Tunnel water seepage and inrush monitoring system based on image recognition and thermal infrared

Abstract

The present invention discloses a tunnel water seepage and water gushing monitoring system based on image recognition and thermal infrared, which belongs to the field of tunnel water seepage and water gushing monitoring and the field of image processing technology. The system of the present invention includes an image acquisition module, an image preprocessing module, a tunnel water seepage and water gushing real-time analysis module and an early warning module; the image acquisition module is used to use a CMO industrial high-speed camera and an optical combined thermal infrared camera to respectively acquire high-definition optical images and thermal infrared images inside the tunnel, and transmit the acquired images to the image preprocessing module; the image preprocessing module is used to receive the images transmitted by the image acquisition module, perform preprocessing according to the image type, and transmit the obtained preprocessing results to the tunnel water seepage and water gushing real-time analysis module; the tunnel water seepage and water gushing real-time analysis module receives the preprocessing results transmitted by the image preprocessing module, and the tunnel water seepage and water gushing real-time analysis system analyzes the preprocessing results.

Description

Technical Field

[0001] The invention relates to the fields of tunnel seepage and water gushing monitoring and image processing technology, and in particular to a tunnel seepage and water gushing monitoring system based on the combination of image recognition and thermal infrared. Background Art

[0002] Tunnel seepage and water inrush is one of the main disasters that threaten the safety of tunnel structures and facilities. Due to the structural characteristics of tunnels such as small space and difficult evacuation, once seepage and water inrush occurs, it will spread very quickly, causing loose debris and high-pressure water to appear in the tunnel in the form of mud, which in turn leads to lining erosion, tunnel bottom sinking, and corrosion of electrical equipment, seriously affecting the safe operation of the tunnel, and causing great destructiveness and danger to the social economy. Therefore, it is extremely necessary to actively explore effective tunnel seepage and water inrush monitoring methods and monitoring systems, and scientifically apply prevention and control measures.

[0003] In the past, the detection method for tunnel seepage and water gushing disasters was manual inspection, that is, personnel were regularly dispatched to enter the tunnel to conduct manual visual inspections or use cameras to determine whether there was a seepage or water gushing incident in the tunnel. This method has a relatively slow detection speed, a small coverage area, and is easily affected by subjective factors, resulting in false detection and missed detection. It is even more impossible to conduct real-time monitoring and early warning of tunnel seepage and water gushing incidents, and it is impossible to intelligently and accurately locate the location of seepage and water gushing, which reduces the accuracy and reliability of tunnel seepage and water gushing monitoring, has a high false alarm rate, and when a seepage or water gushing incident is detected, the monitoring system needs to react for a period of time before responding. During the reaction time, the seepage and water gushing situation gradually worsens, reducing the monitoring effect of the monitoring system. Summary of the invention

[0004] The purpose of the present invention is to provide a tunnel water seepage and water inrush monitoring system based on the combination of image recognition and thermal infrared to solve the problems raised in the above background technology.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: the system includes an image acquisition module, an image preprocessing module, a tunnel water seepage and gushing real-time analysis module and an early warning module;

[0006] The image acquisition module is used to use a CMO industrial high-speed camera and an optical combined thermal infrared camera to respectively acquire high-definition optical images and thermal infrared images of the interior of the tunnel, and transmit the acquired images to an image preprocessing module;

[0007] The image preprocessing module is used to receive the images transmitted by the image acquisition module, perform preprocessing according to the image type, and transmit the obtained preprocessing results to the tunnel seepage and water inrush real-time analysis module;

[0008] The tunnel water seepage and water gushing real-time analysis module is used to receive the preprocessing results transmitted by the image preprocessing module. The tunnel water seepage and water gushing real-time analysis system analyzes the preprocessing results, performs fusion calculation on the analysis results to obtain accurate target point positioning, water seepage and water gushing amount and flow rate, and transmits the obtained target point positioning, water seepage and water gushing amount and flow rate to the early warning module;

[0009] The early warning module is used to receive the target point positioning, seepage and water inrush volume and flow rate transmitted by the tunnel water seepage and water inrush real-time analysis module, determine the tunnel water seepage and water inrush level according to the received content, issue early warning information based on the determination result, and match corresponding prevention and control measures according to the early warning information.

[0010] Further, the image acquisition module includes a high-definition optical image acquisition unit and a thermal infrared image acquisition unit;

[0011] Place a strong light source at the tunnel entrance and exit, install and debug CMOS industrial high-speed cameras, optical combined thermal infrared cameras, thermal infrared image collectors, optical image collectors and computer terminals, and confirm that the monitoring environment is safe. The light source, CMOS industrial high-speed camera and optical combined thermal infrared camera are all laid out within 10 meters from the tunnel entrance and exit. This layout plan meets the requirements of the Technical Specifications for Operation Monitoring of Urban Rail Transit Facilities and the Technical Specifications for Operation of Urban Rail Transit. The device and light source are laid out within 10 meters from the tunnel entrance and exit to meet the high-precision requirements of optical images and thermal infrared images inside the tunnel;

[0012] The high-definition optical image acquisition unit uses a CMOS industrial high-speed camera to acquire high-definition optical images inside the tunnel, and transmits the acquired high-definition optical images to the image preprocessing module through an image data transmission device;

[0013] The thermal infrared image acquisition unit uses an optically combined thermal infrared camera to acquire thermal infrared images inside the tunnel, and transmits the acquired thermal infrared images to the image preprocessing module through an image data transmitter.

