Aircraft Target Detection In Polsar Image Based On Region Segmentation And Multi-Feature Decision

This paper presents a novel aircraft target detection method based on the combination of region segmentation and multifeature decision. First, the maximum entropy threshold is used to segment region of interest (ROI) like runway, apron and related regions from PolSAR image. Then suspected aircraft targets are extracted in ROI with the feature of dissimilation scattering power. Polarimetric scattering properties of targets and their environmental information are used to construct a detection feature for true aircraft decision from the suspected ones. Experimental results with measured data show that the algorithm can effectively detect aircraft targets with less false and missing alarm rates.