PROJECT TITLE :

Urban Area SAR Image Man-Made Target Extraction Based on the Product Model and the Time–Frequency Analysis

ABSTRACT:

This paper proposed an innovative framework to almost automatically extract man-created target from a high-resolution (HR) polarimetric SAR (PolSAR) image of an urban area. The core half of this framework may be a new PolSAR image feature extraction technique, that is developed by combining the spherically invariant random vector (SIRV) product model with the time-frequency (TF) analysis technology. The SIRV product model will higher characterize HR SAR pictures, and also the TF analysis can assist the classification by taking advantages of the anisotropic property to avoid the confusion of natural and man-created targets. Therefore, using this type of extracted features, man-made targets can be simply discriminated with a simple unsupervised K-means classifier. Experimental results demonstrate the effectiveness of the proposed framework, in that man-created targets are extracted with clear contours, and natural surfaces are very continuous and homogenous. Additionally, lots of interesting targets with special scattering performances are highlighted in many rare classes. Their options are worth learning. Above all, as a result of of barely requiring previous information, the framework should be promising during a wide spectrum of applications by providing the speedy man-made target info acquisition of urban areas.