[0014] Further, the image preprocessing module includes a high-definition optical image preprocessing unit and a thermal infrared image preprocessing unit;

[0015] The high-definition optical image preprocessing unit receives the high-definition optical image transmitted by the high-definition optical image acquisition unit, converts the high-definition optical image data into a regional pixel subset, extracts the abnormal region feature value in the regional pixel subset, and transmits the extracted abnormal region feature value to the tunnel water seepage and water inrush real-time analysis module;

[0016] The thermal infrared image preprocessing unit receives the thermal infrared image transmitted by the thermal infrared image acquisition unit, converts the thermal infrared image information into temperature data, selects tunnel temperature data of a certain period of time to fit into a real-time temperature cloud map, and transmits the real-time temperature cloud map to the tunnel water seepage and water gushing real-time analysis module.

[0017] Furthermore, the high-definition optical image preprocessing unit converts the high-definition optical image data into a regional pixel subset, and the specific method for extracting the abnormal region feature value in the regional pixel subset is:

[0018] The high-definition optical image data is converted into pixel data, and the surrounding pixel data that is basically consistent is divided into an area. By comparing the changes in pixel values ​​in each area at different times, the position with the largest change in pixel value is found, and the pixel feature value of the position is extracted.

[0019] Furthermore, the thermal infrared image preprocessing unit converts the thermal infrared image information into temperature data, and the specific method of selecting the tunnel temperature data of a certain period of time to fit into the real-time temperature cloud map is:

[0020] ①. Grayscale processing is performed on the thermal infrared image, and the thermal infrared image is converted into grayscale values ​​of 1-6 levels. Different levels represent different temperatures, and the larger the level, the higher the temperature;

[0021] ②. Based on the grayscale value in ①, the tunnel temperature data within a certain period of time is obtained. When the grayscale value corresponding to the temperature data is 1, the temperature data is represented by blue. When the grayscale value corresponding to the temperature data is 2, the temperature data is represented by cyan. When the grayscale value corresponding to the temperature data is 3, the temperature data is represented by green. When the grayscale value corresponding to the temperature data is 4, the temperature data is represented by yellow. When the grayscale value corresponding to the temperature data is 5, the temperature data is represented by orange. When the grayscale value corresponding to the temperature data is 6, the temperature data is represented by red. According to the temperature data and color division, the temperature data is converted into corresponding color codes to obtain a real-time temperature cloud map.

[0022] Furthermore, the tunnel water seepage and water gushing real-time analysis module includes an optical monitoring unit, a thermal infrared monitoring unit, a fusion calculation unit and a tunnel water seepage and water gushing situation analysis unit;

[0023] The optical monitoring unit receives the abnormal area characteristic value transmitted by the high-definition optical image preprocessing unit, identifies and tracks the abnormal area characteristic value, calculates the position, seepage and gushing amount and flow rate of the suspected water seepage and gushing target point inside the tunnel, and transmits the calculated position, seepage and gushing amount and flow rate of the suspected water seepage and gushing target point inside the tunnel to the fusion calculation unit;

[0024] The thermal infrared monitoring unit receives the real-time temperature cloud map transmitted by the thermal infrared image preprocessing unit, identifies the real-time temperature cloud map, locates the position and range of the abnormal temperature area inside the tunnel, and transmits the position and range of the abnormal temperature area inside the tunnel to the fusion calculation unit;

[0025] The fusion calculation unit receives the position, amount and flow rate of the suspected water seepage and gushing target point inside the tunnel transmitted by the optical monitoring unit and the position and range of the abnormal temperature area inside the tunnel transmitted by the thermal infrared monitoring unit, performs fusion calculation on the received content, and transmits the calculation result, amount and flow rate of water seepage and gushing to the tunnel water seepage and gushing situation analysis unit;

[0026] The tunnel water seepage and water gushing analysis unit receives the calculation results, water seepage and water gushing volume and flow rate transmitted by the fusion calculation unit, eliminates the error area and abnormal conditions according to the calculation results, obtains the precise positioning of the water seepage and water gushing target point in the tunnel, and transmits the precise position of the water seepage and water gushing target point in the tunnel, the water seepage and water gushing volume and flow rate to the early warning module.

[0027] Furthermore, the specific method for the thermal infrared monitoring unit to identify the real-time temperature cloud map is:

[0028] Step 1: Construct a coordinate system with the horizontal position of the tunnel as the horizontal axis and the vertical height of the tunnel as the vertical axis. Different colors in the real-time temperature cloud map represent the temperature distribution. The colors include red, orange, yellow, green, cyan, and blue. The colors from blue to red represent the temperature from low to high. Among them, the background temperature of the tunnel is in the red and orange range, and the seepage water temperature is in the cyan and blue range. Therefore, the temperature at the seepage water location differs from the background temperature of the tunnel by three ranges, so the temperature at the seepage water location is significantly lower than the background temperature of the tunnel.

[0029] Step 2: Based on Step 1, the location of the seepage target point can be accurately identified in the temperature cloud map. In addition, the temperature cloud map can show the flow of water over time and identify temperature anomaly areas with certain temporal and spatial distribution patterns in the cloud map.

