information and fire risk levels corresponding to each type of historical environmental information, with the goal of outputting a fire risk index corresponding to each type of historical environmental information and a second weight corresponding to at least one type of sub-environmental information included in each type of environmental information, training the target second sub-model may include:

[0197] further dividing at least one category of historical environmental information into multiple categories of historical sub-environmental information;

[0198] Based on multiple categories of historical sub-environmental information and the historical fire risk levels corresponding to each category of historical sub-environmental information, the target second sub-model is trained with the goal of outputting the fire risk index corresponding to each category of historical environmental information and the second weight corresponding to at least one category of sub-environmental information included in each category of environmental information.

[0199] In this embodiment, in the actual execution process, the second weight corresponding to each type of sub-environment information can be obtained by training the target second sub-model through the following formula:

[0200] w=(X T X) -1 X T y

[0201] Among them, w is the second weight corresponding to each type of sub-environment information; X is the historical sub-environment information; y is the historical fire risk level corresponding to the historical sub-environment information.

[0202] In some embodiments, a regression model such as a random forest, gradient boosting tree or XGBoost (ie, a target second sub-model) can be trained based on historical sub-environment information to obtain second weights corresponding to each type of sub-environment information.

[0203] In other embodiments, regression models such as random forest, gradient boosting tree or XGBoost can be fused to obtain a fused regression model (i.e., the target second sub-model). The fused regression model can be trained based on historical sub-environment information to obtain the second weights corresponding to each type of sub-environment information.

[0204] In some embodiments, based on the different numerical values ​​corresponding to the historical sub-environment information, the second weights corresponding to each type of sub-environment information can be dynamically adjusted through a Bayesian optimization algorithm, so that the fire risk index corresponding to the target type environmental information obtained based on the second weights corresponding to each type of sub-environment information is more accurate.

[0205] In this embodiment, the following formula can be used:

[0206]

[0207] Dynamically adjust the second weight corresponding to each sub-environment information.

[0208] Among them, w t is the second weight corresponding to each type of sub-environment information; f(w) is the fire risk index model; D is the historical sub-environment information.

[0209] According to the fire risk detection method provided in the embodiment of the present application, by subdividing various types of historical environmental information, multiple historical sub-environmental information included in each historical environmental information is obtained, and the target model is trained with multiple historical sub-environmental information, so that the target model outputs the second weight corresponding to the sub-environmental information, thereby improving the accuracy of the second weight corresponding to the acquired sub-environmental information.

[0210] In the actual execution process, the target first sub-model and the target second sub-model obtained by training are integrated to obtain the target model.

[0211] According to the fire risk detection method provided in the embodiment of the present application, by acquiring multiple types of historical environmental information and the historical fire risk levels corresponding to each type of historical environmental information, the target model is trained so that the target model can learn the correlation between the historical environmental information and the historical fire risk levels corresponding to each type of historical environmental information, so that the target model can output an accurate comprehensive fire risk index, the first weights corresponding to each type of environmental information and the second weights corresponding to each type of sub-environmental information, and the determined first weights corresponding to each type of environmental information and the second weights corresponding to each type of sub-environmental information are scientific and reasonable.

[0212] In the actual implementation process, the comprehensive fire risk index of the target area can be determined jointly through the specific data of various types of environmental information and the degree of influence of various types of environmental information on judging the fire risk situation in the target area.

[0213] In some embodiments, the dependency relationship between various types of environmental data can be constructed through a Bayesian network, and the fusion of various types of environmental data can be achieved through conditional probability.

[0214] In some embodiments, various types of environmental information can be input into the neural network model, and a comprehensive fire risk index can be obtained through the network fusion layer.

[0215] For example, in the actual implementation process, a multi-input model is used, and meteorological information, satellite remote sensing information, and drone monitoring information are used as inputs to the neural network model, which are fused through the fully connected layer to obtain the output comprehensive fire risk index.

[0216] In the process of processing various types of environmental information, graph neural networks can also be used to process data with complex spatial relationships and realize the fusion analysis of various types of environmental information from different sources.

