Study on coastline changes of Yellow River estuary using semi-supervised clustering

This paper proposes an image segmentation strategy based on semi-supervised clustering. Firstly, we extract band information from images to generate pixel datasets; secondly, incorporate the instance-level knowledge in the form of pairwise constraints to assist clustering, and partition two clusters; finally, generate binary images based on clustering results so as to segment land and ocean. Then, we utilize our method to extract coastlines of Yellow River estuary in 1978, 1989, 1999 and 2009, so as to analyse the coastline evolution of Yellow River estuary.