[0030] Furthermore, the specific steps of the optical monitoring unit for determining the position, amount and velocity of the suspected water seepage and gushing target point inside the tunnel, and the fusion calculation unit for performing fusion calculation are as follows:

[0031] Step 1. Separately capture the suspected water seepage and water gushing target points monitored by the optical image data and the temperature abnormality areas monitored by the thermal infrared image;

[0032] Step 2. Fusion and comparison of the position and range of the abnormal area of ​​the two images, calculate the relative error of the target area of ​​the two images. The specific calculation formula is:

[0033] Based on the coordinate system constructed in Step 1, find the center position of the intercepted target area, the farthest position and the closest position of the edge of the target area from the center position, then the coordinates corresponding to the center position are (x1, y1), (x′, y′), and the coordinates corresponding to the closest position and the farthest position of the edge of the target area from the center position are (x2, y2), (x″, y″), (x3, y3), (x″′, y″′);

[0034] The specific calculation formula of relative error ε is:

[0035]

[0036]

[0037] Wherein, d represents the distance between the center positions of the two target areas. When d=0, it means that the two target areas are in the same position. When d≠0, it means that the two target areas are not in the same position. S represents the area of ​​the intercepted target area with (x1, y1) as the center. S′ represents the area of ​​the intercepted target area with (x′, y′) as the center. Respectively represent the circle centered at (x′, y′) and (x1, y1). is the area of ​​a circle with a radius of , and λ represents the area of ​​a circle with (x′, y′) as the center. is the ratio of the target area area to S', β represents the calculated value with (x', y') as the center, is the ratio of the target area of ​​the radius to S, and ε represents the relative error between the two target areas;

[0038] Step 3: Based on step 2, the specific formula for calculating the position, seepage and inrush volume and flow rate of the suspected water inrush target point inside the tunnel is:

[0039] The center position of the suspected water seepage target point inside the tunnel is (x1, y1), and the diffusion range of the target point is obtained by tracking the temperature change of the corresponding position of the real-time temperature cloud map;

[0040] The flow velocity v at the target point of suspected water seepage and inrush inside the tunnel is:

[0041]

[0042] Among them, v represents the flow velocity of seepage water at the corresponding moment;

[0043] The amount of water seepage Q at the target point of suspected water seepage inside the tunnel is:

[0044] Q = Sv;

[0045] Where v represents the flow velocity of the seepage water at the corresponding moment, and S represents the area of ​​the target area intercepted with (x1, y1) as the center;

[0046] Step 4. Set the error threshold based on the field test. If the relative error value is less than the set threshold, the suspected target is set as the actual target and accurately positioned. If the relative error value is greater than the set threshold, it is marked as a detection anomaly and the anomaly is eliminated from the detection results.

[0047] Furthermore, the early warning module includes a receiving unit, an early warning issuing unit and a water seepage and water inrush solving unit;

[0048] The receiving unit is used to receive the precise tunnel seepage and gushing target point position, seepage and gushing volume and flow rate transmitted by the tunnel seepage and gushing situation analysis unit, and upload the received content to the monitoring center, which makes an accurate judgment on the tunnel seepage and gushing related level and transmits the judgment result to the early warning issuing unit;

[0049] The warning issuing unit receives the judgment result transmitted by the receiving unit, issues warning information according to the judgment result, and transmits the issued warning information and the judgment result to the water seepage and water inrush solving unit;

[0050] The water seepage and gushing solution unit receives the warning information and judgment results transmitted by the warning issuing unit, and the monitoring personnel specify matching prevention and control measures in real time according to the judged water seepage and gushing level to ensure effective control of the water seepage and gushing incidents and timely rescue of personnel in the shortest time.

[0051] Furthermore, the method for determining the relevant level of tunnel seepage and water gushing in the water seepage and water gushing solution unit is as follows:

[0052] 1) Taking the tunnel background temperature as the reference value, if the monitored temperature anomaly area is less than 0.2m and the temperature change is greater than 0.5℃, the monitoring system determines that local water seepage has occurred in the tunnel;

[0053] 2) Taking the background temperature of the tunnel as the reference value, if the temperature abnormality area is between 0.2m and 2m and the temperature change is >0.5℃, the monitoring system determines that there is local water leakage in the tunnel;

[0054] 3) Taking the tunnel background temperature as the reference value, if the temperature anomaly area is >2m and the temperature change is >0.5℃, the monitoring system determines that a small-scale water inrush has occurred in the tunnel;

[0055] 4) Taking the tunnel background temperature as the reference value, if the temperature anomaly area is >10m and the temperature change is >0.5℃, the monitoring system will determine that a large-scale gushing or flooding has occurred in the tunnel.

[0056] Compared with the prior art, the beneficial effects achieved by the present invention are:

[0057] 1. Compared with the traditional single monitoring method, the present invention organically combines the two monitoring methods of optical imaging and thermal infrared imaging, and integrates and compares the above two monitoring results through the intelligent analysis system to eliminate the occurrence of false detection and missed detection, so as to achieve accurate quantitative analysis of the target point, providing double insurance for the monitoring of tunnel seepage and water gushing, reducing the missed detection rate and error rate, and making the positioning more accurate and the troubleshooting more accurate.

