Fire-fighting scene simulation method, device, equipment, storage medium and product

Abstract

The invention provides a fire-fighting scene simulation method and device, equipment, a storage medium and a product. The method comprises the following steps: acquiring object information of an entity object existing in a fire scene; according to the object information, an adjustment coefficient corresponding to each entity object is determined, and the adjustment coefficient represents the importance of the entity object in the fire-fighting scene simulation process; according to the adjustment coefficient, determining reduction fineness corresponding to each entity object in the fire-fighting scene; and according to the reduction fineness corresponding to each entity object, performing three-dimensional model reconstruction on each entity object in the fire-fighting scene to obtain a simulation scene of the fire-fighting scene.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a fire scene simulation method, device, equipment, storage medium and product. Background Art

[0002] At present, with the rapid development of IoT sensing technology and digital twin technology, fire protection systems can be built based on IoT technology. IoT sensing technology allows fire protection systems to collect key data information through various sensors (such as smoke sensors, fire sources, light sources, etc.). These data are transmitted to relevant data analysis and processing systems for resource integration and emergency response, helping with fire control. Digital twin technology creates digital copies of physical entities in the scene to achieve real-time monitoring and simulation of entity status and behavior. These technologies can be used to analyze and restore firefighting scenes, providing information reference for scientific command and firefighting force dispatch.

[0003] However, existing technical solutions will uniformly identify and restore all objects in the fire scene, and cannot analyze the situation at the fire scene in real time. Therefore, it is difficult to identify important objects or high-risk sources in time. The calculation amount of restoration is large, and the overall processing efficiency of the solution is low. Summary of the invention

[0004] The embodiments of the present application provide a fire scene simulation method, device, equipment, storage medium and product, which can improve the efficiency of fire scene restoration.

[0005] In a first aspect, the present application provides a fire scene simulation method, comprising:

[0006] Obtain object information of entity objects existing in the fire scene;

[0007] Determine, according to the object information, an adjustment coefficient corresponding to each of the entity objects, wherein the adjustment coefficient represents the importance of the entity object in the fire scene simulation process;

[0008] Determining the restoration precision corresponding to each of the entity objects in the fire scene according to the adjustment coefficient;

[0009] According to the restoration precision corresponding to each of the entity objects, a three-dimensional model of each entity object in the fire scene is reconstructed to obtain a simulation scene of the fire scene.

[0010] In some possible implementations, the adjustment coefficient includes an offset passivation coefficient, and the offset passivation coefficient represents the degree of influence of the entity object on the evolution of the fire scene. The step of determining the adjustment coefficient corresponding to each entity object according to the object information includes:

[0011] Processing the object information through a preset restoration model to determine historical data corresponding to each of the object information, wherein the historical data includes possible changes of the physical object in the fire scene;

[0012] Substitute each of the historical data into a preset offset passivation coefficient calculation formula to calculate the offset passivation coefficient corresponding to each of the entity objects.

[0013] In some possible implementations, the historical data includes combustion characteristic values, and the processing of the object information by using a preset restoration model to determine the historical data corresponding to each of the object information includes:

[0014] respectively extracting the combustion features corresponding to each of the object information;

[0015] The combustion characteristics are processed by a preset restoration model to determine the historical data corresponding to each of the object information.

[0016] In some possible implementations, the step of respectively substituting each of the historical data into a preset offset passivation coefficient calculation formula to calculate an offset passivation coefficient corresponding to each of the entity objects includes:

[0017] Determining a normalization factor corresponding to each combustion characteristic value according to a deviation value between each combustion characteristic value and a preset standard value;

[0018] Each of the combustion characteristic values ​​is normalized by using a normalization factor corresponding to each of the combustion characteristic values ​​to obtain an offset passivation coefficient corresponding to each of the entity objects.

[0019] In some possible implementations, the adjustment coefficient includes a damage coefficient, and the damage coefficient represents the degree of damage of the physical object in the fire scene. The step of determining the adjustment coefficient corresponding to each of the physical objects according to the object information includes:

[0020] Extracting damage features corresponding to each of the physical objects from the object information respectively;

[0021] According to the damage characteristics, a damage coefficient corresponding to each of the physical objects is determined.

[0022] In some possible implementations, the damage features include edge features, corner features, and material features, the edge features represent the degree of damage to the edge of the entity object, the corner features represent the structural stability of the entity object, and the material features represent the material type of the entity object, and determining the damage coefficient corresponding to each of the entity objects according to the damage features includes:

[0023] The edge features, the corner features and the material features are summed to obtain a damage degree value of each entity object in the fire scene;

[0024] The damage degree value of each entity object is corrected according to the camera parameter information to determine the damage coefficient corresponding to each entity object.

[0025] In some possible implementations, the acquiring object information of a physical object existing in the fire scene includes:

[0026] Acquire fire scene images;

[0027] Image recognition is performed on the fire scene image to obtain object information of each entity object in the fire scene.

[0028] In some possible implementations, after reconstructing the three-dimensional model of each entity object in the fire scene according to the restoration precision corresponding to each entity object to obtain a simulation scene of the fire scene, the method further includes:

[0029] Acquiring sound data, temperature data and / or gas data in the fire scene;

[0030] The fire scene sound data, the fire scene temperature data and / or the fire scene gas data are added to the simulation scene.

[0031] In some possible implementations, after adding the fire scene sound data, the fire scene temperature data and / or the fire scene gas data to the simulation scene, the method further includes:

[0032] According to the damage coefficient corresponding to each entity object in the fire environment simulation scene, damage data in the fire scene is obtained;

[0033] Extracting features of the fire scene sound data, the fire scene temperature data, and the fire scene gas data to obtain hazard source data;

[0034] Based on the hazard source data, simulation and deduction are performed according to a preset simulation model to obtain risk development trend data;

[0035] Determining the dangerous area in the fire environment simulation scene according to the damage situation data and risk development trend data;

[0036] An optimal rescue path for the fire scene is determined by a preset shortest path algorithm and the dangerous area, and the optimal rescue path does not pass through the dangerous area.

[0037] In a second aspect, the present application provides a fire scene simulation device, the device comprising:

[0038] An acquisition module, used to acquire object information of entity objects existing in the fire scene;

[0039] A determination module, used to determine the adjustment coefficient corresponding to each of the entity objects according to the object information, wherein the adjustment coefficient represents the importance of the entity object in the fire scene simulation process;

[0040] The determination module is further used to determine the restoration precision corresponding to each of the entity objects in the fire scene according to the adjustment coefficient;

[0041] The reconstruction module is used to reconstruct the three-dimensional model of each entity object in the fire scene according to the restoration precision corresponding to each entity object, so as to obtain a simulation scene of the fire scene.

[0042] In a third aspect, the present application provides a fire scene simulation device, which includes: a processor, and a memory storing computer program instructions; the processor reads and executes the computer program instructions to implement the fire scene simulation method described above.

[0043] In a fourth aspect, the present application provides a computer-readable storage medium having computer program instructions stored thereon, and the computer program instructions, when executed by a processor, implement the fire scene simulation method as described above.

