Du, Y. and Yang, N., 2020. Small port detection based on combination of shoreline feature points in polarimetric SAR Images. In: Bai, X. and Zhou, H. (eds.), Advances in Water Resources, Environmental Protection, and Sustainable Development. Journal of Coastal Research, Special Issue No. 115, pp. 102-105. Coconut Creek (Florida), ISSN 0749-0208.

SAR (Synthetic Aperture Radar) is a coherent imaging radar mechanism working in microwave band. SAR image classification is the main component of remote sensing image classification. It has certain penetrability to the surface vegetation and is increasingly becoming one of the most representative earth observation methods today. This paper presents a method of merging shoreline feature points suitable for small port inspection. Firstly, the polarimetric SAR image level set segmentation algorithm is used to realize the accurate segmentation of land and sea, then it is fused with the classical edge detector based on statistical characteristics, and the rough detection of small ports is realized through the characteristics of less edge and good overall connectivity in the sea surface area. On this basis, the classification is completed iteratively by using complex Wishart statistical features. The algorithm can better improve the edge smoothness and regional consistency, and improve the classification quality of polarimetric SAR images. The experimental results show that the method proposed in this paper can realize high-precision detection of bridges.