A NOVEL SAR TARGET DETECTION ALGORITHM BASED ON CONTEXTUAL KNOWLEDGE

This paper proposes a Synthetic Aperture Radar (SAR) vehicle target detection algorithm based on contextual knowledge. The proposed algorithm flrstly obtains the general classiflcation of SAR image with a Markov Random Field (MRF)-based segmentation algorithm; then modifles the prior target presence probability utilizing terrain types, distances to boundary and target aggregation degree; flnally gains the detection results using improved Cell Averaging- Constant False Alarm Rate (CA-CFAR). Detections with real SAR image data show that the proposed algorithm can efiectively improve target detection rate and reduce false alarms compared with conventional CA-CFAR.