[0058] 2 Compared with traditional monitoring methods that can only effectively identify obvious water seepage and gushing in the tunnel, the thermal infrared monitoring system included in the present invention can monitor the tunnel wall temperature in real time and is highly sensitive to water seepage and gushing. The application of this monitoring method can accurately detect water seepage and gushing in the early stage, thereby effectively improving the response time, accuracy and efficiency to water seepage and gushing disasters in the tunnel, and has the significant advantages of high real-time, high precision and high adaptability.

[0059] 3. With the combined monitoring method of optical images and thermal infrared, in addition to accurately locating the location of tunnel seepage and gushing, the amount and flow rate of seepage and gushing water can also be obtained with the help of real-time calculation of the monitoring system, providing assistance for timely prevention and control of hidden dangers.

[0060] 4. The invention proposes a joint monitoring technology of optical image and thermal infrared imaging, which is a non-contact water seepage alarm technology. The system can be arranged at the entrance and exit of the tunnel. The optical image module and thermal infrared image module used are both self-enclosed designs, which not only ensure the monitoring accuracy, but also enhance the flexibility of the system and the convenience of installation. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0062] Figure 1 It is a schematic diagram of the overall process of the tunnel water seepage and water inrush monitoring system based on image recognition and thermal infrared combination of the present invention;

[0063] Figure 2 It is a structural schematic diagram of a tunnel water seepage and inrush monitoring system based on image recognition and thermal infrared combination according to the present invention;

[0064] Figure 3 It is a structural schematic diagram of a high-precision optical image module and a thermal infrared image module of a tunnel water seepage and water inrush monitoring system based on image recognition and thermal infrared combination of the present invention;

[0065] Figure 4 It is a temperature cloud map of the tunnel water seepage and water inrush monitoring system based on image recognition and thermal infrared combined with the present invention;

[0066] Figure 5 It is an optical image recognition system of the tunnel water seepage and water inrush monitoring system based on the combination of image recognition and thermal infrared of the present invention;

[0067] Figure 6 It is a schematic diagram of the flow of the fusion calculation of optical images and thermal infrared imaging of the tunnel water seepage and water inrush monitoring system based on the combination of image recognition and thermal infrared of the present invention.

[0068] In the figure: 1. Optical combined thermal infrared camera; 2. Image data transmitter; 3. Computer terminal; 4. Light source; 5. Tunnel entrance and exit; 6. Camera bracket; 7. Thermal infrared lens; 8. Optical lens; 9. Thermal infrared image collector; 10. Optical image collector; 11. Thermal infrared data transmission cable; 12. Optical data transmission cable. DETAILED DESCRIPTION

[0069] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0070] See also Figure 1-Figure 6 , the present invention provides a technical solution: the system includes an image acquisition module S1, an image preprocessing module S2, a tunnel water seepage and gushing real-time analysis module S3 and an early warning module S4;

[0071] The image acquisition module S1 is used to use a CMOS industrial high-speed camera and an optical combined thermal infrared camera to respectively acquire high-definition optical images and thermal infrared images inside the tunnel, and transmit the acquired images to the image preprocessing module S2; the image acquisition module S1 includes a high-definition optical image acquisition unit S11 and a thermal infrared image acquisition unit S12;

[0072] Place a strong light source at the tunnel entrance and exit, install and debug CMOS industrial high-speed cameras, optical combined thermal infrared cameras, thermal infrared image collectors, optical image collectors and computer terminals, and confirm that the monitoring environment is safe. The light source, CMOS industrial high-speed camera and optical combined thermal infrared camera are all laid out within 10 meters from the tunnel entrance and exit. This layout plan meets the requirements of the Technical Specifications for Operation Monitoring of Urban Rail Transit Facilities and the Technical Specifications for Operation of Urban Rail Transit. The device and light source are laid out within 10 meters from the tunnel entrance and exit to meet the high-precision requirements of optical images and thermal infrared images inside the tunnel;

[0073] The high-definition optical image acquisition unit S11 uses a CMOS industrial high-speed camera to acquire high-definition optical images inside the tunnel, and transmits the acquired high-definition optical images to the image preprocessing module S2 through an image data transmission device;

[0074] The thermal infrared image acquisition unit S12 uses an optically combined thermal infrared camera to acquire thermal infrared images inside the tunnel, and transmits the acquired thermal infrared images to the image preprocessing module S2 through an image data transmitter.

[0075] The image preprocessing module S2 is used to receive the image transmitted by the image acquisition module S1, perform preprocessing according to the image type, and transmit the obtained preprocessing results to the tunnel water seepage and water inrush real-time analysis module S3; the image preprocessing module S2 includes a high-definition optical image preprocessing unit S21 and a thermal infrared image preprocessing unit S22;

[0076] The high-definition optical image preprocessing unit S21 receives the high-definition optical image transmitted by the high-definition optical image acquisition unit S11, converts the high-definition optical image data into a regional pixel subset, extracts the abnormal region feature value in the regional pixel subset, and transmits the extracted abnormal region feature value to the tunnel water seepage and water inrush real-time analysis module S3. The specific method is: converting the high-definition optical image data into pixel data, dividing the surrounding pixel data that are basically consistent into an area, finding the position where the pixel value changes greatly by comparing the changes in the pixel values ​​of each area at different times, and extracting the pixel feature value of the position;

