Fixed and mobile cooperative detection system and method for highway tunnel fire

Abstract

The application relates to the technical field of fire detection, in particular to a fixed and mobile cooperative detection system and method for highway tunnel fire, which comprises the following steps: presetting monitoring points in a tunnel to be detected, shooting infrared images of monitoring sections of the monitoring points in the tunnel in real time, dividing the infrared images into regions, obtaining high-temperature regions in the infrared images at each moment, obtaining temperature change sequences and trend change coefficients of the high-temperature regions, obtaining temperature drift coefficients of the high-temperature regions by combining the high-temperature regions and temperature value change conditions of preset adjacent regions of the high-temperature regions in the infrared images at each moment in a preset period, obtaining space-time characteristic values of the infrared images at each moment, obtaining fire possibility coefficients of the high-temperature regions, and screening suspected fire regions; and mobilizing a mobile detection unmanned vehicle to shoot a video for evaluating a fire condition in the tunnel to be detected. The application aims to improve the accuracy of highway tunnel fire detection.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of fire detection, in particular to a fixed and mobile cooperative detection system and method for highway tunnel fire. BACKGROUND

[0002] With the development of society, the rapid increase in the number of highway tunnels has brought new challenges to safety management. Due to the particularity of the tunnel structure, when a disaster occurs, the traditional fire smoke detector sensing technology often responds slowly and is difficult to effectively deal with abnormal vehicle behavior, slow disaster handling response speed, and other problems, making it difficult for personnel to evacuate, and thus threatening tunnel safety management.

[0003] At present, infrared thermal imaging and visual image are often used to identify fire hazards in various places in the tunnel, so as to quickly judge potential dangers and achieve fire warning. However, for a long highway tunnel, the road section is long and the environment is complex, and the wind speed in the tunnel is fast, so when using traditional methods to monitor the fire in the tunnel using infrared images, temperature drift is likely to occur, which may cause deviation in the judgment of the actual fire point position and affect the accuracy of the monitoring. SUMMARY

[0004] In view of the above, it is necessary to provide a fixed and mobile cooperative detection system and method for highway tunnel fire, which improves the accuracy of fire monitoring in the highway tunnel compared with the traditional highway tunnel fire detection method:

[0005] In a first aspect, the embodiments of the present application provide a fixed and mobile cooperative detection method for highway tunnel fire, which comprises the following steps:

[0006] Pre-set each monitoring point in the tunnel to be detected, and real-time shoot infrared images of each monitoring point on the monitoring section in the tunnel;

[0007] The infrared images are divided into regions, and each high-temperature region in each infrared image at each time is obtained according to the distribution of temperature values of all regions in each infrared image at each time. The temperature change sequence of each high-temperature region is obtained according to the position distribution and temperature value of all high-temperature regions in each infrared image at each time. The trend change coefficient of each high-temperature region is obtained according to the length and trend change intensity of the temperature change sequence and the temperature value of each high-temperature region. The temperature drift coefficient of each high-temperature region is obtained according to the change of the temperature value of each high-temperature region and its preset adjacent region in each infrared image at each time within a preset time period. The space-time feature value of each infrared image at each time is obtained according to the time at which the temperature value of each high-temperature region in each infrared image at each time has the maximum mutation and the distance from the high-temperature region to the edge of the infrared image to which the high-temperature region belongs. Thus, the fire possibility coefficient of each high-temperature region is obtained, which is used to screen each suspected fire region from the high-temperature regions in each infrared image at each time.

[0008] The distance between each mobile detection unmanned vehicle in the tunnel to be detected and each monitoring point corresponding to the suspected fire region is obtained, and the mobile detection unmanned vehicle is mobilized to shoot a video for evaluating the fire condition in the tunnel to be detected.

[0009] In one embodiment, the process of obtaining the high-temperature region is as follows:

[0010] The segmentation threshold of the temperature value of all regions in all infrared images at each time is obtained, and a region with a temperature value greater than or equal to the segmentation threshold is regarded as a high-temperature region.

[0011] In one embodiment, the process of obtaining the temperature change sequence is as follows:

[0012] The high-temperature region with the minimum temperature value in each infrared image at each time is recorded as the lowest high-temperature region in each infrared image at each time. The center points of each high-temperature region in each infrared image at each time and the center point of the lowest high-temperature region are connected, and the temperature values of all high-temperature regions through which the connecting line passes are arranged in order from each high-temperature region to the lowest high-temperature region to form the temperature change sequence of each high-temperature region in each infrared image at each time.