[0044] In a fifth aspect, the present application provides a computer program product. When the instructions in the computer program product are executed by a processor of an electronic device, the electronic device executes the fire scene simulation method described above.

[0045] The fire scene simulation method, device, equipment, storage medium and product provided in the embodiment of the present application obtain the object information of the physical objects existing in the fire scene, determine the adjustment coefficient representing the importance of each object in the fire scene simulation process, and determine the restoration precision corresponding to each of the physical objects in the fire scene according to the adjustment coefficient. For important physical objects, the subsequent restoration precision is high, while the restoration precision of unimportant physical objects is low. Then, according to the restoration precision corresponding to each of the physical objects, a three-dimensional model is reconstructed for each physical object in the fire scene to obtain a simulation scene of the fire scene. Since the unimportant physical objects are restored with low precision and calculation amount, the efficiency of fire scene restoration can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The present application can be better understood from the following description of the specific embodiments of the present application in conjunction with the accompanying drawings, in which:

[0047] Other features, objects and advantages of the present application will become more apparent by reading the following detailed description of non-limiting embodiments with reference to the accompanying drawings, in which the same or similar reference numerals represent the same or similar features.

[0048] Figure 1 is a flow chart of a fire scene simulation method provided by an embodiment of the present application;

[0049] Figure 2 is a flow chart of a fire scene simulation method provided by another embodiment of the present application;

[0050] Figure 3 It is an overall flow chart of the fire scene simulation process provided by an embodiment of the present application;

[0051] Figure 4 It is a structural schematic diagram of a fire scene simulation device provided by an embodiment of the present application;

[0052] Figure 5 It is a schematic diagram of the hardware structure of the fire scene simulation device provided in the embodiment of the present application. DETAILED DESCRIPTION

[0053] The features and exemplary embodiments of various aspects of the present application will be described in detail below. In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain the present application, rather than to limit the present application. For those skilled in the art, the present application can be implemented without the need for some of these specific details. The following description of the embodiments is only to provide a better understanding of the present application by illustrating the examples of the present application.

[0054] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the statement "include..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0055] At present, with the rapid development of IoT sensing technology and digital twin technology, IoT sensing devices in fire protection systems can obtain IoT data information such as fire sources, smoke, and personnel conditions. Through unified data aggregation and processing, they can be displayed on the data visualization system page to facilitate viewing of fire emergency situations, which helps the fire control room to integrate resources, optimize emergency response plans, and unify dispatch and command, so as to properly handle sudden fire incidents in a timely manner. Nowadays, many visual fire emergency systems mainly monitor abnormal data through data sensors of equipment such as smoke, fire sources, and light sources, and transmit the data to the data analysis and processing system for alarm notification. At the same time, these systems will display sensor equipment data in real time on the large screen and access appropriate on-site video data so that the fire intelligence center can view the fire scene in real time and conduct unified dispatch and command.

[0056] However, existing technical solutions have problems such as insufficient real-time scene restoration and analysis, untimely high-risk source alarms, and insufficient optimization of rescue plan path strategies. For example, the current status of the fire scene is only displayed through video transmission, but the real-time scene model is not restored according to the current status of the fire scene; at the same time, the existing technical system also fails to combine the scene's real-time physical equipment, IoT special data and other contents for visualization, such as high-risk source alarms and rescue plan optimization. Only relatively simple analysis of video surveillance data is performed, and it is impossible to restore and analyze the on-site fire scene in real time. In addition, the optimal evacuation plan is only generated based on sensor data, lacking a combined analysis of the real-time data of the fire scene and the sensor analysis data, and does not provide a visual display of the optimal evacuation path plan selected in combination with the hazard source alarm coefficient.

[0057] Through research, the inventors found that there is a need for a visualization generation method for fire emergency scene mapping and analysis alarms. This method can combine real-time scene model restoration, physical equipment, IoT data and other contents for visualization, and at the same time analyze the more appropriate rescue plan for the current fire scene through an algorithm model, and display the rescue path on a visualization page, thereby improving the efficiency and accuracy of emergency rescue work.

[0058] In order to solve the problems of the prior art, the embodiments of the present application provide a fire scene simulation method, device, equipment, storage medium and product. The fire scene simulation method provided by the embodiments of the present application is first introduced below.

[0059] Figure 1 FIG. 1 is a flow chart of a fire scene simulation method provided by an embodiment of the present application. Figure 1 As shown, the method includes the following steps S101 to S104.

[0060] Step S101: Obtain object information of entity objects existing in the fire scene.

[0061] In a specific implementation, the data of physical objects in the fire scene can be obtained through sensors, scanners, or databases with pre-stored data. These data may include information such as the location, shape, and size of the object. The acquired data is processed and analyzed to identify the various physical objects in the fire scene. In this way, the object information of the physical objects in the fire scene, such as location coordinates, object type, importance, etc., can be obtained. This information will be used in the subsequent simulation process.

[0062] As another example, image processing or point cloud processing technology may be used to identify various physical objects in a fire scene.

[0063] Step S102: determining the adjustment coefficient corresponding to each of the above-mentioned physical objects according to the above-mentioned object information, wherein the above-mentioned adjustment coefficient represents the importance of the physical object in the fire scene simulation process.

[0064] In a specific implementation, the importance of each entity object is evaluated according to the object information obtained from step S101. According to the evaluation result, an adjustment coefficient is determined for each entity object. This coefficient reflects the importance of the object in the simulation process and will affect the precision and priority of subsequent simulations.

[0065] Step S103: Determine the restoration precision corresponding to each of the above entity objects in the fire scene according to the above adjustment coefficient.

[0066] In a specific implementation, the physical objects in the fire scene are divided into different levels of precision according to the adjustment coefficient determined in step S102. Generally speaking, objects with higher importance will have higher precision requirements, and vice versa. Thus, the restoration precision corresponding to each physical object is set.

[0067] Step S104: Reconstructing a three-dimensional model of each entity object in the fire scene according to the restoration precision corresponding to each entity object, to obtain a simulation scene of the fire scene.

[0068] In a specific implementation, according to the restoration precision set in step S103, the corresponding restoration precision is used to reconstruct the three-dimensional model of each entity object, and the reconstructed three-dimensional models of each entity object are combined according to the actual situation to obtain a simulation scene of the fire scene.

[0069] As another example, a corresponding three-dimensional model may be generated according to the entity object information obtained in step S101 and the precision requirement set in step S103. The details and precision of each model may vary according to its importance in the scene and usage requirements.

[0070] After the above steps are completed, the above simulation scene can also be displayed.

[0071] In a specific implementation, the three-dimensional model generated in step S104 is placed in a virtual environment of a fire scene, and then the three-dimensional model in the virtual environment is rendered into a visual simulation scene using a graphics rendering technology, and the simulation scene is displayed.

