This paper proposes an automatic unsupervised method for change detection at pixel level of Landsat-5 TM images based on spectral angle mapper (SAM). In most existing studies, conventional use of SAM does not take into account contextual information of a pixel. The proposed method incorporates spatio-contextual information both at feature and decision level for improved change detection accuracy. First, a similarity image is created using contextsensitive spectral angle mapper, and then it is segmented into two segments changed and unchanged using k-means algorithm to create a change map. The quantitative as well as qualitative comparison of the experiment results shows that the proposed method gives better results than the other existing method.
CITATION STYLE
Moughal, T. A., & Yu, F. (2014). An automatic unsupervised method based on context-sensitive spectral angle mapper for change detection of remote sensing images. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 8933, 151–162. https://doi.org/10.1007/978-3-319-14717-8_12
Mendeley helps you to discover research relevant for your work.