Research On Remote Sensing Detection Method Of Strip Underground Targets In Mountains

Underground target detection technology has important application value and can provide help for the construction and development of underground space.At the same time,the detection and positioning of underground mineral resources and sites can promote the sustainable use of mineral resources and effectively guide the development of archaeological work.In the face of the development trend of the underground development of major strategic facilities in countries around the world,the underground target detection technology is also of great military significance,which can improve the long-range precision strike ability of our army and guide the development of camouflage and construction of military facilities in China.At present,there are many underground target detection methods,among which the infrared remote sensing detection method has attracted much attention due to their wide-area,flexibility and good application prospects.Based on the infrared remote sensing technology,this thesis carries out the research on the detection and inversion methods of strip underground targets.Meanwhile,it realizes the detection of the localization of the vent using the hyperspectral anomaly detection algorithm.Firstly,according to the theory of heat transfer,the generation mechanism of infrared disturbance of underground target is analyzed,the mathematical model of background/target infrared information field is constructed,and the infrared detectability of underground target is explored in combination with simulation experiment.At the same time,the characteristic difference between the vents and the background are analyzed,and the detectability of the vent is studied based on the characteristics of hyperspectral data.Secondly,aiming at the interference of solar radiation and surface vegetation that the infrared detection method of underground target will be affected by,this thesis filters it out through theoretical analysis and modeling calculation.According to the principle of energy superposition,combining infrared radiation data and simulated radiation data to filter out the background energy,and the background noise is suppressed based on the distance criterion.A detection method for strip underground targets in mountain based on position correlation and gradient information is proposed,and the target detection position is obtained by fitting and correcting the detection results.And the proportion of the displacement detection result is 1.74%.Furthermore,the optimization theory is applied,combining real infrared radiation data and simulated radiation data to construct the objective function,and the improved conjugate gradient method is used to inversely solve the buried depth information of the target.The relative error of inversion is less than 3.8%.Thirdly,based on the differences between the vent and the background spectral characteristics and spatial characteristics,an improved low-rank sparse decomposition hyperspectral anomaly detection method combined with spatial correlation is proposed,and the location of the vent is detected by combining visible light data.The horizontal relative error between the detected location and the real location is 0.56 m.Finally,a research platform for detection and inversion of underground targets is designed and realized,and the demonstration and verification of detection and positioning functions of strip underground targets and vents in mountain are completed.