Novel Feature Extractions Method of Full Polarimetric SAR Imagery for Automatic Target Recognition

摘要

Synthetic aperture radar (SAR) is efficient for high resolution imaging analysis even in adverse weather or darkness situations. There are various approaches for automatic target recognition (ATR) of SAR images aiming at monitoring unidentified ships or aircrafts. In addition, various types of analyses using full polarimetric data have been developed recently because it can provide valuable information to identify structure of targets, such as vegetation, urban, sea surface areas. This paper proposes a novel feature extraction method for SAR ATR issues, in exploiting full polarimetric SAR data, where a signal to noise ratio (SNR) level for each polarimetric SAR image is appropriately considered in target area extraction. The experimental data, assuming 1/100 scale model of X-band radar, demonstrate that our proposed method significantly improves an accuracy of feature extraction and classification by SOM-based ATR method, even in severely noisy situations.