[0072] The above implementation method provided by the embodiment of the present application obtains the object information of the physical objects existing in the fire scene, determines the adjustment coefficient representing the importance of each object in the fire scene simulation process, and determines the restoration precision corresponding to each of the above physical objects in the fire scene according to the above adjustment coefficient. For important physical objects, the subsequent restoration precision is high, while the restoration precision of unimportant physical objects is low. Then, according to the above restoration precision corresponding to each of the above physical objects, a three-dimensional model is reconstructed for each physical object in the fire scene to obtain a simulation scene of the fire scene. Since the unimportant physical objects are restored with low precision and calculation amount, the efficiency of fire scene restoration can be improved.

[0073] In order to obtain an accurate offset passivation coefficient, in some embodiments, the adjustment coefficient includes an offset passivation coefficient, and the offset passivation coefficient indicates the influence of the entity object on the evolution of the fire scene. The above S102 may include steps A1 to B1:

[0074] Step A1: The object information is processed by a preset restoration model to determine the historical data corresponding to each of the object information, where the historical data includes possible changes of the entity object in the fire scene.

[0075] In a specific implementation, a preset restoration model is preset, and based on the preset restoration model, the current data of the entity object is combined with the preset restoration model, processed to generate historical data, and the historical data is calculated based on the current data of the entity object through an algorithm.

[0076] As another example, the preset restoration model can predict the evolution trend of the object through a statistical model, a machine learning model, or an empirical formula.

[0077] Step B1: Substitute each of the above historical data into a preset offset passivation coefficient calculation formula to calculate the offset passivation coefficient corresponding to each of the above objects.

[0078] In a specific implementation, a mathematical formula for calculating the offset passivation coefficient is predefined according to the scene requirements and the simulation purpose. The historical data generated in step A1 is used as input to the offset passivation coefficient calculation formula. According to the defined formula, the offset passivation coefficient corresponding to each entity object is calculated.

[0079] The above implementation method provided by the embodiment of the present application uses a preset restoration model to process the above object information, thereby determining the historical data corresponding to each of the above object information, substituting the historical data into a preset offset passivation coefficient calculation formula, and using the formula to calculate the corresponding offset passivation coefficient. An accurate offset passivation coefficient can be obtained.

[0080] In order to obtain more targeted historical data, in some embodiments, the historical data includes combustion characteristic values, and the above A1 may include steps A2 to B2:

[0081] Step A2: respectively extract the corresponding combustion features from each of the above object information.

[0082] In the specific implementation, if the scene data is based on images or videos, the various objects in the fire scene must first be identified and classified using computer vision technology. Specifically, this can be achieved through object detection algorithms. Each detected object is segmented to determine its boundaries and position in the scene. Then, based on the current state of the detected object, the possible combustion characteristics of each object are obtained.

[0083] Step B2: Process the above combustion characteristics through a preset restoration model to determine the historical data corresponding to each of the above object information.

[0084] In the specific implementation, a suitable restoration model is selected in advance according to the scene requirements and data characteristics. The combustion characteristics of each object extracted in step A2 are used as the input of the model. The model processes these input data into an estimate of the combustion state of the object within a certain period of time. Thus, the historical data corresponding to each of the above object information is determined.

[0085] As another example, the generated historical data may be in the form of a time series, describing changes in the combustion characteristics of each object within a preset time period, such as changes in temperature over time, fluctuations in flame size, etc.

[0086] The above implementation provided by the embodiment of the present application extracts the corresponding combustion features in the object information and determines the historical data corresponding to each of the above object information by the combustion features. By extracting the combustion features and determining the historical data based on them, more targeted historical data can be obtained.

[0087] In order to make the processing of subsequent steps more accurate, in some embodiments, the above B1 may include steps A3 to B3:

[0088] Step A3: determining a normalization factor corresponding to each of the above combustion characteristic values ​​according to a deviation value between each of the above combustion characteristic values ​​and a preset standard value.

[0089] In a specific implementation, a reference value of the combustion state is preset according to the model requirements, that is, a preset standard value. For each combustion characteristic value, the deviation between its current value and the preset standard value is calculated. Then, the normalization factor corresponding to each of the above combustion characteristic values ​​is determined.

[0090] Step B3: normalizing each of the above combustion characteristic values ​​by using the normalization factors corresponding to each of the above combustion characteristic values, and obtaining the offset passivation coefficient corresponding to each of the above objects.

[0091] In a specific implementation, each combustion characteristic value is normalized using the normalization factor determined in step A3, and the normalization result is determined as the normalization factor corresponding to each of the above combustion characteristic values.

[0092] As another example, the ratio or difference between the normalized characteristic value and the preset standard value may be simply used as a measure of the offset passivation coefficient.

[0093] In the above implementation method provided by the embodiment of the present application, the deviation value between each of the above combustion characteristic values ​​and the preset standard value determines the normalization factor corresponding to each of the above combustion characteristic values. Then, each of the above combustion characteristic values ​​is normalized using the normalization factor to obtain the offset passivation coefficient corresponding to each of the above objects. The combustion characteristic values ​​are normalized using the normalization factor to obtain the offset passivation coefficient, so that the processing of the subsequent steps is more accurate.

[0094] In order to speed up the processing, in some embodiments, the adjustment coefficient includes a damage coefficient, and the damage coefficient indicates the degree of damage of the physical object in the fire scene. The above S102 may include steps A4 to B4:

[0095] Step A4: extracting damage features corresponding to each of the above-mentioned physical objects from each of the above-mentioned object information respectively.

[0096] In a specific implementation, according to the requirements of the fire scene and the nature of the object, the features that can be used to describe the degree of damage of the physical object are determined. The actual damage feature data of each physical object in the fire scene is extracted from each of the above object information.

[0097] As another example, the features used to describe the degree of damage of the physical object may include the degree of burning, the degree of deformation, the degree of material damage, the structural integrity, etc.

[0098] Step B4: Determine the damage coefficient corresponding to each of the above-mentioned physical objects according to the above-mentioned damage characteristics.

[0099] In the specific implementation, the calculation formula of the damage coefficient is preset according to the actual needs and scenarios. For example, a linear or nonlinear function can be used to map the damage feature to a numerical representation of the degree of damage. Based on the calculation formula of the damage coefficient, the damage coefficient corresponding to each of the above-mentioned entity objects is calculated according to the above-mentioned damage features.

[0100] The above implementation method provided by the embodiment of the present application extracts the damage features corresponding to each of the above physical objects from the object information, and determines the damage coefficient corresponding to the physical object according to the damage features. The restoration precision is further determined by the damage condition of the physical object, and the restoration precision can be adjusted according to the damage condition of the physical object to speed up the processing.

[0101] In order to obtain an accurate damage coefficient, in some embodiments, the damage features include edge features, corner features and material features. The edge features represent the degree of damage to the edge of the entity object, the corner features represent the structural stability of the entity object, and the material features represent the material type of the entity object. The above B4 may include steps A5 to B5:

[0102] Step A5: summing the edge features, the corner features and the material features to obtain a damage degree value of each entity object in the fire scene.

[0103] In the specific implementation, the edge feature represents the degree of damage at the edge of the entity object. It can be obtained through image processing algorithms or sensor data. The corner feature can reflect the structural stability of the entity object. It is usually obtained through structural analysis or simulation of physical models. The material feature can describe the material type of the entity object and its response characteristics to fire. Then, the above three feature values ​​can be summed according to certain weights to obtain the total damage value of each entity object.

