A SAR Ship Target Detection Method Based on the Analysis of Region Stability Characteristics
Yali Wang, Daojin Chen, Bingqi Zhu, Yi Jian, Yi Liang · 2022 3rd China International SAR Symposium (CISS) · 2022
Traditional SAR ship target detection method could not work well in non-uniform background clutter conditions, and it has low detection accuracy, slow speed, and many false alarms. To solve these problems, this paper proposes a SAR ship target detection method based on the analysis of region stability characteristics. The method uses gravitational field enhancement technology to enhance the contrast between sea background and ship targets, and suppress the interference of speckle noise. To improve detection speed and reduce calculation, a frequency tuning saliency mechanism is introduced. Finally, maximum stable extreme value detection strategy is adopted to reduce false alarms. Experimental results verify the effectiveness of the proposed method.