Concrete dam crack inspection method based on unmanned aerial vehicle carrying infrared combined optical camera

Abstract

The invention discloses a concrete dam crack inspection method based on an unmanned aerial vehicle carrying an infrared combined optical camera. Core equipment comprises a multi-rotor unmanned aerial vehicle carrying a global positioning system, an infrared thermal imager and a high-resolution optical camera. The invention aims to improve the concrete dam crack detection efficiency and accuracy. According to the method, an infrared camera and an optical camera are carried by an unmanned aerial vehicle to obtain dam surface temperature distribution and visible light images. By means of the infrared thermal imaging technology, a thermal abnormal area caused by the crack can be recognized, and the optical camera provides fine crack form and position information. In the inspection process, an infrared thermal imaging instrument is used for detecting the whole dam and positioning an abnormal area, and then a method of combining the infrared thermal imaging instrument and an optical camera is used for respectively detecting internal and external damages of the abnormal area. The method overcomes the problems of high labor cost, low detection efficiency and difficulty in complete coverage of a traditional inspection method, is especially suitable for rapid and accurate detection of large-area concrete dam surface cracks, and provides effective technical support for safety maintenance of dams.

Description

Technical Field

[0001] The present invention belongs to the field of dam safety detection and relates to a concrete dam crack inspection method based on a drone equipped with an infrared thermal imager combined with an optical camera. Background Art

[0002] As a large-scale water conservancy infrastructure, concrete dams undertake many important functions such as water storage, power generation, and flood control. The structural safety of concrete dams is of vital importance. However, cracks may occur during the construction of large-volume concrete. In addition, various forms of cracks may often appear on the surface of concrete dams due to the long-term influence of the natural environment and external loads during operation and service. If these cracks are not discovered and repaired in time, they may cause dam leakage, structural aging, and even more serious engineering accidents such as dam collapse. Therefore, regular and accurate inspection and monitoring of cracks in concrete dams is an important means to ensure their safe operation.

[0003] Traditional methods of detecting cracks in concrete dams mainly rely on manual inspections, usually performed by technicians through visual inspections or using portable instruments to measure the surface of the dam. This method has the following shortcomings:

[0004] (1) Low efficiency: Manual inspection involves visual inspection and instrument measurement of a large area, which is labor-intensive and time-consuming. Especially for large or inaccessible areas, the inspection efficiency is low;

[0005] (2) High subjectivity: Manual inspection is easily affected by the experience and judgment of the inspector, and there are subjective errors, especially when the crack is small or hidden in a visual blind spot, it may be overlooked;

[0006] (3) Incomplete coverage: Due to the large size and complex terrain of the dam, some areas are difficult to reach, especially at high altitudes or underwater. Manual inspections are difficult to achieve full coverage and are prone to missed inspections;

[0007] (4) High safety risk: Manual inspection requires technicians to come into close contact with the surface of the dam, especially when inspecting at high altitudes or near water, which poses a great safety hazard.

[0008] With the rapid development of drone technology, using drones equipped with various sensors for structural inspection has become a new trend. Drones have the advantages of flexibility, low cost, and wide coverage, and can efficiently complete inspection tasks in complex terrain. In particular, drones can fly at high altitudes and can easily cover inaccessible areas of the dam.

[0009] However, single optical imaging technology has certain limitations in crack detection. For example, optical cameras can only record information within the visible light range and cannot effectively detect potential hidden dangers in the internal structure of the dam. In addition, factors such as ambient light, shadows, and reflections may also affect the quality of optical images, resulting in reduced detection accuracy.

[0010] Infrared thermal imaging technology is a non-contact detection technology based on temperature distribution. It determines the temperature change by detecting the thermal radiation on the surface of the object, and can effectively identify the temperature abnormality area caused by cracks. The infrared thermal imaging method detects the internal cracks of concrete based on the principle of infrared radiation. When there are cracks inside the concrete, it will cause local changes in the thermal conductivity of the concrete. During the external thermal excitation process, the cracks will hinder the transfer of heat, making the temperature distribution inside the concrete uneven, which will eventually cause the surface temperature to change. Summary of the invention

[0011] The present invention provides a concrete dam crack inspection method based on an unmanned aerial vehicle equipped with an infrared thermal imager combined with an optical camera, so as to solve the problems of low efficiency, strong subjectivity, incomplete coverage and high safety risks of traditional manual inspection, and realize rapid and accurate detection and evaluation of dam cracks.

[0012] The technical solution adopted by the present invention is as follows:

[0013] A method for inspecting cracks in a concrete dam based on a drone equipped with an infrared combined with an optical camera, comprising:

[0014] Step 1: Establish an inspection system

[0015] The inspection system consists of a high-resolution optical camera and a high-sensitivity infrared thermal imager, which are mounted on a multi-rotor drone platform. Through a pre-planned route, the drone performs a full-coverage inspection task on the surface of the concrete dam. The drone's flight altitude h, the camera's field of view θ, and the image width L image The following formula must be met to ensure that the image overlap rate and resolution meet the detection requirements:

[0016]

[0017] Among them, Overlap is the image overlap rate, which is usually 60%-80%.

