Target feature extraction algorithm for SAR images of complex background based on corner estimation

Abstract

Aiming at the problem that the Synthetic Aperture Radar (SAR) image is difficult to target detection for the speckle noise, this paper proposes a SAR image target feature extraction method based on the corner information estimation. Firstly, the image is preprocessed by spatial filtering technique. Then the target detection algorithm based on pulse coupled neural network (PCNN) is used to detect the target of SAR image, and the target in the image is better detected. Finally, through Harris operator extracts the target corner point and performs a least squares estimation to obtain the target corner size. The simulation results show that compared with the traditional angle estimation algorithm, the proposed SAR image feature extraction algorithm based on the rotation angle estimation has higher precision and less error.