[0077] The thermal infrared image preprocessing unit S22 receives the thermal infrared image transmitted by the thermal infrared image acquisition unit S12, converts the thermal infrared image information into temperature data, selects the tunnel temperature data of a certain period of time to fit into a real-time temperature cloud map, and transmits the real-time temperature cloud map to the tunnel seepage and water inrush real-time analysis module S3. The specific method is as follows:

[0078] ①. Grayscale processing is performed on the thermal infrared image, and the thermal infrared image is converted into grayscale values ​​of 1-6 levels. Different levels represent different temperatures, and the larger the level, the higher the temperature;

[0079] ②. Based on the grayscale value in ①, the tunnel temperature data within a certain period of time is obtained. When the grayscale value corresponding to the temperature data is 1, the temperature data is represented by blue. When the grayscale value corresponding to the temperature data is 2, the temperature data is represented by cyan. When the grayscale value corresponding to the temperature data is 3, the temperature data is represented by green. When the grayscale value corresponding to the temperature data is 4, the temperature data is represented by yellow. When the grayscale value corresponding to the temperature data is 5, the temperature data is represented by orange. When the grayscale value corresponding to the temperature data is 6, the temperature data is represented by red. According to the temperature data and color division, the temperature data is converted into corresponding color codes to obtain a real-time temperature cloud map.

[0080] The tunnel water seepage and water gushing real-time analysis module S3 is used to receive the preprocessing results transmitted by the image preprocessing module S2. The tunnel water seepage and water gushing real-time analysis system analyzes the preprocessing results, performs fusion calculation on the analysis results to obtain accurate target point positioning, water seepage and water gushing amount and flow rate, and transmits the obtained target point positioning, water seepage and water gushing amount and flow rate to the early warning module S4; the tunnel water seepage and water gushing real-time analysis module S3 includes an optical monitoring unit S31, a thermal infrared monitoring unit S32, a fusion calculation unit S33 and a tunnel water seepage and water gushing situation analysis unit S34;

[0081] The optical monitoring unit S31 receives the abnormal area characteristic value transmitted by the high-definition optical image preprocessing unit S21, identifies and tracks the abnormal area characteristic value, calculates the position, seepage and gushing amount and flow rate of the suspected water seepage and gushing target point inside the tunnel, and transmits the calculated position, seepage and gushing amount and flow rate of the suspected water seepage and gushing target point inside the tunnel to the fusion calculation unit S33;

[0082] The thermal infrared monitoring unit S32 receives the real-time temperature cloud map transmitted by the thermal infrared image preprocessing unit S22, identifies the real-time temperature cloud map, locates the position and range of the abnormal temperature area inside the tunnel, and transmits the position and range of the abnormal temperature area inside the tunnel to the fusion calculation unit S33. The specific method for the thermal infrared monitoring unit S32 to identify the real-time temperature cloud map is:

[0083] Step 1: Construct a coordinate system with the horizontal position of the tunnel as the horizontal axis and the vertical height of the tunnel as the vertical axis. Different colors in the real-time temperature cloud map represent the temperature distribution. The colors include red, orange, yellow, green, cyan, and blue. The colors from blue to red represent the temperature from low to high. Among them, the background temperature of the tunnel is in the red and orange range, and the seepage water temperature is in the cyan and blue range. Therefore, the temperature at the seepage water location differs from the background temperature of the tunnel by three ranges, so the temperature at the seepage water location is significantly lower than the background temperature of the tunnel.

[0084] Step 2: Based on Step 1, the location of the seepage target point can be accurately identified in the temperature cloud map. In addition, the temperature cloud map can show the flow of water over time and identify the temperature anomaly area with a certain temporal and spatial distribution pattern in the cloud map;

[0085] The fusion calculation unit S33 receives the position, amount and flow rate of the suspected water seepage and gushing target point inside the tunnel transmitted by the optical monitoring unit S31 and the position and range of the abnormal temperature area inside the tunnel transmitted by the thermal infrared monitoring unit S32, performs fusion calculation on the received content, and transmits the calculation result, amount and flow rate of water seepage and gushing to the tunnel water seepage and gushing situation analysis unit S34;

[0086] The tunnel water seepage and gushing situation analysis unit S34 receives the calculation results, water seepage and gushing amount and flow rate transmitted by the fusion calculation unit S33, removes the error area and abnormal conditions according to the calculation results, obtains the precise positioning of the water seepage and gushing target point in the tunnel, and transmits the precise tunnel internal water seepage and gushing target point position, water seepage and gushing amount and flow rate to the early warning module S4;

[0087] The specific steps of the optical monitoring unit S31 for calculating the position, seepage and gushing amount and flow rate of the target point suspected of water seepage and gushing inside the tunnel, and the fusion calculation unit S33 for performing fusion calculation are as follows:

[0088] Step 1. Separately capture the suspected water seepage and water gushing target points monitored by the optical image data and the temperature abnormality areas monitored by the thermal infrared image;

[0089] Step 2. Fusion and comparison of the position and range of the abnormal area of ​​the two images, calculate the relative error of the target area of ​​the two images. The specific calculation formula is:

[0090] Based on the coordinate system constructed in Step 1, find the center position of the intercepted target area, the farthest position and the closest position of the edge of the target area from the center position, then the coordinates corresponding to the center position are (x1, y1), (x′, y′), and the coordinates corresponding to the closest position and the farthest position of the edge of the target area from the center position are (x2, y2), (x″, y″), (x3, y3), (x″′, y″′);