[0217] According to the fire risk monitoring method provided in the embodiment of the present application, by inputting various types of environmental information into the target model, the comprehensive fire risk index corresponding to the target area can be quickly obtained without the need for manual calculation by the user, thereby reducing time consumption, improving fire risk monitoring efficiency, and improving the user experience.

[0218] It is understandable that, in the actual implementation process, it is necessary to obtain multiple types of environmental information in real time and process the obtained multiple types of environmental information in real time to understand the fire risk situation in the target area.

[0219] In some embodiments, step 120 may further include:

[0220] Through the stream processing framework, at least one type of environmental information at the current collection moment is processed in real time to obtain the comprehensive fire risk index of the target area.

[0221] In this embodiment, during the actual execution process, stream processing frameworks such as Apache Kafka and Apache Flink can be used to process the continuously collected multiple types of environmental information in real time, and calculate the comprehensive fire risk index corresponding to the multiple types of environmental information at the current collection moment.

[0222] In the actual execution process, a target time period can be set, and the multiple types of environmental information within the target time period can be treated as a batch of data and processed in the same data window to improve the response speed and processing efficiency of the multiple types of environmental information obtained.

[0223] The specific duration of the target time period can be determined based on actual conditions and is not limited in this application.

[0224] In some embodiments, the comprehensive fire danger index may be processed based on the following formula:

[0225]

[0226] Among them, I fusion (t i ) is the comprehensive fire risk index corresponding to at least one type of environmental information at the current collection moment; I stream (t) is the comprehensive fire risk index corresponding to the period t, t i In the time period t; n is the data window size.

[0227] According to the fire risk monitoring method provided in the embodiment of the present application, multiple types of environmental data acquired can be processed in real time through a stream processing framework, and the acquired data can be quickly responded to and processed to obtain a comprehensive fire risk index, thereby improving data processing efficiency and understanding the fire risk situation in the target area in real time.

[0228] In some embodiments, step 120 may further include:

[0229] Through the target calculation framework, at least one type of environmental information at the current collection moment is processed to obtain a comprehensive fire risk index.

[0230] In this embodiment, the target computing framework is a framework that processes at least one type of environmental information.

[0231] The target computing framework includes, but is not limited to, computing frameworks such as Compute Unified Device Architecture (CUDA) and Open Computing Language (OpenCL).

[0232] In the actual execution process, the target calculation framework can use the Graphics Processing Unit (GPU) to accelerate the speed of calculating the comprehensive fire risk index of the target area based on multiple types of environmental information and environmental information at the current acquisition moment.

[0233] In the actual implementation process, for example, when the target environmental information is meteorological information, the fire risk index corresponding to the meteorological information can be calculated based on the following formula:

[0234]

[0235] Among them, I parallel is the fire danger index corresponding to the calculated meteorological information, I opt is the fire risk index corresponding to the meteorological information; N is the number of calculation units.

[0236] According to the fire risk monitoring method provided in the embodiment of the present application, by calculating the fire risk index corresponding to the target type environmental information in various types of environmental information, the calculation efficiency of the fire risk index corresponding to the target type environmental information is effectively improved, the fire risk situation of the target area can be quickly obtained, and the user experience is improved.

[0237] In some embodiments, step 130 may further include:

[0238] Based on the comprehensive fire danger index, determine the fire danger level corresponding to the target area;

[0239] Based on the fire risk level corresponding to the target area, a fire warning is issued to the target area.

[0240] In this embodiment, the fire risk level is a warning level of a possible fire in the target area.

[0241] The specific number of fire risk levels included can be based on user customization or determined based on actual conditions. For example, the fire risk levels can be divided into 3 levels or 5 levels, which is not limited in this application.

[0242] In some embodiments, the fire risk level may be divided based on reference factors such as the comprehensive fire risk index, meteorological bureau regulations, and fire risk classification rules, and the degree of danger and prevention requirements corresponding to each fire risk level may be determined.

[0243] As shown in Table 1, fire risk levels can be divided into five levels.

[0244] Table 1

[0245]

[0246] In the actual implementation process, the fire risk level corresponding to the target area can be determined based on the range of the comprehensive fire risk index.