[0013] In one embodiment, the process of obtaining the trend change coefficient is as follows:

[0014] The product of the length and trend intensity of the temperature change sequence of each high-temperature region is calculated.

[0015] The trend change coefficient is the sum of the temperature value of each high-temperature region and the product.

[0016] In one embodiment, the process of obtaining the temperature drift coefficient is as follows:

[0017] calculating a difference value of temperature values between any two adjacent time points within the preset time period for each high-temperature region; and calculating an accumulated value of the difference values between any two adjacent time points within the preset time period for each high-temperature region;

[0018] calculating a mean value of the difference values between any two time points within the preset time period for each preset neighboring region of each high-temperature region; and calculating an accumulated sum of the mean values within the preset time period for all preset neighboring regions of each high-temperature region;

[0019] calculating a product value of the trend change coefficient and the accumulated sum, mapping the product value as a first positive number; and taking the temperature drift coefficient as a ratio of the accumulated value and the first positive number.

[0020] In one embodiment, the process of obtaining the spatio-temporal characteristic value is as follows:

[0021] arranging temperature values of each high-temperature region in each infrared image at each time point in time sequence before each time point, to obtain a temperature sequence of each high-temperature region at each time point, and obtaining a maximum mutation point in each temperature sequence;

[0022] numbering all time points in time sequence; and calculating a minimum value in sequence numbers of time points at which maximum mutation points of all high-temperature regions in each infrared image at each time point are located;

[0023] calculating an average value of distances from a center point of each high-temperature region to four edges of an infrared image to which the high-temperature region belongs; and calculating an accumulated value of the average values corresponding to all high-temperature regions in each infrared image at each time point;

[0024] the spatio-temporal characteristic value is a ratio of the accumulated value and the minimum value.

[0025] In one embodiment, the process of obtaining the fire possibility coefficient is as follows: mapping the temperature drift coefficient as a second positive number, and taking the fire possibility coefficient as a ratio of the spatio-temporal characteristic value and the second positive number.

[0026] In one embodiment, the process of obtaining the suspected fire region is as follows:

[0027] obtaining a segmentation threshold value of fire possibility coefficients of all high-temperature regions in all infrared images at each time point, and taking high-temperature regions with fire possibility coefficients greater than or equal to the segmentation threshold value as suspected fire regions.

[0028] In one embodiment, the mobile detection unmanned vehicle is arranged to shoot videos of each suspected fire region, for evaluating a fire situation in a tunnel to be detected, including:

[0029] acquire the positions of the monitoring points corresponding to each suspected fire area, and record the positions as the monitoring points of each suspected fire area;

[0030] control the mobile detection unmanned vehicle closest to the monitoring point of each suspected fire area to go to the position of the monitoring point of each suspected fire area, and then shoot a video of the scene for assisting the staff to determine whether a fire occurs in the tunnel to be detected and whether to start the tunnel fire emergency plan.

[0031] In a second aspect, the embodiments of the present application further provide a highway tunnel fire fixed and mobile cooperative detection system, which comprises a memory, a processor, and a computer program stored in the memory and running on the processor, and the processor implements the steps of the highway tunnel fire fixed and mobile cooperative detection method according to any one of the above embodiments when executing the computer program.

[0032] The present application has at least the following beneficial effects:

[0033] The present application can exclude non-fire point areas by extracting high-temperature areas, focus on high-temperature areas that may have fire hazards, and improve the pertinence of fire monitoring; the length of the smoke diffusion path and the temperature variation trend on the smoke diffusion path are used to simulate the smoke diffusion path, and then the temperature value is superimposed to obtain the trend variation coefficient of the high-temperature area, which reflects the possibility of each high-temperature area being a fire point, and then the temperature drift coefficient is obtained by combining the temperature variation of the high-temperature area and the temperature variation of the neighboring area of the high-temperature area, which effectively measures the possibility of each high-temperature area being caused by temperature drift; considering that the smoke diffusion to a non-fire area may cause a high-temperature area caused by smoke to be misjudged as a fire point area, the space-time feature value is obtained by the characteristics of smoke diffusion in the tunnel and the position characteristics of the fire point, which can effectively avoid misjudgment caused by smoke movement and further improve the accuracy of the position judgment of the fire point; the fire possibility coefficient is obtained, which comprehensively considers the temperature drift coefficient and the space-time feature value, can comprehensively and accurately evaluate the possibility of each high-temperature area occurring a fire, provides a scientific and reliable quantitative index for screening suspected fire areas, and helps to accurately select the area most likely to have a fire from numerous high-temperature areas, and provides a clear target for subsequent accurate monitoring and emergency treatment; the distance from the mobile detection unmanned vehicle to the position of the monitoring point corresponding to each suspected fire area is obtained, the mobile detection unmanned vehicle is flexibly dispatched, the on-site video of the suspected fire area can be shot in time, the staff is provided with intuitive and accurate fire situation information, which helps to quickly and accurately determine whether a fire occurs and timely start the tunnel fire emergency plan, and maximizes the reduction of fire loss. BRIEF DESCRIPTION OF DRAWINGS

[0034] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, and the advantages thereof, the accompanying drawings that need to be used in the description of the embodiments or the prior art will be briefly introduced. Obviously, the accompanying drawings in the following description only only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained without creative work based on these drawings.

[0035] Figure 1 The step flow chart of the fixed and mobile cooperative detection method of the highway tunnel fire provided by an embodiment of the present application is shown in the following figure.

[0036] Figure 2 The schematic diagram of the screening process of the suspected fire area is shown in the following figure. DETAILED DESCRIPTION

[0037] In the description of the embodiments of the present application, the words such as "exemplary", "or", "for example" are used to represent an example, illustration or description. Any embodiment or design scheme described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as more preferred or more advantageous than other embodiments or design schemes. In fact, the words such as "exemplary", "or", "for example" are intended to present the relevant concept in a specific manner.

[0038] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one skilled in the art to which the present application belongs. The terms used in the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application. It should be understood that, in the present application, unless otherwise specified, " / " means or.

[0039] In addition, it should be noted that the terms "first", "second" in the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence.

[0040] The specific scheme of the fixed and mobile cooperative detection system and method of the highway tunnel fire provided by the present application will be specifically described below in combination with the accompanying drawings.

[0041] Please refer to Figure 1 which shows the step flow chart of the fixed and mobile cooperative detection method of the highway tunnel fire provided by an embodiment of the present application. The method comprises the following steps:

[0042] Step 1, preset each monitoring point in the tunnel to be detected, and real-time shoot the infrared image of each monitoring point in the monitoring section of the tunnel.

[0043] For each interval distance N, a fixed infrared thermal imaging monitoring point is set for the tunnel to be detected, and an infrared thermal imaging camera is used to take an infrared image of the monitoring section in the tunnel every interval time T. The collected infrared images are divided into z regions.

[0044] In this embodiment, the values of N, T and z are 20 m, 0.5 s and 225 respectively. The values of N, T and z are preset by humans, and the implementer can adjust them according to the actual shooting range of the infrared thermal imaging camera and the actual length of the tunnel to be detected. The present application does not make special limitations.

[0045] Step 2, divide each infrared image into regions, obtain each high-temperature region in each infrared image at each time through the distribution of temperature values of all regions in each infrared image at each time, obtain the temperature change sequence of each high-temperature region through the position distribution and temperature value of all high-temperature regions in each infrared image at each time, obtain the trend change coefficient of each high-temperature region through the length and trend change intensity of the temperature change sequence, and obtain the temperature drift coefficient of each high-temperature region by combining the temperature values of each high-temperature region and the temperature value change of each high-temperature region and its preset adjacent region in each infrared image at each time within the previous preset period.

[0046] For the infrared image in the highway tunnel, when there is no vehicle passing, the overall temperature change is weak, and only a certain change will occur with the change of external temperature, such as higher temperature in summer and lower temperature at night, but the change speed is slow. When the fire point appears, the fire point will rapidly rise in temperature, and the area near the fire point will also have a temperature rise. Specifically, in the infrared image, the gray value of the pixel points in the local area will rise sharply and show a diffusion trend. However, due to the temperature drift, there may be low-temperature pixel points in the high-temperature region, or high-temperature pixel points in the originally low-temperature region, so it is difficult to distinguish the real fire location, and therefore it is necessary to distinguish.