[0104] Step B5: Correct the damage degree value of each entity object according to the camera parameter information to determine the damage coefficient corresponding to each entity object.

[0105] In a specific implementation, according to the camera parameter information, the modified damage degree value is determined by considering the influence of the object on the damage degree within the camera viewing angle. Based on this modified damage degree value, the final damage coefficient is obtained.

[0106] The above implementation method provided in the embodiment of the present application sums the edge features, corner features and material features, comprehensively considers to obtain the damage degree value of the entity object, and then corrects the above damage degree value of each entity object according to the camera parameter information to obtain the final damage coefficient. An accurate damage coefficient can be obtained.

[0107] In order to accurately obtain the object information of each physical object, in some embodiments, the above S101 may include steps A6 to B6:

[0108] Step A6: Acquire fire scene images.

[0109] In a specific implementation, a camera or sensor suitable for a fire scene, such as a high-resolution camera or an infrared sensor, can be used to capture real-time image data in the scene, and then these fire scene images can be quickly acquired.

[0110] Step B6: Perform image recognition on the fire scene image to obtain object information of each entity object in the fire scene.

[0111] In the specific implementation, computer vision technology is used to detect and locate various physical objects in the fire scene. Then, feature extraction is performed on each detected object to obtain object information of each physical object in the fire scene.

[0112] In the above implementation provided by the embodiment of the present application, after acquiring the fire scene image, image recognition is performed on the fire scene image, thereby obtaining object information of each entity object in the fire scene. By using image recognition technology, the object information of each entity object can be accurately obtained.

[0113] In order to obtain a more complete fire environment simulation scene, you can refer to Figure 2 In some implementations, after the above S104, the above method may further include steps S201 to S202:

[0114] Step S201: Acquire the sound data, temperature data and / or gas data in the fire scene.

[0115] In a specific implementation, professional sensors may be used to obtain sound data, temperature data and / or gas data in the above-mentioned firefighting scenarios.

[0116] Step S202: adding the fire scene sound data, the fire scene temperature data and / or the fire scene gas data to the simulation scene.

[0117] In a specific implementation, the sound, temperature and gas data obtained from various sensors are converted into a computer-processable format, such as a digital signal or a specific data structure. Computer graphics are used to create a simulation scene of the fire environment, and the sound, temperature and gas data collected in step S201 are embedded into the corresponding position and time point in the virtual scene. Thus, a fire environment simulation scene is obtained.

[0118] In the above implementation provided by the embodiment of the present application, after obtaining the sound data, temperature data and / or gas data in the above fire scene, the above data are added to the simulation scene of the fire scene, thereby obtaining a fire environment simulation scene. By adding environment-related data to the original simulation scene, a more complete fire environment simulation scene can be obtained.

[0119] In order to obtain a reasonable optimal rescue path, in some embodiments, after the above S202, the above method may further include steps A7 to E7:

[0120] Step A7: According to the damage coefficient corresponding to each entity object in the fire environment simulation scene, the damage data in the fire scene is obtained.

[0121] In a specific implementation, the damage condition of each object in the simulation process is calculated according to the damage coefficient of each entity object, and the damage condition data in the fire scene is obtained.

[0122] Step B7: extract features from the fire scene sound data, the fire scene temperature data and the fire scene gas data to obtain hazard source data.

[0123] In a specific implementation, the sound data of the fire scene obtained from step S202 is subjected to spectrum analysis to extract key sound features such as frequency, intensity and duration, and to identify sound features related to fire, such as flame explosion, object collapse, etc. The temperature data obtained from step S202 is analyzed to identify the fire source, hot spot and its change trend, so as to detect abnormal temperature change patterns, indicating possible fire development or hot spot. The gas data obtained from step S202 is analyzed to detect changes in the concentration of dangerous gases such as smoke and CO2, and to identify possible situations where the concentration of dangerous gases exceeds the standard. The above data are combined to obtain the hazard source data.

[0124] Step C7: Based on the above hazard source data, simulation is performed according to the preset simulation model to obtain risk development trend data.

[0125] In the specific implementation, an appropriate preset simulation model is selected, such as a fire diffusion model, a heat conduction model, etc., and simulation is performed based on the hazard source data extracted in step B7. The sound characteristics, temperature distribution, and gas concentration data are input into the model to simulate the development trend of the fire and scene changes to obtain risk development trend data.

[0126] Step D7: Determine the dangerous area in the fire environment simulation scenario based on the damage data and risk development trend data.

[0127] In a specific implementation, the damage data in step A7 and the risk development trend data in step C7 are combined to determine the dangerous areas in the fire scene, and a corresponding danger level is assigned to each area according to the degree of damage, fire development speed and danger.

[0128] Step E7: Determine the optimal rescue path for the fire scene by using a preset shortest path algorithm and the dangerous area, where the optimal rescue path does not pass through the dangerous area.

[0129] In a specific implementation, a preset shortest path algorithm is used to calculate an optimal rescue path based on the dangerous area determined in step D7, ensuring that the optimal rescue path does not pass through the dangerous area.

[0130] The above implementation method provided by the embodiment of the present application obtains damage data in the fire scene according to the damage coefficient corresponding to each entity object in the above fire environment simulation scene, and then extracts features of the above fire scene sound data, the above fire scene temperature data and the above fire scene gas data to obtain hazard source data, and simulates and deduces through a preset simulation model to obtain risk development trend data, thereby determining the dangerous area in the above fire environment simulation scene. Then, the preset shortest path algorithm is used to determine the optimal rescue path of the fire scene. A reasonable optimal rescue path can be obtained.

[0131] In one embodiment of the present application, it is necessary to collect various types of data related to the environment through sensors in the fire environment. Considering the data quality problems that may exist in the historical collected data, such as noise, sensor missing and other abnormal data, first of all, the collected data such as temperature, humidity, air pressure and other environmental data, smoke or gas concentration data, video / image data, sound data, personnel location data, equipment status data and other information will be pre-processed, including deduplication, sorting and other operations, and filtering algorithms, smoothing algorithms and the like will be used to repair data quality problems for abnormal values, and interpolation algorithms will be used to fill in missing data values ​​to ensure the validity of historical data.

[0132] In this embodiment, the existing recognition algorithm is used to identify the physical objects in the fire environment, that is, the various items in the fire scene. However, in this embodiment, the offset passivation coefficient is added for the common model recognition algorithms such as image recognition algorithms, stereo vision algorithms, etc. Specifically, the first step is to identify the image or video item information collected from the fire scene through the preset identifyMessage() method, including item name, item material, item shape, item location, burning speed, burning time and other information. The burning speed is usually obtained by comparing the length of the material before and after burning of the object per unit time. The burning time is usually obtained by identifying the material of the object and analyzing and comparing its current status with the burning speed of the material object and the shape of the object. The fire emergency restoration algorithm model B is obtained by regularly training and optimizing the restoration algorithm with high-quality fire emergency data. The historical data C of the corresponding time node is obtained by inputting the object information obtained in the identifyMessage() method, such as the name of the object, the material of the object, the shape of the object, the location of the object, the burning speed, the burning time, the fire spread path and other detailed information. Finally, C is used as a parameter and substituted into the sigmoid() function to obtain an offset passivation coefficient between 0 and 1. The simplified process of the sigmoid() function calculation is:

[0133] f(x)=frac(x)*(1+e^{-k0(x-x_0)})

[0134] The f(x) offset passivation coefficient is a function of the input variable x, that is, the item information. The function contains constants k0 and x_0. k0 is the constant that controls the slope, and x_0 determines the translation position of the function curve on the x-axis. Its value can be obtained by restoring the algorithm model B based on the material and burning time in the item information.

