A new change-detection method in high-resolution remote sensing images based on a conditional random field model. Issue 5 (3rd March 2016)

• Record Type: Journal Article
• Authors: Cao, Guo; Zhou, Licun; Li, Yupeng
• Is Part Of: International journal of remote sensing. Volume 37:Issue 5(2016)
• Page Start: 1173
• Page End: 1189
• Publication Date: 2016-03-03
• Subjects: Change detection -- conditional random field (CRF) -- fully-connected CRF (FCCRF) -- fuzzy C-means (FCM); Remote sensing -- Periodicals; Télédétection -- Périodiques; 621.3678
• DOI: 10.1080/01431161.2016.1148284
• Languages: English
• ISSNs: 0143-1161

Abstract

ABSTRACT: A new change-detection method for remote sensing images based on a conditional random field (CRF) model is proposed in this paper. The method artfully uses memberships of Fuzzy C-means as unary potentials in the fully connected CRF (FCCRF) model without training parameters, and pairwise potentials of the CRF model are defined by a linear combination of Gaussian kernels, with which a highly efficient approximate inference algorithm can be used. The proposed FCCRF model is expressed on the complete set of pixels in both the observed multitemporal images, which can incorporate long range contextual information of remote-sensing images and enable greatly refined change-detection results. Experimental results demonstrate that the proposed approach leads to more accurate pixel-level change-detection performance and is more robust against noise than traditional algorithms.