[0047] Specifically, for the real fire point, the actual fire location is fixed, and because the fire speed is fast, when the fire occurs, the temperature at the location of the fire point changes relatively little, and the relative fire will cause the temperature of the area around the fire point to rise sharply. When temperature drift occurs, high temperature points may appear in areas other than the fire point, but because it is caused by temperature drift, the stability of the high temperature point is low, the temperature changes greatly, and because the high temperature point is not the real fire point, the temperature of the surrounding area will not be affected by it and the overall temperature change is small. Specifically, the gray level of the real fire point in the infrared image changes little, but the gray level of the pixel points in the local area around it changes rapidly. In addition, due to the continuous operation of the ventilation system in the highway tunnel, the smoke generated when the fire occurs will move along with the wind direction in the tunnel, and the smoke of the fire usually has a high temperature, which will cause the temperature at the location of the smoke to increase, but as the smoke moves, the temperature of the smoke will also decrease to some extent, and because of the operation of the ventilation system, the area affected by the temperature of the smoke will be consistent with the direction of the smoke, that is, in the infrared image, the high temperature area has high continuity and is distributed more concentratedly, and the temperature gradually decreases from the fire point to the direction of the ventilation wind direction.

[0048] To characterize the temperature change in a single area, the mean value of the gray values of all pixel points in each area is taken as the temperature value of each area. Taking the s-th moment as an example, the segmentation threshold of the temperature values of all areas in all infrared images at the s-th moment is obtained, and the area with a temperature value greater than the segmentation threshold is taken as a high temperature area, and the remaining area is taken as a low temperature area. The difference value of the temperature values between any two adjacent moments within a predetermined period before the s-th moment is calculated for each area, and the mean value of the difference values between all adjacent moments within the predetermined period before the s-th moment is taken as the temperature change rate of each area at the s-th moment. The high temperature area with the smallest temperature value in each infrared image at the s-th moment is recorded as the lowest high temperature area in each infrared image at the s-th moment. The center points of each high temperature area and the center point of the lowest high temperature area in each infrared image at the s-th moment are connected, and the temperature values of all high temperature areas passed by the connecting line are arranged in order from each high temperature area to the lowest high temperature area to form a temperature change sequence of each high temperature area in each infrared image at the s-th moment. It should be noted that if there are multiple high temperature areas with the same minimum temperature value, then each high temperature area is connected to the multiple high temperature areas with the minimum temperature value, and the temperature change sequence is constructed, and then the longest temperature change sequence is selected as the temperature change sequence of each high temperature area.

[0049] In this embodiment, the temperature values of all regions in all infrared images at the s-th moment are taken as inputs, and the cross-validation method is used to output the segmentation threshold of the temperature values of all regions in all infrared images at the s-th moment. The method for obtaining the segmentation threshold by cross-validation is a known technology, and will not be described herein. As other embodiments, on the basis of being able to obtain the segmentation threshold of the temperature values of all regions in all infrared images at the s-th moment, the implementer can use other existing technologies, such as Otsu threshold segmentation algorithm, global threshold segmentation, etc., which are not specially limited in this application.

[0050] In this embodiment, the difference value between the temperature values is the absolute value of the difference. As other embodiments, on the basis of being able to measure the difference between the temperature values, the implementer can use other calculation methods, such as the square of the difference, etc., which are not specially limited in this application.

[0051] Further, the trend change coefficient of each high-temperature region in each infrared image at each moment is obtained by combining the length and trend change intensity of the temperature change sequence of each high-temperature region in each infrared image at each moment, and the expression is:

[0052] In the formula, represents the trend change coefficient of the i-th high-temperature region in the p-th infrared image at the s-th moment; represents the length of the temperature change sequence of the i-th high-temperature region in the p-th infrared image at the s-th moment; represents the trend intensity of the temperature change sequence of the i-th high-temperature region in the p-th infrared image at the s-th moment; represents the temperature value of the i-th high-temperature region in the p-th infrared image at the s-th moment. The trend intensity in the STL (Seasonal and Trend decomposition using Loess) decomposition algorithm can be obtained by the trend intensity calculation formula in the STL decomposition algorithm, which is a known technology and will not be described herein.

