Coastline extraction based on multi-scale segmentation and multi-level inheritance classification

Detailed management of the coastline is critical to development coastal states. However, current classification relatively weak. This study proposed an automatic method detect coastlines with category attributes based on multi-scale segmentation and multi-level inheritance classification. Fully integrating advantages classification, it solved problems that traditional methods could not solve, such as extracting categorical attributes, cultivation ponds are easily affected by tidal flats, complex terrain. The Chinese GF-2 satellite images used extract various types in Jiaozhou Bay its surrounding areas harbor-wharf coastline, silt pond rocky sandy coastline. Compared human interpretation, found extracted our different 10.104 km 0.099 2.677 8.831 0.218 Furthermore, compared object-based region growing edge detection (OBRGIE) method, increased 13.52%, 2.16%, 14.48%, 52.57%, 22.97%, respectively. results show algorithmically more reasonable, accurate, powerful. It can provide data support for refined management.