[0091] The specific calculation formula of relative error ε is:

[0092]

[0093]

[0094] Wherein, d represents the distance between the center positions of the two target areas. When d=0, it means that the two target areas are in the same position. When d≠0, it means that the two target areas are not in the same position. S represents the area of ​​the intercepted target area with (x1, y1) as the center. S′ represents the area of ​​the intercepted target area with (x′, y′) as the center. Respectively represent the circle centered at (x′, y′) and (x1, y1). is the area of ​​a circle with a radius of , and λ represents the area of ​​a circle with (x′, y′) as the center. is the ratio of the target area area to S', β represents the calculated value with (x', y') as the center, is the ratio of the target area of ​​the radius to S, and ε represents the relative error between the two target areas;

[0095] Step 3: Based on step 2, the specific formula for calculating the position, seepage and inrush volume and flow rate of the suspected water inrush target point inside the tunnel is:

[0096] The center position of the suspected water seepage target point inside the tunnel is (x1, y1), and the diffusion range of the target point is obtained by tracking the temperature change of the corresponding position of the real-time temperature cloud map;

[0097] The flow velocity v at the target point of suspected water seepage and inrush inside the tunnel is:

[0098]

[0099] Among them, v represents the flow velocity of seepage water at the corresponding moment;

[0100] The amount of water seepage Q at the target point of suspected water seepage inside the tunnel is:

[0101] Q = Sv;

[0102] Where v represents the flow velocity of the seepage water at the corresponding moment, and S represents the area of ​​the target area intercepted with (x1, y1) as the center;

[0103] Step 4. Set the error threshold based on the field test. If the relative error value is less than the set threshold, the suspected target is set as the actual target and accurately positioned. If the relative error value is greater than the set threshold, it is marked as a detection anomaly and the anomaly is eliminated from the detection results.

[0104] The early warning module S4 is used to receive the target point location, seepage and gushing volume and flow rate transmitted by the tunnel seepage and gushing real-time analysis module S3, determine the tunnel seepage and gushing level according to the received content, issue early warning information based on the determination result, and match corresponding prevention and control measures according to the early warning information; the early warning module S4 includes a receiving unit S41, an early warning issuing unit S42 and a seepage and gushing solution unit S43;

[0105] The receiving unit S41 is used to receive the precise tunnel seepage target point location, seepage volume and flow rate transmitted by the tunnel seepage analysis unit S34, and upload the received content to the monitoring center. The monitoring center makes an accurate judgment on the tunnel seepage level and transmits the judgment result to the early warning release unit S42.

[0106] The warning issuing unit S42 receives the judgment result transmitted by the receiving unit S41, issues warning information according to the judgment result, and transmits the issued warning information and the judgment result to the water seepage and water inrush solving unit S43;

[0107] The seepage and gushing solution unit S43 receives the warning information and judgment results transmitted by the warning issuing unit S42. The monitoring personnel specify matching prevention and control measures in real time according to the judged seepage and gushing level to ensure effective control of the seepage and gushing incident and timely rescue of personnel in the shortest time. The determination method of the tunnel seepage and gushing related levels in the seepage and gushing solution unit S43 is as follows:

[0108] 1) Taking the tunnel background temperature as the reference value, if the monitored temperature anomaly area is less than 0.2m and the temperature change is greater than 0.5℃, the monitoring system determines that local water seepage has occurred in the tunnel;

[0109] 2) Taking the background temperature of the tunnel as the reference value, if the temperature abnormality area is between 0.2m and 2m and the temperature change is >0.5℃, the monitoring system determines that there is local water leakage in the tunnel;

[0110] 3) Taking the tunnel background temperature as the reference value, if the temperature anomaly area is >2m and the temperature change is >0.5℃, the monitoring system determines that a small-scale water inrush has occurred in the tunnel;

[0111] 4) Taking the tunnel background temperature as the reference value, if the temperature anomaly area is >10m and the temperature change is >0.5℃, the monitoring system will determine that a large-scale gushing or flooding has occurred in the tunnel.

[0112] 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.