[0018] Step 2: Use the inspection system to obtain data

[0019] Infrared radiation is one of the most widespread electromagnetic wave radiations existing in nature. It is based on the fact that any object at normal temperature will have random motions of its molecules and atoms and continuously radiate infrared energy. The more intense the motions of the molecules and atoms, the greater the radiated energy; conversely, the smaller the radiated energy. Objects above 0°K will radiate infrared rays due to the motions of their own molecules. The famous Planck's law shows that there is a certain relationship among temperature, wavelength, and energy. The total infrared energy increases rapidly with the increase of temperature; the peak wavelength shifts towards the short wave with the increase of temperature. According to the Stefan-Boltzmann law, when the temperature changes, the total infrared energy is proportional to the fourth power of the absolute temperature. When there is a small change in temperature, it will cause a large change in the total energy, which can be expressed by the following formula: P = gσT 4 ; where: P is the radiation power (W / cm2); σ is a constant, σ = 5.673×10 -12 W / (cm 2 ·K 4 ); T is the thermodynamic temperature (K) of the object surface; g is the emissivity, 0 < g < 1. The paths of heat diffusion and transfer within the object will, due to different thermophysical properties of material conduction, either be blocked and piled up or pass through unobstructed, and finally corresponding "hot zones" and "cold zones" will be formed on the object surface. This temperature difference phenomenon from the inside to the surface is the basic principle of infrared detection.

[0020] During the flight of the unmanned aerial vehicle (UAV), the infrared thermal imager can detect temperature anomalies caused by cracks or material defects. First, use the infrared thermal imager to detect and monitor the entire dam and locate the abnormal areas; after identifying the abnormal areas, then use the UAV to approach the identified abnormal areas at close range. For the abnormal areas, use a method combining the infrared thermal imager and the optical camera to monitor the internal and external damages of the abnormal areas respectively. The optical camera captures the geometric details of the concrete surface and accurately records the crack morphology using its high-resolution characteristics. The geometric features such as the width and length of the cracks are quantified by calibrating the camera parameters:

[0021]

[0022] Among them, L is the actual crack length, p is the pixel size, D is the shooting distance, and f is the camera focal length.

[0023] Step 3: Real-time data fusion

[0024] During the flight of the UAV, synchronously collect infrared and optical data, and perform real-time fusion by combining the characteristics of the two imaging methods:

[0025] ① Preliminary infrared screening

[0026] By setting the temperature difference threshold ΔT th , screen out the areas where cracks may exist:

[0027] ΔT=T max -T min >ΔT th

[0028] The screened target areas are marked as suspected crack areas for further analysis by optical imaging.

[0029] ②Optical precision recognition

[0030] Detailed analysis of the optical image of the suspected area to extract the specific geometric parameters of the crack, including the length L crack , Width W crack , and combined with infrared information to verify the authenticity of the cracks.

[0031] The present invention is suitable for crack inspection of large concrete dams with complex terrain and difficult close access. It can not only overcome the defects of traditional manual inspection methods, but also improve inspection efficiency, coverage and accuracy, providing more advanced technical means for structural safety monitoring of dams. This method combines the flexibility of drones, the temperature detection capability of infrared thermal imaging, and the high-resolution imaging function of optical cameras, which can achieve efficient and accurate detection of dam cracks and provide strong guarantees for the safe operation of dams. Crack information can be directly obtained based on real-time complementary analysis of infrared and optical, providing an efficient and economical solution for dam structural health monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 Schematic diagram of equipping a drone with an infrared thermal imager and an optical camera;

[0033] Figure 2 This is a schematic diagram of the overall inspection of the dam using an infrared thermal imager;

[0034] Figure 3 Schematic diagram of infrared thermal imager and optical camera detection of local abnormal areas;

[0035] In the figure: 1-Global Positioning System (GPS); 2-High-resolution optical camera; 3-Infrared thermal imager

[0036] 4-Multi-rotor drone; 5-Dam; 6-Cracks in dam concrete; 7-Anomaly detection area; 8-Satellite. DETAILED DESCRIPTION

[0037] The present invention provides a method for monitoring dam concrete cracks using a drone equipped with infrared and optical cameras. The specific implementation scheme is as follows:

[0038] 1. UAV platform construction

[0039] like Figure 1The UAV-carrying platform includes: a multi-rotor UAV platform 1, which provides flexible flight capabilities; an infrared thermal imager 3, which is used to capture the temperature distribution on the concrete surface; a high-resolution optical camera 2, which is used to obtain crack geometry details; a global positioning system (GPS) to achieve precise positioning; and a real-time data transmission module to support data return and on-site analysis.

[0040] The thermal sensitivity of the infrared camera is required to reach NETD < 50mK to ensure the capture of small temperature differences. The resolution of the optical camera must be ≥ 20MP to ensure clear recording of the geometric characteristics of the cracks. The inspection height h and the route coverage rate Coverage meet the formula:

[0041]

[0042] Among them, Overlap is the image overlap rate, which is usually 60%-80%.