[0247] For example, when the fire risk index is 25, the fire risk level corresponding to the target area is determined to be level 4.

[0248] In some embodiments, the target area may be further divided into multiple grids, and a comprehensive fire risk index of each grid may be calculated based on multiple types of environmental information of each grid, and a fire warning may be issued based on the comprehensive fire risk index of each grid.

[0249] like Figure 3 As shown, taking the forest area of ​​City A, which is more than 4 million mu, as an example, the forest area of ​​City A accounts for B% of the total land area of ​​the city, while the forest area of ​​County C in City A accounts for D% of the forest area of ​​City A. The forest area of ​​County C can be used as a key defense area.

[0250] City A may be located in a temperate continental monsoon climate, with more rainfall from May to October each year, and less rainfall from October to May of the following year. Therefore, October to May of the following year can be designated as the annual forest fire prevention and control season.

[0251] In the actual implementation process, the forest area of ​​City A can be divided into multiple grids. Based on the real-time data of various environmental information of each grid, the comprehensive fire risk index of each grid can be calculated to determine the fire risk level corresponding to different grids in the forest area of ​​City A.

[0252] In the actual implementation process, after the fire risk level of the target area is determined, the fire risk level can be fed back to the user in a timely manner.

[0253] Taking the fire risk index of 10 as an example, it can be seen from the pre-divided fire risk level table that when the fire risk index is 10, the corresponding fire risk level is 2, and the target area can continue to be monitored.

[0254] Taking the fire risk index of 35 as an example, it can be seen from the pre-divided fire risk level table that when the fire risk index is 35, the corresponding fire risk level is 5. An alarm message can be output to the administrator of the target area, indicating that the target area is extremely flammable and a fire is likely to occur.

[0255] According to the fire monitoring method provided in the embodiment of the present application, the fire risk level of the target area can be effectively determined by calculating the comprehensive fire risk index, so that the subsequent early warning actions can be determined based on the actual situation of the fire risk level. When the fire risk level is high, the administrator of the target area can be reminded in time to pay attention to the fire risk situation in the target area, take countermeasures in time, and reduce unnecessary losses.

[0256] In some embodiments, based on the fire risk level corresponding to the target area, providing a fire warning to the target area may further include:

[0257] When the fire risk level corresponding to the target area is the target fire risk level, collecting image information of the target area;

[0258] Process the image information through the image recognition model to obtain the flame information in the image information;

[0259] Based on the flame information, a fire warning is issued to the target area.

[0260] In this embodiment, the target fire risk level is a level that requires prompting an administrator of the target area to pay close attention to the fire risk situation in the target area.

[0261] The specific level of the target fire risk level can be determined based on user customization or based on actual conditions. For example, the target fire risk level is level 3 or above, or the target fire risk level can be level 5, which is not limited in this application.

[0262] The image information is an image captured by an image sensor of the target area.

[0263] In the actual implementation process, an image sensor can be set in the target area to collect image information, or an unmanned aerial vehicle can be used to collect image information of the target area.

[0264] Of course, the image information of the target area may also be collected in any other feasible manner, which is not limited in this application.

[0265] In the actual implementation process, when the fire risk level corresponding to the target area is higher than the target fire risk level, the inspection frequency of the target area can be increased to obtain more image information so as to keep an eye on the situation of the target area.

[0266] When the fire risk level corresponding to the target area is not higher than the target fire risk level, the target area can continue to be monitored, and the comprehensive fire risk index of the target area can be updated in real time to update the fire risk level of the target area in real time.

[0267] The image recognition model is used to extract features from image information and identify whether the image information includes fire risk and other information.

[0268] Image recognition models can include machine learning models and neural network models, etc.

[0269] During the actual execution process, the image recognition model can be determined based on the actual situation. For example, the image recognition model can be an image recognition model based on the Transformer structure and a model such as AlexNet, which is not limited in this application.

[0270] In the actual implementation process, the acquired image information can be input into the image recognition model to determine the flame situation in the image information, thereby determining whether there is a fire in the target area, so as to be able to respond to the fire in time and extinguish the fire quickly.