[0053] It should be noted that the longer the temperature change sequence of the i-th high-temperature region, the more likely the temperature change sequence of the i-th high-temperature region is formed by the movement of smoke when the i-th high-temperature region is the ignition point. The greater the trend intensity of the temperature change sequence of the i-th high-temperature region, the more the path of the temperature change sequence of the i-th high-temperature region conforms to the rule that the temperature gradually decreases from the ignition point to the direction of the ventilation wind direction. At the same time, the high temperature itself is still a necessary condition for judging the fire. Therefore, the greater the trend change coefficient, the more likely the i-th high-temperature region is the ignition point.

[0054] Furthermore, by using the trend change coefficients of each high-temperature region in each infrared image at each time point, and combining the temperature changes of each high-temperature region and its preset neighboring regions in each infrared image at each time point over a previous preset time period, the temperature drift coefficient of each high-temperature region in each infrared image at each time point is obtained, expressed as:

[0055] In the formula, This represents the temperature drift coefficient of the i-th high-temperature region in the p-th infrared image at time s. This represents the total number of moments within a preset time period preceding and adjacent to the s-th moment; , These represent the temperature values ​​of the i-th high-temperature region in the p-th infrared image at the s-th time, respectively, at the v-th and v-1-th times within the preset time period; J represents the trend change coefficient of the i-th high-temperature region in the p-th infrared image at time s; J represents the total number of preset neighboring regions of the i-th high-temperature region in the p-th infrared image at time s. This represents the rate of temperature change of the j-th preset neighbor region of the i-th high-temperature region in the p-th infrared image at time s. This indicates the operation of taking the absolute value; α represents a preset positive number used to avoid the denominator being 0. The value of α is preset by the user and can be set by the implementer. In this embodiment, the value of α is 0.01.

[0056] In this embodiment, the length of the preset time period is 10 seconds. The length of the preset time period is preset by the user and can be set by the implementer according to the actual situation. This application does not impose any special restrictions.

[0057] In this embodiment, the preset neighboring region of the i-th high-temperature region is the 8-neighborhood of the i-th high-temperature region. The implementer can set the neighboring region of the i-th high-temperature region according to the actual situation.

[0058] It should be noted that: the smaller the overall temperature change rate of the neighborhood of the i-th high-temperature region, the more obvious the temperature change of the i-th high-temperature region in the short term, and the lower the trend change coefficient of the i-th high-temperature region, the more likely the i-th high-temperature region is located due to temperature drift, and the less likely it represents the location of the ignition point itself.

[0059] Step 3: By combining the time when the temperature of each high-temperature region in each infrared image at each time point was the moment when the temperature value of each region changed the most, and the distance of each high-temperature region to the edge of its respective infrared image, the spatiotemporal feature values ​​of each infrared image at each time point are obtained.

[0060] Because there are many ventilation systems in the tunnel and the wind speed in the tunnel is fast, when the fire occurs, the smoke generated will quickly move due to the wind speed in the tunnel, and thus the remaining sections in the tunnel can also appear smoke. Because the smoke itself has a high temperature, when the smoke moves away from the fire point and enters the monitoring range of another infrared thermal imaging camera, a high-temperature area will also appear in the infrared image, and the distribution rule is similar to the area where the fire point is located. Therefore, when the step 2 is used for calculation, the high-temperature area caused by the smoke is easily mistaken as the area where the fire point is located, and thus the judgment of the position of the fire point is wrong, which delays the fire. Therefore, further analysis is needed.

[0061] Specifically, for the fire position in the tunnel, fire can occur in each area in the tunnel, and thus when the fire occurs, the position of the fire point is usually far away from the boundary of the monitoring range of a single monitoring point. Relatively, when a high-temperature area caused by smoke appears in the monitoring range of a monitoring point, because the smoke spreads from a position other than the monitoring point, when the smoke first enters the monitoring range of the monitoring point, the distance between the smoke and the boundary of the monitoring range is close. In addition, for the entire tunnel, when the fire occurs, temperature mutation occurs in the local tunnel area, and the temperature change in the area where the fire point is located appears earliest, and as the smoke spreads, the temperature change in the remaining tunnel areas gradually occurs. Therefore, the time when the temperature change in the area other than the fire point occurs is later than that in the fire point.

