A fire scene information inversion method based on smoke traces

Abstract

The present invention discloses a fire scene information inversion method based on smoke traces, which belongs to the field of fire investigation technology and includes the following steps: S1, construction of basic fire scene information; S2, identification of fire scene smoke traces; S3, inversion of fire scene information; wherein, S101, establishment of a three-dimensional boundary model of the fire scene; S102, establishment of a three-dimensional model of combustibles in the fire scene; S201, obtaining the grayscale value distribution of smoke traces; S202, construction of a three-dimensional structure of smoke trace data; S203, normalization of the grayscale values ​​to obtain the probability distribution of smoke traces; S301, inversion calculation of flame height information; S302, inversion calculation of fire source power information; S303, inversion calculation of flame temperature information. The present invention adopts the above-mentioned fire scene information inversion method based on smoke traces, utilizes a fire scene smoke trace identification method and a fire scene information inversion method, realizes the inversion calculation of fire scene information, assists manual fire investigation work, and realizes the automation and intelligence of fire inversion.

Description

Technical Field

[0001] The invention relates to the technical field of fire investigation, and in particular to a fire scene information inversion method based on smoke traces. Background Art

[0002] Fire investigation is an important responsibility of firefighting agencies. The main content of fire cause investigation is to discover, collect and analyze various trace evidence. The formation and retention of various fire trace evidence have regularity and particularity. According to their different formation and retention processes and characteristics, they can directly or indirectly prove the fire spread path, fire point, fire cause, fire nature, etc.

[0003] Data analysis, hypothesis proposing and hypothesis testing are important links in fire investigation work to trace the source of the fire. However, this process currently relies mainly on manual analysis, inference and demonstration, which is heavily dependent on the work experience of investigators and lacks automated and intelligent methods for fire location inference based on fire dynamics theory. Summary of the invention

[0004] The purpose of the present invention is to provide a fire scene information inversion method based on smoke traces, realize the inversion calculation of fire scene information, assist manual fire investigation work, and realize the automation and intelligence of fire inversion.

[0005] To achieve the above object, the present invention provides a fire scene information inversion method based on smoke traces, comprising the following steps:

[0006] S1. Construction of basic fire scene information;

[0007] S2, identification of smoke traces in fire scene;

[0008] S3, fire scene information inversion;

[0009] Among them, the construction of basic fire scene information includes the following sub-steps:

[0010] S101. Based on the drawing information of the fire scene, a three-dimensional boundary model of the fire scene is established using a three-dimensional space modeling tool, including walls and openings;

[0011] S102. Establish a three-dimensional model of the combustibles in the fire scene according to the distribution of the combustibles in the fire scene, including wood and plastics commonly found in the fire scene;

[0012] Fire smoke trace recognition includes the following sub-steps:

[0013] S201, extracting photos of smoke marks from the fire scene, and performing grayscale processing on the obtained photos of smoke marks from the fire scene using a grayscale calculation method to obtain grayscale value distribution of the smoke marks;

[0014] S202, constructing a three-dimensional structure of smoke trace data, and constructing a digital smoke trace image of the entire fire scene in a three-dimensional modeling tool according to the calculated grayscale distribution;

[0015] S203, normalizing the grayscale values ​​to obtain the probability distribution of the smoke mark, and taking the boundary where the probability distribution is 0.5 as the outer contour of the mark;

[0016] Fire scene information inversion includes the following sub-steps:

[0017] S301, flame height information inversion calculation;

[0018] S302, fire source power information inversion calculation;

[0019] S303: flame temperature information inversion calculation.

[0020] Preferably, the basic information modeling of the fire scene is based on the boundary and combustible information of the fire scene, and the basic input information of the fire scene is constructed using the developed dedicated three-dimensional model platform;

[0021] Fire scene smoke trace recognition is used to process fire scene smoke trace photos, quantify the smoke traces, identify the smoke intensity distribution information, and construct a three-dimensional smoke trace distribution field;

[0022] Fire scene information inversion is used to process smoke distribution information. The smoke trace inversion model is used to invert and calculate the smoke traces in each part to obtain the fire source size, fire source power, and temperature distribution data corresponding to each smoke trace, and finally integrate the fire scene parameter information of the entire fire scene.