[0271] The flame information includes: the existence of flame in the target area in the image information or the absence of flame in the target area in the image information.

[0272] In the actual execution process, the flame information included in the image information can be analyzed to determine whether flames are generated in the target area, so as to perform corresponding operations.

[0273] like Figure 4 As shown, when the comprehensive fire risk index is calculated to be 30 through the acquisition of various types of environmental information, it can be determined that the fire risk level is level 5. At this time, the inspection frequency of the target area can be increased, such as increasing the frequency of manual inspections, image sensor inspections, and drone inspections, and obtaining image information collected during the inspection process, and analyzing the image information of the target area.

[0274] For example, when image information is processed through an image recognition model and it is determined that a flame has occurred in a target area, the discovery of a fire in the target area is characterized and information such as the discovery of a fire in the target area and the extent of the fire is reported. After receiving the information about the fire, medium or large drones can be dispatched to the target area to extinguish the fire based on the extent of the fire, and personnel can be arranged to check to ensure that there are no remaining sparks that could cause the fire to reignite, and to ensure that the comprehensive fire risk index can be restored to a safe range.

[0275] According to the fire monitoring method provided in the embodiment of the present application, by comparing the correlation between the fire risk level of the target area and the given target fire risk level, the safety level of the target area can be effectively determined. When the fire risk of the target area is high, the inspection frequency of the area is increased, and image information of the target area is collected to increase attention to the target area, so as to timely discover the acquisition of the target area and process the acquisition in time, thereby reducing the speed of fire spread and reducing unnecessary losses.

[0276] In some embodiments, processing the image information through an image recognition model to obtain flame information in the image information includes:

[0277] Performing at least one of Gaussian low-pass filtering, hue-saturation-brightness block homomorphic filtering color image enhancement, and image segmentation based on a K-means model in a brightness-color opposition dimension color space on the image information to obtain first image information;

[0278] The first image information is processed by an image recognition model to obtain flame information in the first image information.

[0279] In this embodiment, the first image information is high-quality image information obtained after the image information is processed by an image processing technology.

[0280] Image processing can include operations such as image denoising, image enhancement, and image restoration on image information.

[0281] In the actual implementation process, image information can be processed through methods such as mean filtering, median filtering, and deep learning models.

[0282] Of course, in the actual implementation process, the image information can also be processed by any other feasible method to obtain high-quality image information, and this application does not limit it.

[0283] According to the fire monitoring method provided in the embodiment of the present application, by performing image processing on the acquired image information, the image quality of the image information is effectively improved, thereby improving the accuracy of the flame information obtained by subsequent recognition of the first image information.

[0284] In some embodiments, performing image processing on the image information to obtain the first image information includes:

[0285] Performing Gaussian low-pass filtering on the image information to obtain first image information;

[0286] In this embodiment, the Gaussian low-pass filtering process can transform the image information from the spatial domain to the frequency domain.

[0287] Gaussian low-pass filtering can filter out the high-frequency part of the frequency domain image and pass the low-frequency part, which can remove or weaken the image noise, thereby reducing the noise while retaining the edge information.

[0288] The first image information is an image obtained after noise reduction is performed on the image information.

[0289] In the actual implementation process, the following formula can be used:

[0290] G(u,v)=H(u,v)F(u,v)

[0291] Perform Gaussian low-pass filtering on the image information.

[0292] Among them, F(u, v) is the Fourier transform of the image information; H(u, v) is the transfer function of the Gaussian filter; G(u, v) is the Fourier transform of the image information after noise reduction.

[0293] The transfer function of the Gaussian filter can be determined based on the following formula:

[0294]

[0295] Among them, H(u, v) is the transfer function of the Gaussian filter, D0 is the cutoff frequency of the Gaussian low-pass filter, which is a non-negative quantity; D(u, v) represents the distance from the origin of the frequency plane to the point (u, v), u is the horizontal coordinate of the point; v is the vertical coordinate of the point.

[0296] The distance from the origin of the frequency plane to the point (u, v) can be calculated based on the following formula:

[0297]

[0298] Among them, D(u, v) is the distance from the origin of the frequency plane to the point (u, v); u is the horizontal coordinate of the point; v is the vertical coordinate of the point; M is the row of image information; N is the column of image information.