[0135] Through the above calculation, an offset passivation coefficient f(x) will be obtained. The simplified process of restoration using the offset passivation coefficient is y=f(x)*functionX(x), where f(x) is the offset passivation coefficient, and the functionX(x) method is the historical data feature value of the object generated by the point cloud reconstruction algorithm to generate a three-dimensional model. The value calculated by the two is the weighted value of the object. The weighted value will be continuously adjusted according to the accuracy of the historical data of the emergency algorithm model B, and the weights of some important parameters such as building structure, building materials, smoke concentration, temperature and other information will be strengthened, while the weights of some parameters such as firefighting facility alarms and fire extinguishing equipment will be weakened. Through the weighted value y, the impact of different item information on the fire scene can be analyzed. The information of items that have a greater impact on the evolution of the fire will be restored quickly and finely, and the information of items that have a smaller impact on the evolution of the fire will be restored in a coarse-grained manner. Since some items are restored in a coarse-grained manner, the amount of calculation is reduced. This method greatly improves the restoration efficiency of the entire fire scene model.

[0136] In the specific processing process, the key feature information, such as object edges, corners, materials, etc., is first extracted from the image or video through the visual algorithm. The object edge is used as parameter E: the integrity of the object edge is marked, ranging from 0 to 1, 0 means complete damage, 1 means complete integrity, and the edge information of the object is obtained through the edge detection algorithm, including the edge shape and structure of the object. The specific calculation formula can be simplified as follows:

[0137] EdgeMatching()=Q / (Q+FP+FN)

[0138] Where Q is the number of correctly matched edges, FP is the number of falsely detected edges, and FN is the number of missed edges. The corner point is used as the parameter F: it identifies the stability of the corner point, ranging from 0 to 1, 0 means that all corner points are damaged, and 1 means that all corner points are intact.

[0139] The stability of corner points in images / videos can be determined by feature analysis. The corresponding analysis method is:

[0140] VarianceStab()=Var(N) / (k3*Average(N))

[0141] Var(N) is the variance of the grayscale values ​​of pixels around the corners, Average(N) is the mean of the grayscale values ​​of pixels around the corners, and k3 is a constant used to adjust the sensitivity of stability. The material type is used as parameter M: different materials are encoded according to the material type, such as 1 for wood, 2 for metal, 3 for plastic, etc. These parameters facilitate the subsequent rapid modeling of feature objects.

[0142] Since the acquired image will be affected by the distortion caused by the camera shooting, it is necessary to calculate the internal and external parameters of the camera based on the feature points in the image and the existing camera calibration algorithm. The internal parameters include information such as focal length and distortion parameters, and the external parameters include the camera viewing angle parameters. The focal length is used as parameter F to identify the focal length of the camera; the distortion is used as parameter D to identify the camera distortion process; and the camera viewing angle is used as parameter V to identify the position and direction of the camera.

[0143] Based on the feature information extracted from the image or video, combined with the internal and external parameter information of the camera, the damage coefficient is integrated and the model is reconstructed through the long-term trained computing model. The damage coefficient is used as the parameter K. It is a coefficient calculated comprehensively based on the material properties of the feature object, the degree of combustion, the location distance and the duration of the fire, that is, the time factor T. The simplified process method for determining the damage coefficient parameter K is:

[0144] K=[calculateMaterial(E,T)+calculatePoint(F,T)+getMaterial(M,T)]*

[0145] getCamera(F,D,V)

[0146] Among them, calculateMaterial(E, T) function is the object edge information calculation function, which measures the edge information of the object at a specified time; calculatePoint(F, T) function is the object corner information calculation function, which measures the corner information of the object at a specified time; getMaterial(M, T) is the material calculation function, which measures the material data information at a specified time; getCamera(F, D, V) is the camera parameter calculation function, which measures the camera parameter information.

[0147] The damage coefficient K is a coefficient calculated based on the edge information, corner information, material properties, location distance, and fire duration of the feature object. The damage coefficients of feature objects with different material properties and different fire durations will vary. By using the damage coefficient K, the restoration efficiency of the three-dimensional model can be improved. For example, for the solid wood table in the center of the living room of a fire scene, a damage coefficient value K1 is determined after calculation in the above manner. The larger the K1 value, the longer the fire time, the less texture and surface details of the corresponding object, and the lower the precision when reconstructing the three-dimensional model. This coefficient can be used to determine the precision of the restoration of the corresponding object during three-dimensional modeling, thereby improving the restoration efficiency of the three-dimensional model of the entire fire scene. At the same time, after the corresponding model of the fire scene is reconstructed in this way, the damage coefficient curve can be used to deduce the evolution of the model fire.

[0148] Based on the data sent back by the smoke and temperature sensors, we can screen, analyze and process them, issue real-time alarms for locations with abnormal smoke concentrations and temperatures, and promptly notify the firefighting center in the relevant nearby area. We strive to receive news as soon as a fire occurs and conduct emergency firefighting operations in a timely manner.

[0149] According to the gas data corresponding to the fire scene collected by the gas sensor, this method adds a hazard source coefficient indicator to this module for chemical analysis of various sensor data, such as whether the content of flammable and explosive gases such as coal gas and natural gas at the fire scene exceeds the standard, and whether the content of other toxic gases continues to rise. This will serve as a data basis for the subsequent visualization generation method of the optimal solution for emergency scenarios.

[0150] The data information captured by the sound and position sensors at the emergency fire scene is transmitted to the data collector, and then the sound and position information data in the data collector is processed, such as noise reduction and / or data alignment, to improve the data quality and accuracy; then the processed data is analyzed, such as explosion sounds, cracking sounds, rescue sounds, and the location information of the people to be rescued, so that firefighters can carry out accurate rescue in a timely manner through the analyzed data.

[0151] Based on the information obtained from the above calculations, the scene is restored through the reconstructed model to simulate and reproduce the two- and three-dimensional information of the fire scene, including panel data such as smoke, temperature, gas concentration, and sound from the two-dimensional fire panel sensor, while reproducing the three-dimensional scene fire data.