[0113] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A tunnel water seepage and flooding monitoring system based on image recognition and thermal infrared, characterized by: The system comprises an image acquisition module (S1), an image preprocessing module (S2), a tunnel water seepage and water gushing real-time analysis module (S3) and an early warning module (S4); The image acquisition module (S1) is used to respectively acquire high-definition optical images and thermal infrared images inside the tunnel using a CMOS industrial high-speed camera and an optical combined thermal infrared camera, and transmit the acquired images to an image preprocessing module (S2); The image acquisition module (S1) comprises a high-definition optical image acquisition unit (S11) and a thermal infrared image acquisition unit (S12); Place a strong light source at the tunnel entrance and exit, install and debug CMOS industrial high-speed cameras, optical combined thermal infrared cameras, thermal infrared image collectors, optical image collectors and computer terminals, and confirm that the monitoring environment is safe. The light source, CMOS industrial high-speed camera and optical combined thermal infrared camera are all laid out within 10 meters from the tunnel entrance and exit; The high-definition optical image acquisition unit (S11) uses a CMOS industrial high-speed camera to acquire a high-definition optical image inside the tunnel, and transmits the acquired high-definition optical image to the image preprocessing module (S2) through an image data transmission device; The thermal infrared image acquisition unit (S12) uses an optically combined thermal infrared camera to acquire a thermal infrared image inside the tunnel, and transmits the acquired thermal infrared image to the image preprocessing module (S2) via an image data transmission device; The image preprocessing module (S2) is used to receive the image transmitted by the image acquisition module (S1), perform preprocessing according to the image type, and transmit the obtained preprocessing results to the tunnel seepage and water inrush real-time analysis module (S3); The image preprocessing module (S2) includes a high-definition optical image preprocessing unit (S21) and a thermal infrared image preprocessing unit (S22); The high-definition optical image preprocessing unit (S21) receives the high-definition optical image transmitted by the high-definition optical image acquisition unit (S11), converts the high-definition optical image data into a regional pixel subset, and extracts the abnormal region feature value in the regional pixel subset, and the specific method is as follows: Convert high-definition optical image data into pixel data, divide the surrounding pixel data that are basically the same into an area, find the location where the pixel value changes greatly by comparing the changes in the pixel values ​​of each area at different times, and extract the pixel feature value of the location; The extracted abnormal area characteristic values ​​are transmitted to the tunnel water seepage and water inrush real-time analysis module (S3); The thermal infrared image preprocessing unit (S22) receives the thermal infrared image transmitted by the thermal infrared image acquisition unit (S12), converts the thermal infrared image information into temperature data, and selects tunnel temperature data of a certain period of time to fit into a real-time temperature cloud map. The specific method is as follows: ①. Grayscale processing is performed on the thermal infrared image, and the thermal infrared image is converted into grayscale values ​​of 1-6 levels. Different levels represent different temperatures, and the larger the level, the higher the temperature; ②. Based on the grayscale value in ①, the tunnel temperature data within a certain period of time is obtained. When the grayscale value corresponding to the temperature data is 1, the temperature data is represented by blue. When the grayscale value corresponding to the temperature data is 2, the temperature data is represented by cyan. When the grayscale value corresponding to the temperature data is 3, the temperature data is represented by green. When the grayscale value corresponding to the temperature data is 4, the temperature data is represented by yellow. When the grayscale value corresponding to the temperature data is 5, the temperature data is represented by orange. When the grayscale value corresponding to the temperature data is 6, the temperature data is represented by red. According to the temperature data and color division, the temperature data is converted into the corresponding color code to obtain a real-time temperature cloud map; The real-time temperature cloud map is transmitted to the tunnel water seepage and water inrush real-time analysis module (S3); The tunnel water seepage and water gushing real-time analysis module (S3) is used to receive the preprocessing results transmitted by the image preprocessing module (S2), and the tunnel water seepage and water gushing real-time analysis system analyzes the preprocessing results, performs fusion calculation on the analysis results to obtain accurate target point positioning, water seepage and water gushing amount and flow rate, and transmits the obtained target point positioning, water seepage and water gushing amount and flow rate to the early warning module (S4); The tunnel water seepage and water gushing real-time analysis module (S3) comprises an optical monitoring unit (S31), a thermal infrared monitoring unit (S32), a fusion calculation unit (S33) and a tunnel water seepage and water gushing situation analysis unit (S34); The optical monitoring unit (S31) receives the abnormal area characteristic value transmitted by the high-definition optical image preprocessing unit (S21), identifies and tracks the abnormal area characteristic value, calculates the position, seepage and gushing amount and flow rate of the suspected water seepage and gushing target point inside the tunnel, and transmits the calculated position, seepage and gushing amount and flow rate of the suspected water seepage and gushing target point inside the tunnel to the fusion calculation unit (S33); The thermal infrared monitoring unit (S32) receives the real-time temperature cloud map transmitted by the thermal infrared image preprocessing unit (S22), identifies the real-time temperature cloud map, locates the position and range of the abnormal temperature area inside the tunnel, and transmits the position and range of the abnormal temperature area inside the tunnel to the fusion calculation unit (S33); The fusion calculation unit (S33) receives the position of the suspected water seepage and gushing target point inside the tunnel transmitted by the optical monitoring unit (S31), the water seepage and gushing amount and flow rate, and the position and range of the abnormal temperature area inside the tunnel transmitted by the thermal infrared monitoring unit (S32), performs fusion calculation on the received content, and transmits the calculation result, water seepage and gushing amount and flow rate to the tunnel water seepage and gushing situation analysis unit (S34); The tunnel water seepage and gushing situation analysis unit (S34) receives the calculation results, water seepage and gushing amount and flow rate transmitted by the fusion calculation unit (S33), removes the error area and abnormal conditions according to the calculation results, obtains the precise positioning of the water seepage and gushing target point in the tunnel, and transmits the precise position of the water seepage and gushing target point in the tunnel, the water seepage and gushing amount and flow rate to the early warning module (S4); The early warning module (S4) is used to receive the target point location, seepage and gushing volume and flow rate transmitted by the tunnel water seepage and gushing real-time analysis module (S3), determine the tunnel water seepage and gushing level according to the received content, issue early warning information based on the determination result, and match corresponding prevention and control measures according to the early warning information.