[0299] like Figure 5 As shown, in the actual implementation process, the cutoff frequency D0 of the Gaussian low-pass filter can be set to -50, the filter radius can be set to 5, the image information can be fast Fourier transformed, and then the high-frequency components of F(u, v) can be attenuated by using the low-pass characteristics of H(u, v). The smoothed image g(x, y) is obtained by inverse Fourier transform of G(u, v), that is, the first image information is obtained.

[0300] According to the fire risk monitoring method provided in the embodiment of the present application, by performing Gaussian low-pass filtering on the image information, the color image is effectively smoothed and denoised, thereby improving the image quality of the acquired first image information and improving the accuracy of subsequent image recognition.

[0301] In some embodiments, performing image processing on the image information to obtain the first image information includes:

[0302] The image information is subjected to a hue-saturation-brightness block homomorphic filtering color image enhancement process to obtain first image information.

[0303] In this embodiment, the hue represents the frequency of the color.

[0304] Saturation refers to the depth of the color.

[0305] The hue-saturation-intensity (HSI) color space is more in line with the visual characteristics of the human eye, can eliminate the influence of brightness in color flame images, and better describe the perception of color.

[0306] Homomorphic filtering is used to enhance or restore images with uneven illumination.

[0307] Homomorphic filtering can use the illumination or reflectance model of an image as the basis for frequency domain processing, and improve the quality of image information by compressing the brightness range and enhancing the contrast.

[0308] In the actual implementation process, the image information can be enhanced by the block homomorphic filtering color image enhancement method of HIS to obtain a first image with better quality.

[0309] According to the fire monitoring method provided in the embodiment of the present application, by performing block homomorphic filtering color image enhancement processing on the hue-saturation-brightness of the image information, the acquired first image information is made more consistent with the nonlinear characteristics of the human eye corresponding to the continuous reading of the flame image, thereby alleviating the technical problem of directly performing Fourier transform on the flame image, resulting in distortion of the image information, and improving the accuracy of subsequent image recognition.

[0310] In some embodiments, the image information may be subjected to Gaussian low-pass filtering and hue-saturation-brightness block homomorphic filtering color image enhancement processing to obtain first image information, so as to obtain first image information of higher quality.

[0311] According to the fire monitoring method provided in the embodiment of the present application, by performing image processing on the image information, the image quality of the acquired first image information is effectively improved, thereby improving the accuracy and precision of subsequent image recognition.

[0312] In some embodiments, performing image processing on the image information to obtain the first image information includes:

[0313] Based on the K-means model of the brightness-color opposition dimension color space, the image information is segmented to obtain the first image information.

[0314] In this embodiment, the brightness-color opposition dimension (Lab) color space includes one brightness channel and two color channels.

[0315] Among them, L is brightness, indicating a change from pure black to pure white, a indicates a change from red to green, and b indicates a change from yellow to blue.

[0316] The brightness-color opposition dimensional color space is able to represent the colors perceived by the human eye.

[0317] The brightness-color opposition dimension color space can make up for the uneven color distribution of the RGB space, and the color space is independent of light and equipment.

[0318] The brightness-color opposition dimension color space achieves better color image segmentation results by retaining a wide color gamut and rich colors.

[0319] The K-means model measures similarity based on the distance between points.

[0320] During the application of the K-means model, based on the set number of categories K and the iteration termination condition, the value of the cluster center is continuously updated iteratively to obtain a better segmentation result.

[0321] The specific value of K can be determined based on user customization or actual conditions, and is not limited in this application.

[0322] In the actual implementation process, the K-means model of the brightness-color opposition dimension color space converts the image information from the RGB color space to the Lab color space, determines the value of K, such as K is 4, and then calculates the distance between each pixel of the a and b components and the cluster center, and divides the pixels closest to the cluster center into the cluster closest to the cluster center. According to the pixels of each cluster of the a and b components, the cluster center is updated, and it is continuously iterated until the iteration termination condition is reached to obtain the first image information.

