Title of Paper:A novel coastline detection method of remote sensing imagery with local gradient based on hybrid active contour model

Hits:

Date of Publication:2015-01-01

Journal:ICIC Express Letters

Included Journals:EI、Scopus

Volume:9

Issue:6

Page Number:1651-1656

ISSN No.:1881803X

Abstract:Weak boundaries often occur in remote sensing images and may cause incorrect coastline detection because of boundary leakage. In this paper, a novel coastline detection method is proposed by hybrid active contour model (ACM). We propose an active contour model using local gradient and local gradient mean reduction to solve the problem of low contrast in image pixels and larger local gradient mean on remote sensing images. Two restriction functions are incorporated into Region-Scalable Fitting (RSF) ACM. The minimization of the energy associated with this modified active contour model is implemented within a level set framework. The local gradient means are lower than those of RSF. Then these two local gradient means are utilized by a spoon force into Geodesic Active Contour (GAC) model. The results on remote sensing images compared with other methods are presented. It is shown that the proposed method can achieve better coastline detection in remote sensing images. ? 2015 ICIC International.