[0023] Preferably, step S201 specifically comprises: extracting the RGB (R, G, B) value of each pixel point, and calculating the gray value of each point using the following conversion formula: :

[0024] (1).

[0025] Preferably, step S202 specifically includes: constructing a smoke trace dataset Gray (x, y, z) in the fire scene using the calculated grayscale value of each point, inputting the dataset into a three-dimensional modeling platform, and obtaining a three-dimensional distribution of the smoke trace in the fire scene.

[0026] Preferably, step S203 is specifically as follows: calculating the maximum value Max_Gray and the minimum value Min_Gray of the grayscale data set Gray (x, y, z), the normalized grayscale distribution of each point is L (x, y, z) = Gray (x, y, z) / (Max_Gray-Min_Gray), and taking the boundary of the normalized distribution of 0.5 as the boundary contour of the smoke mark.

[0027] Preferably, step S301 is specifically as follows: according to the obtained grayscale distribution of the smoke trace, the outer contour curve and the lower edge contour of the smoke trace are extracted, and the size of the fire source is directly obtained through the width of the lower edge contour. D , use formula (2) to fit the outer contour curve of the smoke trace to obtain the average flame height corresponding to the smoke trace Z f ;

[0028] ; (2);

[0029] in, Z The ordinate represents the outer contour curve of the smoke trace; x The horizontal axis represents the outer contour curve of the smoke trace.

[0030] Preferably, step S302 specifically includes: after obtaining the average flame height according to step S301, using formula (3) to calculate the power information of the fire source;

[0031] (3);

[0032] in, Indicates the heat release rate of the fire source; represents the specific heat capacity at constant pressure; Indicates the air density; represents the ambient temperature; g represents the acceleration due to gravity.

[0033] Preferably, step S303 is specifically as follows: after obtaining the fire source power information according to step S302, the center temperature distribution information at different flame heights is obtained using the following formula;

[0034] Continuous flame zone: (4);

[0035] Intermittent flame zone: (5);

[0036] Buoyancy plume area: (6);

[0037] in, Indicates the flame center temperature at different flame heights;

[0038] According to the radial temperature Gaussian distribution relationship, the radial flame temperature at different heights of the flame is obtained. ;

[0039] (7);

[0040] in, is a constant; b Indicates the average height of the flame Z f The relevant constants; r Indicates radial distance.

[0041] Therefore, the present invention adopts the above-mentioned fire scene information inversion method based on smoke traces to realize the inversion calculation of fire scene information, assist manual fire investigation work, and realize the automation and intelligence of fire inversion.

[0042] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 It is a grayscale conversion diagram of fire scene smoke traces and a schematic diagram of smoke trace contour recognition in an embodiment of a fire scene information inversion method based on smoke traces of the present invention;

[0044] Figure 2 It is a schematic diagram of a process of calculating the flame height by using smoke traces in an embodiment of a fire scene information inversion method based on smoke traces of the present invention;

[0045] Figure 3 It is a schematic diagram of temperature distribution information obtained by inverting and calculating smoke traces in an embodiment of a fire scene information inversion method based on smoke traces of the present invention. DETAILED DESCRIPTION

[0046] The technical solution of the present invention is further described below through the accompanying drawings and embodiments.

[0047] Unless otherwise defined, technical or scientific terms used in the present invention shall have the common meanings understood by one having ordinary skills in the field to which the present invention belongs.

[0048] Embodiment 1

[0049] The present invention provides a fire scene information inversion method based on smoke traces, comprising the following steps:

[0050] S1. Construction of basic fire scene information: Based on the boundary and combustible information of the fire scene, the basic input information of the fire scene is constructed using the developed dedicated 3D model platform;

[0051] The construction of basic fire scene information includes the following sub-steps:

[0052] S101. Based on the drawing information of the fire scene, use a three-dimensional space modeling tool to establish a three-dimensional boundary model of the fire scene, including walls, openings, etc.;

[0053] S102. Establish a three-dimensional model of combustibles in the fire scene according to the distribution of combustibles in the fire scene, including common materials such as wood and plastic in the fire scene;

[0054] S2, fire scene smoke trace recognition, used to process fire scene smoke trace photos, quantify the smoke traces, identify the smoke intensity distribution information, and construct a three-dimensional smoke trace distribution field;

[0055] Fire smoke trace recognition includes the following sub-steps:

[0056] S201, extracting photos of smoke marks from the fire scene, and performing grayscale processing on the obtained photos of smoke marks from the fire scene using a grayscale calculation method to obtain grayscale value distribution of the smoke marks; Figure 1 Specifically, the following formula is used to calculate the gray value of each pixel point: :

[0057] (1).