[0323] According to the fire risk monitoring method provided in the embodiment of the present application, through the K-means model based on the brightness-color opposition dimension color space, the K-means model performs clustering in the brightness-color opposition dimension color space, effectively segments the flame area in the image information from the complex background, can cope with different lighting conditions and complex backgrounds, and improve the accuracy of flame recognition.

[0324] In some embodiments, the image information can also be subjected to at least two of the following processes: Gaussian low-pass filtering, hue-saturation-brightness block homomorphic filtering color image enhancement processing, and K-means model based on brightness-color opposition dimensional color space to segment the image information and obtain the first image information.

[0325] In this embodiment, the image information may be subjected to Gaussian low-pass filtering and hue-saturation-brightness block homomorphic filtering color image enhancement processing to obtain the first image information.

[0326] The image information may be subjected to Gaussian low-pass filtering and segmented based on a K-means model of a brightness-color opposition dimension color space to obtain first image information.

[0327] The image information can be subjected to hue-saturation-brightness block homomorphic filtering color image enhancement processing and segmented based on a K-means model of a brightness-color opposition dimension color space to obtain first image information.

[0328] The image information may also be subjected to Gaussian low-pass filtering, hue-saturation-brightness block homomorphic filtering color image enhancement processing, and segmented based on a K-means model of a brightness-color opposition dimension color space to obtain first image information.

[0329] In the actual implementation process, the corresponding processing method can be selected based on the specific situation of the image information, and this application does not limit it.

[0330] According to the fire monitoring method of the embodiment of the present application, the image information is subjected to Gaussian low-pass filtering, hue-saturation-brightness block homomorphic filtering color image enhancement processing, and the image information is segmented based on the K-means model of the brightness-color opposition dimensional color space, thereby effectively improving the quality of the image information and thus improving the accuracy of subsequent image recognition.

[0331] The fire monitoring method provided in the embodiment of the present application can be executed by a fire monitoring device. In the embodiment of the present application, the fire monitoring device provided in the embodiment of the present application is described by taking the fire monitoring method executed by the fire monitoring device as an example.

[0332] The embodiment of the present application also provides a fire monitoring device.

[0333] like Figure 6 As shown, the fire monitoring device includes: a first processing module 610 , a second processing module 620 and a third processing module 630 .

[0334] The first processing module 610 is used to obtain at least one type of environmental information of the target area at the current collection time, where the types of environmental information include: types corresponding to meteorological information, types corresponding to satellite remote sensing information, and types corresponding to drone monitoring information;

[0335] The second processing module 620 is used to perform weighted prediction on various types of environmental information to obtain a comprehensive fire risk index;

[0336] The third processing module 630 is used to issue a fire warning to the target area based on the comprehensive fire risk index.

[0337] According to the fire monitoring device provided in the embodiment of the present application, by acquiring the meteorological information, satellite remote sensing information and drone monitoring information of the target area, the comprehensive fire risk index corresponding to the multiple types of environmental information at the current collection time is effectively determined, so that based on the actual value of the comprehensive fire risk index, the target area is warned, the fire risk monitoring cost is reduced, the accuracy of the monitoring results is improved, the probability of missed reports and false alarms is reduced, and timely attention can be paid to areas where fire may occur or areas where fire has occurred, reducing the huge losses caused by the spread of fire due to undetected fire.

[0338] In some embodiments, the second processing module 620 may also be used to:

[0339] Obtain the fire risk index corresponding to each type of environmental information and the first weight corresponding to each type of environmental information;

[0340] The fire risk index corresponding to each type of environmental information is processed based on each first weight to obtain a comprehensive fire risk index.

[0341] In some embodiments, the second processing module 620 may also be used to:

[0342] Classifying target environmental information among various types of environmental information to obtain at least one type of sub-environmental information;

[0343] Obtain the fire line index corresponding to each type of sub-environment information and the second weight corresponding to each type of sub-environment information,

[0344] The fire line index corresponding to each type of sub-environment information is processed based on each second weight to obtain the fire line index corresponding to the target type of environment information.

[0345] In some embodiments, the second processing module 620 may also be used to:

[0346] Various types of environmental information are input into the target model to obtain the comprehensive fire risk index output by the target model.