[0152] The damage coefficient obtained by dynamic data module analysis K = [calculateMaterial(E, T) + calculatePoint(F, T) + getMaterial(M, T)] * getCamera(F, D, V) is integrated into the simulation capability. According to the different materials of the three-dimensional scene buildings, the edge information and corner point information of the collected objects are used through the fire edge analysis function - calculateMaterial() and the fire corner point analysis function - calculatePoint() to calculate the burning degree of objects of different materials in the fire scene and other information. The subsequent fire development trend is accurately deduced in the visual interface, providing the fire command and dispatch center with real-time fire scenes and providing a favorable data basis for decision-making.

[0153] According to the real-time results of data analysis and treatment sent back by various sensors such as smoke, temperature, gas, and sound position in the perception data module, high-risk alarms for dangerous sources are displayed on the visualization screen, such as high-risk alarms for gas area concentration, high-risk alarms for smoke and harmful substances concentration, and high-risk alarms for building collapse, so that the command center can quickly formulate emergency dispatch plans. For example, for high-risk alarms for gas or natural gas types, the main switch corresponding to the gas or natural gas pipeline in the corresponding area will be immediately closed through the API interface, and the details of the building structure will be displayed to the rescue personnel in a visual form. At the same time, the rescue personnel will be informed of the location of the vents and windows in a timely manner, and ventilation and personnel evacuation will be carried out in a timely manner; for toxic and harmful gases, the rescue personnel will be informed to carry air purification equipment to purify the toxic gases as much as possible, and the personnel in the toxic area will be evacuated to avoid secondary disasters.

[0154] Collect relevant data of the rescue scene through sensors, pictures, videos, etc., including map data, hazard source location, gas data, personnel distribution information data, etc., and perform data cleaning and processing on the collected data; use the collected map data and the damage coefficient calculated by governance analysis to restore the actual situation of the scene rescue, and simulate and deduce the hazard source data received in real time based on historical data to predict the possible impact range of the hazard source; according to the actual fire data situation, combined with the Dijkstra algorithm, input the damage coefficient and high-risk hazard source information as parameters into the algorithm model to obtain the optimal rescue path for the current emergency fire scene; display the calculated optimal rescue path on the page in a visual way, and display the restored scene, high-risk hazard source distribution, and map information on the visualization page; ensure real-time update of data, keep the path information and related data on the visualization page synchronized with the actual situation, and the rescue commander can adjust the parameters according to different situations or view the optimal rescue path at different time nodes.

[0155] The overall process can be referred to Figure 3 , through 301 dynamic data processing, the feature data in the collected pictures or videos is extracted. Perform edge, corner and material analysis of objects, and optimize and restore them through damage coefficients. Use filtering and smoothing algorithms on 303 historical fire data to repair historical data to ensure the validity and accuracy of the data. Then use the preset algorithm model 302 to use the offset passivation coefficient for rapid restoration. After the model restoration is completed 304, prepare for data fusion. Through 305 perception data governance, perception data such as gas, smoke and temperature are aggregated and governed, and the coefficient value of the hazard source is analyzed. Through other preset algorithm models 306, the restored model and perception data are comprehensively rendered and displayed, and a high-risk alarm is issued for the hazard source. At the same time, the optimal path can also be displayed. Finally, 307 visualization is completed.

[0156] The above-mentioned method of the embodiment of the present application improves the accuracy of the model restoration in the fire emergency scene by training the algorithm model with the passivation offset coefficient. At the same time, the method does not simply display the dynamic data such as pictures / videos in the fire emergency scene, but extracts the dynamic data, and then analyzes the data using the damage coefficient parameter curve to restore the model in real time, so that the fire emergency command center can view the fire scene in a visual way. At the same time, for the perception data obtained by the sensor equipment, it is not only for data summary and display in the visualization system, but also for adding the hazard source coefficient, analyzing and processing the perception data such as coal gas, toxic gas, and building materials, and giving high-risk warnings for different degrees of hazard sources, and showing them in a visual form at the same time, timely reminding the fire emergency personnel to avoid firefighters from being injured during the rescue process to the greatest extent. Finally, through the comprehensive and coordinated calculation of the algorithm model analysis, a scheme for mapping and analyzing the alarm of the current fire emergency scene is formed, and the path of the scheme is presented in a visual form, which helps the commander of the fire emergency center to mobilize personnel.

[0157] Based on the fire scene simulation method provided in the above embodiment, the present application also provides a specific implementation of a fire scene simulation device. Please refer to the following embodiment.

[0158] See first Figure 4 The fire scene simulation device 400 provided in the embodiment of the present application includes the following modules:

[0159] The acquisition module 401 is used to acquire object information of physical objects existing in the fire scene.

[0160] The determination module 402 is used to determine the adjustment coefficient corresponding to each of the entity objects according to the object information, and the adjustment coefficient represents the importance of the entity object in the fire scene simulation process.

[0161] The determination module 402 is further used to determine the restoration precision corresponding to each of the entity objects in the fire scene according to the adjustment coefficient.

[0162] The reconstruction module 403 is used to reconstruct the three-dimensional model of each entity object in the fire scene according to the restoration precision corresponding to each entity object, so as to obtain a simulation scene of the fire scene.

[0163] The above implementation method provided by the embodiment of the present application obtains the object information of the physical objects existing in the fire scene, determines the adjustment coefficient representing the importance of each object in the fire scene simulation process, and determines the restoration precision corresponding to each of the above physical objects in the fire scene according to the above adjustment coefficient. For important physical objects, the subsequent restoration precision is high, while the restoration precision of unimportant physical objects is low. Then, according to the above restoration precision corresponding to each of the above physical objects, a three-dimensional model is reconstructed for each physical object in the fire scene to obtain a simulation scene of the fire scene. Since the unimportant physical objects are restored with low precision and calculation amount, the efficiency of fire scene restoration can be improved.

[0164] As an implementation of the present application, the fire scene simulation device 400 may further include:

[0165] The processing module is used to process the above object information through a preset restoration model to determine the historical data corresponding to each of the above object information, and the above historical data includes possible changes of the physical object in the fire scene.

[0166] The calculation module is used to substitute each of the above historical data into a preset offset passivation coefficient calculation formula to calculate the offset passivation coefficient corresponding to each of the above objects.

[0167] The above implementation method provided by the embodiment of the present application uses a preset restoration model to process the above object information, thereby determining the historical data corresponding to each of the above object information, substituting the historical data into a preset offset passivation coefficient calculation formula, and using the formula to calculate the corresponding offset passivation coefficient. An accurate offset passivation coefficient can be obtained.

[0168] As an implementation of the present application, the fire scene simulation device 400 may further include:

[0169] The extraction module is used to extract the corresponding combustion features from each of the above object information.

[0170] The processing module is used to process the above combustion characteristics through a preset restoration model to determine the historical data corresponding to each of the above object information.

[0171] The above implementation provided by the embodiment of the present application extracts the corresponding combustion features in the object information and determines the historical data corresponding to each of the above object information by the combustion features. By extracting the combustion features and determining the historical data based on them, more targeted historical data can be obtained.

[0172] As an implementation of the present application, the fire scene simulation device 400 may further include:

[0173] The determination module is used to determine the normalization factor corresponding to each of the above combustion characteristic values ​​according to the deviation value between each of the above combustion characteristic values ​​and the preset standard value.

