The Detection Of Airport Runways Area Based On Scattering Characteristics In Polsar Image

Polarimetric Synthetic Aperture Radar(PolSAR)is a high-resolution imaging radar that can obtain high-resolution radar images similar to optical photography in weather conditions with extremely low visibility.The relative movement of the radar and the target is used to synthesize a smaller real antenna aperture with a data processing method to synthesize a larger equivalent antenna aperture radar,also known as a comprehensive aperture radar.Synthetic aperture radar is characterized by high resolution,can work around the clock,and can effectively identify camouflage and penetration cover.Compared with other optical imaging systems,it can be protected from various weather conditions such as rain,dust,fog,and light.So that the target can be detected.Among the country’s important aviation infrastructures,as a typical facility carrying key tasks such as aircraft take-off and landing,positioning,and maintenance,the airport has extremely important strategic significance in various fields such as military and civil.In this paper,the runway is the core facility of the airport,and it is the most reference feature in the airport detection.The runway can be used to detect the airport.In this paper,based on the research on the scattering characteristics of PolSAR images,analysis of the physical characteristics and statistical information contained in PolSAR images,two algorithms are proposed to realize the airport runway area detection in PolSAR images.The first algorithm proposes a regional PolSAR airport runway detection method.First,the SLIC(Simple Linear Iterative Clustering)algorithm is used to perform super-pixel segmentation on polarized SAR images;then the new three-component decomposition and polarization scattering entropy are used to coarsely classify the images,and then the improved K-means clustering is used to combine iteration with the degree of difference The method completes the fine classification,and finally uses the runway’s scattering characteristics and geometric features to extract the complete airport area from the classification results.While improving the calculation efficiency,it also maintains complete image details.The second algorithm is improved in the traditional PolSAR airport runway identification part,and a new identification method is proposed.The PolSAR scattering feature is combined with the texture feature LBP(Local binary patterns)feature to identify the runway area.First,a new scattering feature is used to perform threshold segmentation on polarized SAR images.The LBP feature is constructed in this area and sent to the SVM classifier for classification.Then the airport runway is detected and identified,and a real airport is obtained Runway area.The above two algorithms were tested experimentally using multiple sets of polar PolSAR data collected by the UAVSAR system.The results show that both methods can effectively detect airport runways,have complete structures,clear edges,and low false alarm and missed alarm rates.However,the second method is more computationally efficient,takes less time,and has a lower false alarm detection rate in the target area.