[0347] In some embodiments, the apparatus may further include a fourth processing module, configured to:

[0348] Obtain multiple types of historical environmental information and the fire risk levels corresponding to each type of historical environmental information;

[0349] Based on multiple categories of historical environmental information and the fire risk levels corresponding to each category of historical environmental information, the target model is trained with the goal of outputting a comprehensive fire risk index, a first weight corresponding to each category of environmental information, and a second weight corresponding to at least one category of sub-environmental information included in each category of environmental information.

[0350] In some embodiments, the third processing module may also be used to:

[0351] Based on the comprehensive fire danger index, determine the fire danger level corresponding to the target area;

[0352] Based on the fire risk level corresponding to the target area, a fire warning is issued to the target area.

[0353] In some embodiments, the third processing module may also be used to:

[0354] When the fire risk level corresponding to the target area is the target fire risk level, collecting image information of the target area;

[0355] Process the image information through the image recognition model to obtain the flame information in the image information;

[0356] Based on the flame information, a fire warning is issued to the target area.

[0357] In some embodiments, the third processing module may also be used to:

[0358] Performing at least one of Gaussian low-pass filtering, hue-saturation-brightness block homomorphic filtering color image enhancement, and image segmentation based on a K-means model in a brightness-color opposition dimension color space on the image information to obtain first image information;

[0359] The first image information is processed by an image recognition model to obtain flame information in the first image information.

[0360] The fire monitoring device in the embodiment of the present application can be an electronic device, or a component in the electronic device, such as an integrated circuit or a chip. The electronic device can be a terminal, or other devices other than a terminal. Exemplarily, the electronic device can be a mobile phone, a tablet computer, a laptop computer, a PDA, a vehicle-mounted electronic device, a mobile Internet device (Mobile Internet Device, MID), an augmented reality (augmented reality, AR) / virtual reality (virtual reality, VR) device, a robot, a wearable device, an ultra-mobile personal computer (ultra-mobile personal computer, UMPC), a netbook or a personal digital assistant (personal digital assistant, PDA), etc. It can also be a server, a network attached storage (Network Attached Storage, NAS), a personal computer (personal computer, PC), a television (television, TV), a teller machine or a self-service machine, etc., and the embodiment of the present application does not make specific limitations.

[0361] The fire monitoring device in the embodiment of the present application may be a device having an operating system. The operating system may be an Android operating system, an IOS operating system, or other possible operating systems, which are not specifically limited in the embodiment of the present application.

[0362] The fire monitoring device provided in the embodiment of the present application can achieve Figures 1 to 5 To avoid repetition, the various processes implemented by the method embodiment are not described here.

[0363] In some embodiments, Figure 7 As shown, an embodiment of the present application further provides an electronic device 700, including a processor 701, a memory 702, and a computer program stored in the memory 702 and executable on the processor 701. When the program is executed by the processor 701, each process of the above-mentioned fire monitoring method embodiment is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be described here.

[0364] It should be noted that the electronic devices in the embodiments of the present application include the mobile electronic devices and non-mobile electronic devices mentioned above.

[0365] The embodiment of the present application also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the various processes of the above-mentioned fire monitoring method embodiment are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0366] The processor is the processor in the electronic device described in the above embodiment. The readable storage medium includes a computer readable storage medium, such as a computer read-only memory ROM, a random access memory RAM, a magnetic disk or an optical disk.

[0367] An embodiment of the present application also provides a computer program product, including a computer program, which implements the above-mentioned fire monitoring method when executed by a processor.

[0368] The processor is the processor in the electronic device described in the above embodiment. The readable storage medium includes a computer readable storage medium, such as a computer read-only memory ROM, a random access memory RAM, a magnetic disk or an optical disk.

[0369] An embodiment of the present application further provides a chip, which includes a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the various processes of the above-mentioned fire monitoring method embodiment, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0370] It should be understood that the chip mentioned in the embodiments of the present application can also be called a system-level chip, a system chip, a chip system or a system-on-chip chip, etc.