2. The tunnel water seepage and water inrush monitoring system based on image recognition and thermal infrared according to claim 1 is characterized by: The specific method for the thermal infrared monitoring unit (S32) to identify the real-time temperature cloud map is: Step 1: Construct a coordinate system with the horizontal position of the tunnel as the horizontal axis and the vertical height of the tunnel as the vertical axis. Different colors in the real-time temperature cloud map represent the temperature distribution. The colors include red, orange, yellow, green, cyan, and blue. The colors from blue to red represent the temperature from low to high. Among them, the background temperature of the tunnel is in the red and orange range, and the seepage water temperature is in the cyan and blue range. Therefore, the temperature at the seepage water location differs from the background temperature of the tunnel by three ranges, so the temperature at the seepage water location is significantly lower than the background temperature of the tunnel. Step 2: Based on Step 1, the location of the seepage target point can be accurately identified in the temperature cloud map. In addition, the temperature cloud map can show the flow of water over time and identify temperature anomaly areas with certain temporal and spatial distribution patterns in the cloud map.

3. The tunnel water seepage and water inrush monitoring system based on image recognition and thermal infrared according to claim 2 is characterized by: The specific steps of the optical monitoring unit (S31) for determining the position, amount and velocity of the suspected water seepage and water gushing target point inside the tunnel, and the fusion calculation unit (S33) for performing fusion calculation are as follows: Step 1. Separately capture the suspected water seepage and water gushing target points monitored by the optical image data and the temperature abnormality areas monitored by the thermal infrared image; Step 2. Fusion and comparison of the position and range of the abnormal area of ​​the two images, calculate the relative error of the target area of ​​the two images. The specific calculation formula is: Based on the coordinate system constructed in Step 1, find the center position of the intercepted target area, the farthest position and the closest position of the edge of the target area from the center position, then the coordinates corresponding to the center position are (x1, y1), (x′, y′), and the coordinates corresponding to the closest position and the farthest position of the edge of the target area from the center position are (x2, y2), (x″, y″), (x3, y3), (x″′, y″′); The specific calculation formula of relative error ε is: Wherein, d represents the distance between the center positions of the two target areas. When d=0, it means that the two target areas are in the same position. When d≠0, it means that the two target areas are not in the same position. S represents the area of ​​the intercepted target area with (x1, y1) as the center. S′ represents the area of ​​the intercepted target area with (x′, y′) as the center. λ and β represent the proportional coefficients. ε represents the relative error between the two target areas. Step 3: Based on step 2, the specific formula for calculating the position, seepage and inrush volume and flow rate of the suspected water inrush target point inside the tunnel is: The center position of the suspected water seepage target point inside the tunnel is (x1, y1), and the diffusion range of the target point is obtained by tracking the temperature change of the corresponding position of the real-time temperature cloud map; The flow velocity v at the target point of suspected water seepage and inrush inside the tunnel is: Among them, v represents the flow velocity of seepage water at the corresponding moment; The amount of water seepage Q at the target point of suspected water seepage inside the tunnel is: Q = Sv; Where v represents the flow velocity of the seepage water at the corresponding moment, and S represents the area of ​​the target area intercepted with (x1, y1) as the center; Step 4. Set the error threshold based on the field test. If the relative error value is less than the set threshold, the suspected target is set as the actual target and accurately positioned. If the relative error value is greater than the set threshold, it is marked as a detection anomaly and the anomaly is eliminated from the detection results.

4. The tunnel water seepage and water inrush monitoring system based on image recognition and thermal infrared according to claim 3 is characterized by: The early warning module (S4) includes a receiving unit (S41), an early warning issuing unit (S42) and a water seepage and water gushing solution unit (S43); The receiving unit (S41) is used to receive the precise tunnel seepage and water inrush target point location, seepage and water inrush volume and flow rate transmitted by the tunnel seepage and water inrush situation analysis unit (S34), and upload the received content to the monitoring center, which makes an accurate judgment on the tunnel seepage and water inrush related level and transmits the judgment result to the early warning release unit (S42); The warning issuing unit (S42) receives the judgment result transmitted by the receiving unit (S41), issues warning information according to the judgment result, and transmits the issued warning information and the judgment result to the water seepage and water inrush solution unit (S43); The water seepage and gushing solution unit (S43) receives the warning information and judgment results transmitted by the warning issuing unit (S42), and the monitoring personnel specify matching prevention and control measures in real time according to the judged water seepage and gushing level.

5. The tunnel water seepage and flooding monitoring system based on image recognition and thermal infrared according to claim 4 is characterized in that: The method for determining the relevant level of tunnel seepage and water gushing in the water seepage and water gushing solution unit (S43) is as follows: 1) Taking the tunnel background temperature as the reference value, if the monitored temperature anomaly area is less than 0.2m and the temperature change is greater than 0.5℃, the monitoring system determines that local water seepage has occurred in the tunnel; 2) Taking the background temperature of the tunnel as the reference value, if the temperature abnormality area is between 0.2m and 2m and the temperature change is >0.5℃, the monitoring system determines that there is local water leakage in the tunnel; 3) Taking the tunnel background temperature as the reference value, if the temperature anomaly area is >2m and the temperature change is >0.5℃, the monitoring system determines that a small-scale water inrush has occurred in the tunnel; 4) Taking the tunnel background temperature as the reference value, if the temperature anomaly area is >10m and the temperature change is >0.5℃, the monitoring system will determine that a large-scale gushing or flooding has occurred in the tunnel.

Images