[0058] S202, constructing a three-dimensional structure of smoke trace data, and constructing a digital smoke trace image of the entire fire scene in a three-dimensional modeling tool according to the calculated grayscale distribution; specifically: using the calculated grayscale value of each point, construct a smoke trace data set Gray (x, y, z) in the fire scene, and input the data set into the three-dimensional modeling platform to obtain the three-dimensional distribution of the smoke trace of the fire scene.

[0059] S203, normalize the grayscale values ​​to obtain the probability distribution of the smoke trace, and take the boundary with a probability distribution of 0.5 as the outer contour of the trace; specifically: calculate the maximum value Max_Gray and the minimum value Min_Gray of the grayscale data set Gray (x, y, z), the normalized grayscale distribution of each point is L (x, y, z) = Gray (x, y, z) / (Max_Gray-Min_Gray), and take the boundary with a normalized distribution of 0.5 as the boundary contour of the smoke trace.

[0060] S3, fire scene information inversion, is used to process smoke distribution information. The smoke trace inversion model is used to invert and calculate the smoke traces of each part, and the fire source size, fire source power, and temperature distribution data corresponding to each smoke trace are obtained. Finally, the fire scene parameter information of the entire fire scene is integrated. Figure 2 shown.

[0061] Fire scene information inversion includes the following sub-steps:

[0062] S301, flame height information inversion calculation; specifically: according to the obtained grayscale distribution of the smoke trace, extract the outer contour curve and the lower edge contour of the smoke trace, and directly obtain the size of the fire source through the width of the lower edge contour D , use formula (2) to fit the outer contour curve of the smoke trace to obtain the average flame height corresponding to the smoke trace Z f ;

[0063] ; (2);

[0064] in, Z The ordinate represents the outer contour curve of the smoke trace; x The horizontal axis represents the outer contour curve of the smoke trace.

[0065] S302, fire source power information inversion calculation; specifically: after obtaining the average flame height according to step S301, the power information of the fire source is calculated using formula (3);

[0066] (3);

[0067] in, Indicates the heat release rate of the fire source; represents the specific heat capacity at constant pressure; Indicates the air density; represents the ambient temperature; g represents the acceleration due to gravity.

[0068] S303, flame temperature information inversion calculation, the temperature distribution information schematic diagram obtained by inversion calculation of smoke traces is as follows Figure 3 As shown, from bottom to top, they are the continuous flame area, the intermittent flame area, and the buoyancy plume area. Specifically, after the fire source power information is obtained according to step S302, the center temperature distribution information at different flame heights is obtained using the following formula;

[0069] Continuous flame zone: (4);

[0070] Intermittent flame zone: (5);

[0071] Buoyancy plume area: (6);

[0072] in, Indicates the flame center temperature at different flame heights;

[0073] According to the radial temperature Gaussian distribution relationship, the radial flame temperature at different heights of the flame is obtained. ;

[0074] (7);

[0075] in, is a constant; b Indicates the average height of the flame Z f The relevant constants; r Indicates radial distance.

[0076] Therefore, the present invention adopts the above-mentioned fire scene information inversion method based on smoke traces to realize the inversion calculation of fire scene information, assist manual fire investigation work, and realize the automation and intelligence of fire inversion.