[0174] The normalization module is used to normalize each of the above combustion characteristic values ​​by using the normalization factors corresponding to each of the above combustion characteristic values, so as to obtain the offset passivation coefficient corresponding to each of the above objects.

[0175] In the above implementation method provided by the embodiment of the present application, the deviation value between each of the above combustion characteristic values ​​and the preset standard value determines the normalization factor corresponding to each of the above combustion characteristic values. Then, each of the above combustion characteristic values ​​is normalized using the normalization factor to obtain the offset passivation coefficient corresponding to each of the above objects. The combustion characteristic values ​​are normalized using the normalization factor to obtain the offset passivation coefficient, so that the processing of the subsequent steps is more accurate.

[0176] As an implementation of the present application, the fire scene simulation device 400 may further include:

[0177] The extraction module is used to extract the damage features corresponding to each of the above-mentioned physical objects from each of the above-mentioned object information.

[0178] The determination module is used to determine the damage coefficient corresponding to each of the above-mentioned physical objects according to the above-mentioned damage characteristics.

[0179] The above implementation method provided by the embodiment of the present application extracts the damage features corresponding to each of the above physical objects from the object information, and determines the damage coefficient corresponding to the physical object according to the damage features. The restoration precision is further determined by the damage condition of the physical object, and the restoration precision can be adjusted according to the damage condition of the physical object to speed up the processing.

[0180] As an implementation of the present application, the fire scene simulation device 400 may further include:

[0181] The summing module is used to sum the edge features, the corner features and the material features to obtain the damage degree value of each entity object in the fire scene.

[0182] The correction module is used to correct the damage degree value of each entity object according to the camera parameter information, and determine the damage coefficient corresponding to each entity object.

[0183] The above implementation method provided in the embodiment of the present application sums the edge features, corner features and material features, comprehensively considers to obtain the damage degree value of the entity object, and then corrects the above damage degree value of each entity object according to the camera parameter information to obtain the final damage coefficient. An accurate damage coefficient can be obtained.

[0184] As an implementation of the present application, the fire scene simulation device 400 may further include:

[0185] The acquisition module is used to acquire fire scene images.

[0186] The recognition module is used to perform image recognition on the fire scene image to obtain object information of each entity object in the fire scene.

[0187] In the above implementation provided by the embodiment of the present application, after acquiring the fire scene image, image recognition is performed on the fire scene image, thereby obtaining object information of each entity object in the fire scene. By using image recognition technology, the object information of each entity object can be accurately obtained.

[0188] As an implementation of the present application, the fire scene simulation device 400 may further include:

[0189] The acquisition module is used to acquire the sound data, temperature data and / or gas data in the above fire scene.

[0190] An adding module is used to add the fire scene sound data, the fire scene temperature data and / or the fire scene gas data to the simulation scene.

[0191] In the above implementation provided by the embodiment of the present application, after obtaining the sound data, temperature data and / or gas data in the above fire scene, the above data are added to the simulation scene of the fire scene, thereby obtaining a fire environment simulation scene. By adding environment-related data to the original simulation scene, a more complete fire environment simulation scene can be obtained.

[0192] As an implementation of the present application, the fire scene simulation device 400 may further include:

[0193] The acquisition module is used to obtain the damage data in the fire scene according to the damage coefficient corresponding to each entity object in the above fire environment simulation scene.

[0194] The extraction module is used to extract features from the fire scene sound data, the fire scene temperature data and the fire scene gas data to obtain hazard source data.

[0195] The deduction module is used to perform simulation deduction based on the above-mentioned hazard source data and according to a preset simulation model to obtain risk development trend data.

[0196] The determination module is used to determine the dangerous area in the above-mentioned fire environment simulation scenario based on the above-mentioned damage data and risk development trend data.

[0197] The determination module is also used to determine the optimal rescue path for the fire scene through a preset shortest path algorithm and the dangerous area, and the optimal rescue path does not pass through the dangerous area.

[0198] The above implementation method provided by the embodiment of the present application obtains damage data in the fire scene according to the damage coefficient corresponding to each entity object in the above fire environment simulation scene, and then extracts features of the above fire scene sound data, the above fire scene temperature data and the above fire scene gas data to obtain hazard source data, and simulates and deduces through a preset simulation model to obtain risk development trend data, thereby determining the dangerous area in the above fire environment simulation scene. Then, the preset shortest path algorithm is used to determine the optimal rescue path of the fire scene. A reasonable optimal rescue path can be obtained.

[0199] The various modules in the fire scene simulation device provided in the embodiment of the present application can implement the various steps in the above-mentioned fire scene simulation method and achieve the corresponding effects. For the sake of concise description, they will not be repeated here.

[0200] Figure 5 A schematic diagram of the structure of the fire scene simulation hardware provided in an embodiment of the present application is shown.

[0201] The fire scene simulation device may include a processor 501 and a memory 502 storing computer program instructions.

[0202] Specifically, the processor 501 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0203] The memory 502 may include a large capacity memory for data or instructions. By way of example and not limitation, the memory 502 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive or a combination of two or more of these. In appropriate cases, the memory 502 may include a removable or non-removable (or fixed) medium. In appropriate cases, the memory 502 may be inside or outside the integrated gateway disaster recovery device. In a specific embodiment, the memory 502 is a non-volatile solid-state memory.

[0204] The memory may include a read-only memory (ROM), a random access memory (RAM), a disk storage medium device, an optical storage medium device, a flash memory device, an electrical, optical or other physical / tangible memory storage device. Therefore, typically, the memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., a memory device) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the fire scene simulation method according to any one embodiment of the present disclosure.

[0205] The processor 501 implements any one of the fire scene simulation methods in the above embodiments by reading and executing computer program instructions stored in the memory 502 .

[0206] In one example, the fire scene simulation device may further include a communication interface 503 and a bus 510. Figure 5 As shown, the processor 501, the memory 502, and the communication interface 503 are connected via a bus 510 and communicate with each other.

[0207] The communication interface 503 is mainly used to implement communication between various modules, devices, units and / or equipment in the embodiments of the present application.

[0208] Bus 510 includes hardware, software or both, and the parts of online data flow billing equipment are coupled to each other. For example, but not limitation, bus may include accelerated graphics port (AGP) or other graphics bus, enhanced industrial standard architecture (EISA) bus, front-end bus (FSB), hypertransport (HT) interconnection, industrial standard architecture (ISA) bus, infinite bandwidth interconnection, low pin count (LPC) bus, memory bus, micro channel architecture (MCA) bus, peripheral component interconnection (PCI) bus, PCI-Express (PCI-X) bus, serial advanced technology attachment (SATA) bus, video electronics standard association local (VLB) bus or other suitable bus or two or more of these combinations. In appropriate cases, bus 510 may include one or more buses. Although the present application embodiment describes and shows a specific bus, the present application considers any suitable bus or interconnection.

[0209] In addition, in combination with the fire scene simulation method in the above embodiment, the embodiment of the present application can provide a computer storage medium for implementation. The computer storage medium stores computer program instructions; when the computer program instructions are executed by the processor, any fire scene simulation method in the above embodiment is implemented.