[0043] 2. Infrared detection method

[0044] The materials and structures of the concrete panels of a hydropower station are basically consistent within a certain range. When there is no abnormality in the panel, the overall heating is uniform. If the panel has cracks or other abnormalities, due to the different properties of the air and the dam body, heat will accumulate at the cracks during the heat conduction process, causing the temperature in this area to be higher than other parts. Therefore, infrared thermal imaging technology is used to obtain the temperature difference of the panel to analyze the crack condition of the dam. Infrared thermal imaging defect detection work is carried out in front of the hydropower station dam. In order to obtain better detection results, the concrete structure surface of the dam is photographed with an infrared thermal imager during the period of sunlight, such as Figure 2 The distribution of cracks in the dam was determined by analyzing the infrared thermal images obtained.

[0045] The temperature data of the concrete surface is collected by an infrared thermal imager to generate a thermal image sequence, which is then transferred to the software interface in the computer. The surface temperature distribution of the concrete can be observed through the thermal image on the software interface. Different temperature areas correspond to different colors. The brighter the color, the higher the temperature, and the darker the color, the lower the temperature. At the same time, data processing can be performed on the software interface to obtain the temperature distribution on the dam concrete surface, the changes in the surface temperature of the dam concrete over time corresponding to the places where there are cracks inside and where there are no cracks. By analyzing the results, the location of the cracks can be located and the size of the cracks can be determined.

[0046] 3. Optical camera and infrared thermal imager capture crack information

[0047] like Figure 3As shown in the figure, after identifying the abnormal temperature area 7, the optical camera and infrared thermal imager on the drone accurately photograph these areas, capture the morphological information of the cracks, and locate the abnormal area according to the positioning system 8 on the drone. The optical camera can record important information such as the length and width of the cracks, and the infrared thermal imaging can detect internal damage to the concrete. Combining the data and images collected by infrared thermal imaging and optical cameras, it is convenient for later staff to evaluate the severity of the cracks and make maintenance recommendations.

[0048] 4. Data Fusion

[0049] Infrared and optical data are collected synchronously during flight, and the characteristics of the two imaging methods are combined in real time:

[0050] ①Infrared preliminary screening,

[0051] By setting the temperature difference threshold ΔT th , filter out areas where cracks may exist:

[0052] ΔT=T max -T min >ΔT th

[0053] The screened target areas are marked as suspected crack areas for further analysis by optical imaging.

[0054] ②Optical precision identification:

[0055] Detailed analysis of the optical image of the suspected area to extract the specific geometric parameters of the crack, including the length L crack , Width W crack , and combined with infrared information to verify the authenticity of the cracks.

[0056] This technical solution is particularly suitable for crack inspection of concrete dams with large areas and complex geometric shapes. The combined detection method of infrared and optical can significantly improve the inspection efficiency and reduce the missed detection rate and misjudgment rate. In addition, this solution does not require complex post-processing of images. It can directly obtain crack information based on real-time complementary analysis of infrared and optical, providing an efficient and economical solution for dam structure health monitoring.

Claims

1. A method for inspecting cracks in concrete dams based on an unmanned aerial vehicle equipped with an infrared combined with an optical camera, characterized in that: Step 1: Establish an inspection system The inspection system consists of a high-resolution optical camera and a high-sensitivity infrared thermal imager, which are mounted on a multi-rotor UAV platform. Through a pre-planned route, the UAV performs a full-coverage inspection task on the surface of the concrete dam. The flight altitude h of the UAV, the field of view angle θ of the camera, and the image width L image The following formula must be met to ensure that the image overlap rate and resolution meet the detection requirements: Among them, Overlap is the image overlap rate, usually 60%-80%; Step 2: Use the inspection system to obtain data During the flight of the drone, the infrared thermal imager can detect temperature anomalies caused by cracks or material defects. First, the infrared thermal imager is used to monitor the entire dam and locate the abnormal area. After the abnormal area is identified, the drone is used to approach the identified abnormal area at a close distance. The infrared thermal imager and the optical camera are combined to monitor the internal and external damage of the abnormal area. The optical camera captures the geometric details of the concrete surface and uses its high-resolution characteristics to accurately record the crack morphology. The geometric features of the cracks, such as width and length, are quantified by calibrating the camera parameters: Among them, L is the actual crack length, p is the pixel size, D is the shooting distance, and f is the focal length of the camera; Step 3: Real-time data fusion: ①Infrared preliminary screening By setting the temperature difference threshold ΔT th , filter out areas where cracks may exist: ΔT=T max -T min >ΔT th The screened target areas are marked as suspected crack areas for further analysis by optical imaging; ②Optical precision recognition Detailed analysis of the optical image of the suspected area to extract the specific geometric parameters of the crack, including the length L crack , Width W crack , and combined with infrared information to verify the authenticity of the cracks.

Images