[0062] In order to represent the above characteristics, the temperature values of each high-temperature area in each infrared image at the s-th moment before the s-th moment are arranged in time sequence to obtain the temperature sequence of each high-temperature area in each infrared image at the s-th moment, and the maximum mutation point in each temperature sequence is obtained. Then, by the time when the maximum temperature value mutation of each high-temperature area in each infrared image at each moment occurs before, and the distance from each high-temperature area in each infrared image at each moment to the edge of the infrared image, the space-time characteristic value of each infrared image at each moment is obtained, and the expression is:

[0063] In the formula, represents the space-time characteristic value of the p-th infrared image at the s-th moment; represents the number of high-temperature areas in the p-th infrared image at the s-th moment; represents the average value of the distance from the center point of the i-th high-temperature area in the p-th infrared image at the s-th moment to the four edges of the infrared image; min() represents the minimum value operation; represents the serial number of the time when the maximum mutation point in the temperature sequence of the i-th high-temperature area in the p-th infrared image at the s-th moment is located, wherein all the times are numbered in time sequence.

[0064] In this embodiment, the Pettitt mutation point detection algorithm is used to obtain the maximum mutation point in each temperature sequence, specifically: the statistical quantity of each mutation point in each temperature sequence is obtained by the Pettitt mutation point detection algorithm The absolute value of the statistical quantity of the maximum mutation point in each temperature sequence is the maximum mutation point in each temperature sequence. The Pettitt mutation point detection algorithm is a known technology, and will not be described here. As other embodiments, on the basis of being able to obtain the maximum mutation point in each temperature sequence, the implementer can use other existing technologies, such as the Mann-Kendall mutation point detection algorithm, and the present application does not make special restrictions.

[0065] In this embodiment, the distance from the center point of the i-th high-temperature region to the edge of the infrared image is the Euclidean distance.

[0066] It should be noted that: the earlier the temperature mutation appears in the monitoring range of the monitoring point for shooting the p-th infrared image, and the farther the position of the high-temperature region in the monitoring range from the edge of the infrared image, the more likely it is that there is a fire point in the monitoring range of the monitoring point for shooting the p-th infrared image.

[0067] Step 4, obtain the fire possibility coefficient of each high-temperature region in each infrared image at each time by the temperature drift coefficient of each high-temperature region in each infrared image at each time and the space-time characteristic value of each infrared image at each time, for screening each suspected fire region from the high-temperature region in each infrared image at each time; mobilize the mobile detection unmanned vehicle to shoot the video by the distance between each mobile detection unmanned vehicle in the to-be-detected tunnel and the monitoring point corresponding to each suspected fire region, for evaluating the fire situation in the to-be-detected tunnel.

[0068] Further, obtain the fire possibility coefficient of each high-temperature region in each infrared image at each time by the temperature drift coefficient of each high-temperature region in each infrared image at each time and the space-time characteristic value of each infrared image at each time, and the expression is:

[0069] ; in the formula, represents the fire possibility coefficient of the i-th high-temperature region in the p-th infrared image at the s-th time; represents the space-time characteristic value of the p-th infrared image at the s-th time; represents the temperature drift coefficient of the i-th high-temperature region in the p-th infrared image at the s-th time; β represents a predetermined positive number, which is used to avoid the denominator being 0, and the value of β is artificially preset, which can be set by the implementer. In this embodiment, the value of β is 0.01.

[0070] It should be noted that: when the s-th moment, the temperature change of the i-th high-temperature region in the p-th infrared image is less likely to be caused by temperature drift phenomenon, and the p-th infrared image has a higher possibility of having a fire region, then the monitoring range of the monitoring point that takes the p-th infrared image is more likely to have a fire location in the highway tunnel.

[0071] Further, the segmentation threshold of the fire possibility coefficient of all high-temperature regions in all infrared images at the s-th moment is obtained, and the high-temperature region with a fire possibility coefficient greater than or equal to the segmentation threshold is regarded as a suspected fire region. The screening process of the suspected fire region is shown in the schematic diagram as Figure 2

[0072] In this embodiment, the fire possibility coefficient of all high-temperature regions in all infrared images at the s-th moment is taken as input, and the cross-validation method is used to output the segmentation threshold of the fire possibility coefficient of all high-temperature regions in all infrared images at the s-th moment. The method for obtaining the segmentation threshold by cross-validation is a known technology, and will not be described herein. As other embodiments, on the basis of being able to obtain the segmentation threshold of the fire possibility coefficient of all high-temperature regions in all infrared images at the s-th moment, the implementer can use other existing technologies, such as Otsu threshold segmentation algorithm, global threshold segmentation, etc., which are not specially limited by the present application.