[0077] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. A fire scene information inversion method based on smoke traces, characterized in that: The following steps are involved: S1. Construction of basic fire scene information; S2, identification of smoke traces in fire scene; S3, fire scene information inversion; Among them, the construction of basic fire scene information includes the following sub-steps: S101. Based on the drawing information of the fire scene, a three-dimensional boundary model of the fire scene is established using a three-dimensional space modeling tool, including walls and openings; S102. Establish a three-dimensional model of the combustibles in the fire scene according to the distribution of the combustibles in the fire scene, including wood and plastics commonly found in the fire scene; Fire smoke trace recognition includes the following sub-steps: S201, extracting photos of smoke marks from the fire scene, and performing grayscale processing on the obtained photos of smoke marks from the fire scene using a grayscale calculation method to obtain grayscale value distribution of the smoke marks; S202, constructing a three-dimensional structure of smoke trace data, and constructing a digital smoke trace image of the entire fire scene in a three-dimensional modeling tool according to the calculated grayscale distribution; S203, normalizing the grayscale values ​​to obtain the probability distribution of the smoke mark, and taking the boundary of the probability distribution of 0.5 as the outer contour of the mark; Fire scene information inversion includes the following sub-steps: S301, flame height information inversion calculation; specifically: according to the obtained grayscale distribution of the smoke trace, extract the outer contour curve and the lower edge contour of the smoke trace, and directly obtain the size of the fire source through the width of the lower edge contour D , use formula (2) to fit the outer contour curve of the smoke trace to obtain the average flame height corresponding to the smoke trace ; ； （2）； in, Z The ordinate represents the outer contour curve of the smoke trace; x The horizontal coordinate representing the outer contour curve of the smoke trace; S302, fire source power information inversion calculation; S303: flame temperature information inversion calculation.

2. A fire scene information inversion method based on smoke traces according to claim 1, characterized in that: Modeling of basic fire scene information: Based on the boundaries and combustible information of the fire scene, the basic input information of the fire scene is constructed using the developed dedicated 3D model platform; Fire scene smoke trace recognition is used to process fire scene smoke trace photos, quantify the smoke traces, identify the smoke intensity distribution information, and construct a three-dimensional smoke trace distribution field; Fire scene information inversion is used to process smoke distribution information. The smoke trace inversion model is used to invert and calculate the smoke traces in each part to obtain the fire source size, fire source power, and temperature distribution data corresponding to each smoke trace, and finally integrate the fire scene parameter information of the entire fire scene.

3. The method for fire scene information inversion based on smoke traces according to claim 2 is characterized in that: Step S201 specifically includes: extracting the RGB (R, G, B) value of each pixel point, and calculating the gray value of each point using the following conversion formula: : （1）。 4. The method for fire scene information inversion based on smoke traces according to claim 3 is characterized in that: Step S202 specifically includes: constructing a smoke trace dataset Gray (x, y, z) in the fire scene using the calculated grayscale value of each point, inputting the dataset into the three-dimensional modeling platform, and obtaining the three-dimensional distribution of the smoke trace in the fire scene.

5. The method for fire scene information inversion based on smoke traces according to claim 4 is characterized in that: Step S203 is specifically as follows: the maximum value Max_Gray and the minimum value Min_Gray of the grayscale data set Gray (x, y, z) are calculated, the normalized grayscale distribution of each point is L (x, y, z) = Gray (x, y, z) / (Max_Gray-Min_Gray), and the boundary where the normalized distribution is 0.5 is taken as the boundary contour of the smoke mark.

6. The method for fire scene information inversion based on smoke traces according to claim 5 is characterized in that: Step S302 specifically includes: after obtaining the average flame height according to step S301, using formula (3) to calculate the power information of the fire source; （3）； in, Indicates the heat release rate of the fire source; represents the specific heat capacity at constant pressure; Indicates the air density; represents the ambient temperature; g represents the acceleration due to gravity.

7. The method for fire scene information inversion based on smoke traces according to claim 6 is characterized in that: Step S303 is specifically as follows: after obtaining the fire source power information according to step S302, the center temperature distribution information at different flame heights is obtained using the following formula; Continuous flame zone: (4); Intermittent flame zone: (5); Buoyancy plume area: (6); in, Indicates the flame center temperature at different flame heights; According to the radial temperature Gaussian distribution relationship, the radial flame temperature at different heights of the flame is obtained. ; （7）； in, is a constant; b Indicates the average height of the flame The relevant constants; r Indicates radial distance.

Images