[0210] An embodiment of the present application also provides a computer program product, including a computer program, which, when executed, implements any one of the fire scene simulation methods in the above embodiments.

[0211] It should be clear that the present application is not limited to the specific configuration and processing described above and shown in the figures. For the sake of simplicity, a detailed description of the known method is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present application is not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the order between the steps after understanding the spirit of the present application.

[0212] The functional blocks shown in the above-described block diagram can be implemented as hardware, software, firmware or a combination thereof. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, etc. When implemented in software, the elements of the present application are programs or code segments that are used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or a communication link by a data signal carried in a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, optical fiber media, radio frequency (RF) links, etc. The code segment can be downloaded via a computer network such as the Internet, an intranet, etc.

[0213] It should also be noted that the exemplary embodiments mentioned in this application describe some methods or systems based on a series of steps or devices. However, this application is not limited to the order of the above steps, that is, the steps can be performed in the order mentioned in the embodiment, or in a different order from the embodiment, or several steps can be performed simultaneously.

[0214] Aspects of the present disclosure are described above with reference to the flowchart and / or block diagram of the method, device (system) and computer program product according to the embodiment of the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of each box in the flowchart and / or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device to produce a machine so that these instructions executed by the processor of the computer or other programmable data processing device enable the implementation of the function / action specified in one or more boxes of the flowchart and / or block diagram. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field programmable logic circuit. It can also be understood that each box in the block diagram and / or flowchart and the combination of boxes in the block diagram and / or flowchart can also be implemented by dedicated hardware that performs a specified function or action, or can be implemented by a combination of dedicated hardware and computer instructions.

[0215] The above is only a specific implementation of the present application. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the systems, modules and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here. It should be understood that the protection scope of the present application is not limited to this. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in this application, and these modifications or replacements should be included in the protection scope of this application.

Claims

1. A fire scene simulation method, characterized in that: include: Obtain object information of entity objects existing in the fire scene; Determine, according to the object information, an adjustment coefficient corresponding to each of the entity objects, wherein the adjustment coefficient represents the importance of the entity object in the fire scene simulation process; Determining the restoration precision corresponding to each of the entity objects in the fire scene according to the adjustment coefficient; According to the restoration precision corresponding to each of the entity objects, a three-dimensional model of each entity object in the fire scene is reconstructed to obtain a simulation scene of the fire scene.

2. The fire scene simulation method according to claim 1, characterized in that: The adjustment coefficient includes an offset passivation coefficient, and the offset passivation coefficient represents the influence degree of the entity object on the evolution of the fire scene. The adjustment coefficient corresponding to each entity object is determined according to the object information, including: Processing the object information through a preset restoration model to determine historical data corresponding to each of the object information, wherein the historical data includes possible changes of the physical object in the fire scene; Substitute each of the historical data into a preset offset passivation coefficient calculation formula to calculate the offset passivation coefficient corresponding to each of the entity objects.

3. The fire scene simulation method according to claim 2, characterized in that: The historical data includes combustion characteristic values, and the object information is processed by a preset restoration model to determine the historical data corresponding to each of the object information, including: respectively extracting the combustion features corresponding to each of the object information; The combustion characteristics are processed by a preset restoration model to determine the historical data corresponding to each of the object information.

4. The fire scene simulation method according to claim 3, characterized in that: Substituting each of the historical data into a preset offset passivation coefficient calculation formula to calculate the offset passivation coefficient corresponding to each of the entity objects includes: Determining a normalization factor corresponding to each combustion characteristic value according to a deviation value between each combustion characteristic value and a preset standard value; Each of the combustion characteristic values ​​is normalized by using a normalization factor corresponding to each of the combustion characteristic values ​​to obtain an offset passivation coefficient corresponding to each of the entity objects.

5. The fire scene simulation method according to claim 1, characterized in that: The adjustment coefficient includes a damage coefficient, and the damage coefficient represents the degree of damage of the physical object in the fire scene. The determining, according to the object information, the adjustment coefficient corresponding to each of the physical objects includes: Extracting damage features corresponding to each of the physical objects from the object information respectively; According to the damage characteristics, a damage coefficient corresponding to each of the physical objects is determined.

6. The fire scene simulation method according to claim 5, characterized in that: The damage features include edge features, corner features and material features, the edge features represent the degree of damage to the edge of the entity object, the corner features represent the structural stability of the entity object, and the material features represent the material type of the entity object. Determining the damage coefficient corresponding to each of the entity objects according to the damage features includes: The edge features, the corner features and the material features are summed to obtain the degree of damage of each entity object in the fire scene; The damage degree of each entity object is corrected according to the camera parameter information to determine the damage coefficient corresponding to each entity object.

7. The fire scene simulation method according to claim 1, characterized in that: The obtaining of object information of entity objects existing in the fire scene includes: Acquire fire scene images; Image recognition is performed on the fire scene image to obtain object information of each entity object in the fire scene.

8. The fire scene simulation method according to any one of claims 1 to 7, characterized in that: After reconstructing the three-dimensional model of each entity object in the fire scene according to the restoration precision corresponding to each entity object to obtain a simulation scene of the fire scene, the method further includes: Acquiring sound data, temperature data and / or gas data in the fire scene; The sound data, the temperature data and / or the gas data are added to the simulation scene.

9. The fire scene simulation method according to claim 8, characterized in that: After adding the sound data, the temperature data and / or the gas data to the simulation scene, the method further includes: According to the damage coefficient corresponding to each entity object, the damage data in the fire scene is obtained; Extracting features from the sound data, the temperature data, and the gas data to obtain hazard source data; Based on the hazard source data, simulation and deduction are performed according to a preset simulation model to obtain risk development trend data; Determining the dangerous area in the simulation scenario according to the damage situation data and the risk development trend data; An optimal rescue path is determined by using a preset shortest path algorithm and the dangerous area, and the optimal rescue path does not pass through the dangerous area.

10. A fire scene simulation device, characterized in that: The device comprises: An acquisition module, used to acquire object information of entity objects existing in the fire scene; A determination module, used to determine the adjustment coefficient corresponding to each of the entity objects according to the object information, wherein the adjustment coefficient represents the importance of the entity object in the fire scene simulation process; The determination module is further used to determine the restoration precision corresponding to each of the entity objects in the fire scene according to the adjustment coefficient; The reconstruction module is used to reconstruct the three-dimensional model of each entity object in the fire scene according to the restoration precision corresponding to each entity object, so as to obtain a simulation scene of the fire scene.

11. A fire scene simulation device, characterized in that: The device comprises: a processor and a memory storing computer program instructions; the processor reads and executes the computer program instructions to implement the fire scene simulation method according to any one of claims 1 to 9.

12. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer program instructions, and when the computer program instructions are executed by a processor, the fire scene simulation method according to any one of claims 1 to 9 is implemented.

13. A computer program product, characterized in that When the instructions in the computer program product are executed by a processor of an electronic device, the electronic device executes the fire scene simulation method as described in any one of claims 1 to 9.

Images