[0073] Further, if there is a suspected fire region at the s-th moment, the positions of the monitoring points corresponding to each suspected fire region are obtained and recorded as the monitoring points of each suspected fire region. At the same time, the tunnel fire monitoring system obtains the positions of each mobile detection unmanned vehicle in the tunnel to be detected, sends a control signal to control the mobile detection unmanned vehicle closest to the monitoring point of each suspected fire region to go to the position of the monitoring point of each suspected fire region, and then take a video of the scene and return it to the tunnel fire monitoring system to assist the staff to determine whether a fire has occurred in the tunnel to be detected and whether to start the tunnel fire emergency plan.

[0074] Based on the same inventive concept as the above method, the embodiments of the present application also provide a fixed and mobile cooperative detection system for highway tunnel fire, which comprises a memory, a processor and a computer program stored in the memory and running on the processor. The processor executes the computer program to realize the steps of any one of the above fixed and mobile cooperative detection methods for highway tunnel fire.

[0075] ​In summary, the present application can exclude non-fire point areas by extracting high temperature areas, focus on high temperature areas that may have fire hazards, and improve the pertinence of fire monitoring; simulate smoke diffusion paths using the position distribution and temperature values of high temperature areas, and then obtain the trend change coefficient of the high temperature area by superimposing the temperature values, the length of the smoke diffusion path, and the temperature change trend on the smoke diffusion path, reflecting the possibility of each high temperature area being a fire point, and then combining the temperature change of the high temperature area and the temperature change of the neighboring area of the high temperature area to obtain the temperature drift coefficient, effectively measuring the possibility of each high temperature area being caused by temperature drift; considering that smoke diffusion to non-fire areas may cause misjudgment of high temperature areas caused by smoke as fire point areas, the space-time characteristic value is obtained by the characteristics of smoke diffusion in the tunnel and the position characteristics of the fire point, which can effectively avoid misjudgment caused by smoke movement and further improve the accuracy of the position judgment of the fire point; the fire possibility coefficient is obtained, which comprehensively considers the temperature drift coefficient and the space-time characteristic value, can comprehensively and accurately evaluate the possibility of fire in each high temperature area, provides a scientific and reliable quantitative index for screening suspected fire areas, helps to accurately select the area most likely to have a fire from numerous high temperature areas, and provides a clear target for subsequent accurate monitoring and emergency treatment; by detecting the distance from the mobile detection unmanned vehicle to the position of each suspected fire area monitoring point, the mobile detection unmanned vehicle can be flexibly dispatched, the on-site video of the suspected fire area can be taken in time, and intuitive and accurate fire situation information can be provided for the staff, which helps to quickly and accurately judge whether a fire has occurred and timely start the tunnel fire emergency plan, thereby minimizing fire losses.

[0076] The computer program instructions can also be loaded onto a computer, other programmable data processing apparatus, or other devices to cause a series of operational steps to be performed on the computer, other programmable apparatus or other devices to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide processes for implementing the functions / acts specified in the flowchart and / or block diagram block or blocks.

[0077] It is apparent that a person skilled in the art can make a variety of modifications to the application described herein without departing from the spirit and scope of the application. Therefore, the described embodiments are to be considered in all respects as illustrative and not restrictive.