[0371] It should be noted that, in this article, the terms "comprise", "include" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise one..." do not exclude the presence of other identical elements in the process, method, article or device including the element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved, for example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0372] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the relevant technology, can be embodied in the form of a computer software product, which is stored in a storage medium (such as ROM / RAM, a disk, or an optical disk), and includes a number of instructions for a terminal (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0373] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.

[0374] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "illustrative embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials, or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0375] Although the embodiments of the present application have been shown and described, those skilled in the art will appreciate that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present application, and that the scope of the present application is defined by the claims and their equivalents.

Claims

1. A fire monitoring method, characterized in that: include: Acquire at least one type of environmental information of the target area at the current collection time, where the types of environmental information include: types corresponding to meteorological information, types corresponding to satellite remote sensing information, and types corresponding to drone monitoring information; Perform weighted prediction on various types of environmental information to obtain a comprehensive fire risk index; Based on the comprehensive fire risk index, a fire warning is issued to the target area.

2. The fire monitoring method according to claim 1, characterized in that: The weighted prediction of various types of environmental information to obtain a comprehensive fire risk index includes: Obtaining a fire risk index corresponding to each type of environmental information and a first weight corresponding to each type of environmental information; The fire risk index corresponding to each type of the environmental information is processed based on the first weights to obtain the comprehensive fire risk index.

3. The fire monitoring method according to claim 2, characterized in that: The obtaining of the fire risk index corresponding to each type of environmental information includes: Classifying target environmental information among various types of environmental information to obtain at least one type of sub-environmental information; Obtaining the fire risk index corresponding to each type of sub-environment information and the second weight corresponding to each type of sub-environment information, The fire risk index corresponding to each type of the sub-environmental information is processed based on each of the second weights to obtain the fire risk index corresponding to the target type of environmental information.

4. The fire monitoring method according to any one of claims 1 to 3, characterized in that: The weighted prediction of various types of environmental information to obtain a comprehensive fire risk index includes: Various types of environmental information are input into the target model to obtain a comprehensive fire risk index output by the target model.

5. The fire monitoring method according to claim 4, characterized in that: The target model is trained based on the following steps: Obtain multiple types of historical environmental information and fire risk levels corresponding to each type of historical environmental information; Based on the multiple categories of historical environmental information and the fire risk levels corresponding to each category of the historical environmental information, the target model is trained with the goal of outputting the comprehensive fire risk index, the first weight corresponding to each category of the environmental information, and the second weight corresponding to at least one category of sub-environmental information included in each category of the environmental information.

6. The fire monitoring method according to any one of claims 1 to 3, characterized in that: The step of providing a fire warning to the target area based on the comprehensive fire risk index includes: Based on the comprehensive fire risk index, determining the fire risk level corresponding to the target area; Based on the fire risk level corresponding to the target area, a fire warning is issued to the target area.

7. The fire monitoring method according to claim 6, characterized in that: The providing a fire warning to the target area based on the fire risk level corresponding to the target area includes: When the fire risk level corresponding to the target area is the target fire risk level, collecting image information of the target area; Processing the image information through an image recognition model to obtain flame information in the image information; Based on the flame information, a fire warning is issued to the target area.

8. The fire monitoring method according to claim 7, characterized in that: The step of processing the image information by using an image recognition model to obtain flame information in the image information includes: Performing at least one of Gaussian low-pass filtering, hue-saturation-brightness block homomorphic filtering color image enhancement, and image segmentation based on a K-means model in a brightness-color opposition dimension color space on the image information to obtain first image information; The first image information is processed by an image recognition model to obtain the flame information in the first image information.

9. A fire monitoring device, characterized in that: include: A first processing module is used to obtain at least one type of environmental information of the target area at the current collection time, where the types of environmental information include: types corresponding to meteorological information, types corresponding to satellite remote sensing information, and types corresponding to drone monitoring information; The second processing module is used to perform weighted prediction on various types of environmental information to obtain a comprehensive fire risk index; The third processing module is used to issue a fire warning to the target area based on the comprehensive fire risk index.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the fire monitoring method according to any one of claims 1 to 8 is implemented.

Images