Vessel detection via multi-order saliency-based fuzzy fusion of spaceborne and airborne SAR images

This study focuses on vessel target detection by fusing synthetic aperture radar (SAR) remote sensing images collaboratively collected from spaceborne and airborne platforms. Accurate vessel detection is difficult in the presence of inshore interferences and in the case of structured and shaped targets. In this study, we have proposed a new method for the fusion of spaceborne and airborne SAR images based on multi-order superpixel-level saliency and fuzzy logic (MSSFL). We first generated a new global regional contrast map (GRCM) by exploiting the multi-order superpixel-level saliency (MSS). In GRCMs, the vessel targets are well restored, and the background is suppressed. Next, a new fuzzy logic approach based on regional features is presented to fuse the MSS information provided by the GRCMs of the spaceborne and airborne SAR images. This regional feature-based fuzzy fusion can further enhance the vessel target regions and filter out the inshore interference regions. Experimental results using the SAR images of Gaofen-3 satellite and unmanned aerial vehicle show that the proposed MSSFL method yields a higher target-to-clutter ratio of fused images and improved detection performance, compared with the commonly utilized image fusion approaches and the classical detection algorithms solely using spaceborne or airborne SAR images.