Claims

1. A method for coordinated detection of fixed and moving fires in highway tunnels, characterized in that, The method includes the following steps: Pre-set monitoring points inside the tunnel to be monitored, and capture infrared images of the monitored sections of the tunnel in real time. Each infrared image is divided into regions. The distribution of temperature values ​​across all regions in each infrared image at each time point is used to obtain high-temperature regions. The location distribution and temperature values ​​of all high-temperature regions in each infrared image at each time point are used to obtain temperature change sequences for each high-temperature region. The length and trend intensity of these temperature change sequences, combined with the temperature values ​​of each high-temperature region, are used to obtain trend change coefficients for each high-temperature region. The temperature drift coefficients for each high-temperature region are obtained by combining the temperature value changes of each high-temperature region and its preset neighboring regions in each infrared image at each time point over a previous preset period. Finally, the spatiotemporal characteristic values ​​of each infrared image at each time point are obtained by combining the time when the temperature of each high-temperature region in each infrared image at each time point experienced the largest temperature change with the distance from each high-temperature region to the edge of its respective infrared image. This yields the fire probability coefficients for each high-temperature region, which are used to filter suspected fire areas from the high-temperature regions in each infrared image at each time point. By measuring the distance between each mobile detection drone in the tunnel to be detected and the corresponding monitoring point in each suspected fire area, the mobile detection drones are mobilized to capture videos to assess the fire situation in the tunnel to be detected. The process for obtaining the trend change coefficient is as follows: Calculate the product of the length of the temperature change sequence and the trend intensity for each high-temperature region; The trend change coefficient is the sum of the product of the temperature value of each high-temperature region and the coefficient. The process of obtaining the temperature drift coefficient is as follows: Calculate the temperature difference between any two adjacent moments in each high-temperature region within the preset time period; and sum the differences between any two adjacent moments in each high-temperature region within the preset time period. Calculate the mean of the difference values ​​between any two times within the preset time period for each preset neighboring region of each high-temperature region; calculate the sum of the mean values ​​of all preset neighboring regions of each high-temperature region within the preset time period; Calculate the product of the trend change coefficient and the cumulative sum, and map the product value to a first positive number; the temperature drift coefficient is the ratio of the cumulative value to the first positive number; The process for obtaining the spatiotemporal feature values ​​is as follows: Arrange the temperature values ​​of each high-temperature region in each infrared image at each time point according to the time sequence to obtain the temperature sequence of each high-temperature region at each time point, and obtain the maximum abrupt change point in each temperature sequence. Number all moments sequentially; calculate the minimum value among the moment numbers of the moments corresponding to the maximum abrupt change points of all high-temperature regions in each infrared image at each moment. Calculate the average distance from the center point of each high-temperature region to the four edges of its corresponding infrared image; calculate the cumulative value of the average distances for all high-temperature regions in each infrared image at each time point; The spatiotemporal characteristic value is the ratio of the cumulative value to the minimum value.

2. The fixed and mobile coordinated detection method for highway tunnel fires as described in claim 1, characterized in that, The process of obtaining the high-temperature region is as follows: Obtain the segmentation threshold for the temperature values ​​of all regions in all infrared images at each time point, and define the regions with temperature values ​​greater than or equal to the segmentation threshold as high-temperature regions.

3. The fixed and mobile coordinated detection method for highway tunnel fires as described in claim 1, characterized in that, The process of obtaining the temperature change sequence is as follows: The high-temperature region with the lowest temperature value in each infrared image at each time point is recorded as the lowest high-temperature region in each infrared image at each time point. The center point of each high-temperature region in each infrared image at each time point is connected to the center point of the lowest high-temperature region. The temperature values ​​of all high-temperature regions through which the line passes are arranged in the order from each high-temperature region to the lowest high-temperature region to form the temperature change sequence of each high-temperature region in each infrared image at each time point.

4. The fixed and mobile coordinated detection method for highway tunnel fires as described in claim 1, characterized in that, The process of obtaining the fire probability coefficient is as follows: the temperature drift coefficient is mapped to a second positive number, and the fire probability coefficient is the ratio of the spatiotemporal characteristic value to the second positive number.

5. The fixed and mobile coordinated detection method for highway tunnel fires as described in claim 1, characterized in that, The process of obtaining the suspected fire area is as follows: The segmentation threshold of the fire probability coefficient of all high-temperature areas in all infrared images at each time point is obtained, and high-temperature areas with fire probability coefficient greater than or equal to the segmentation threshold are regarded as suspected fire areas.

6. The fixed and mobile coordinated detection method for highway tunnel fires as described in claim 1, characterized in that, The deployment of mobile detection drones to capture videos of suspected fire areas is used to assess the fire situation within the tunnel to be detected, including: The system acquires the real-time location of each mobile detection drone within the tunnel to be detected; it also acquires the location of the monitoring point corresponding to each suspected fire area and records it as the monitoring point for each suspected fire area. Control the mobile detection drones that are closest to the monitoring points in each suspected fire area, and send them to the location of the monitoring points in each suspected fire area to capture video of the scene. This video will help staff determine whether a fire has occurred in the tunnel to be detected and whether the tunnel fire emergency plan should be activated.

7. A fixed and mobile coordinated detection system for highway tunnel fires, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the fixed and mobile collaborative detection method for highway tunnel fires as described in any one of claims 1-6.

Images