Semi-supervised segmentation of multispectral remote sensing image based on spectral clustering

Xiangrong Zhang, Ting Wang, Licheng Jiao, Chun YangProceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE 2009 · DOI

Abstract

In this paper, a new multi-spectral remote sensing image segmentation method based on multi-parameter semi-supervised spectral clustering (STS<sup>3</sup>C) is proposed. Two types of instance-level constraints: must-link and cannot-link are incorporated into spectral cluster to construct semi-supervised spectral clustering in which the self-tuning parameter is applied to avoid the selection of the scaling parameter. Further, when STS<sup>3</sup>C is applied to multi-spectral remote sensing image segmentation, the uniform sampling technique combined with nearest neighbor rule is used to reduce the computation complexity. Segmentation results show that STS<sup>3</sup>C outperforms the semi-supervised spectral clustering with fixed parameter and the well-known clustering methods including k-means and FCM in multi-spectral remote sensing image segmentation.
BibTeX
@inproceedings{zhang2009semi,
  author = {Xiangrong Zhang and Ting Wang and Licheng Jiao and Chun Yang},
  title = {{Semi-supervised segmentation of multispectral remote sensing image based on spectral clustering}},
  booktitle = {Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE},
  year = {2009},
  doi = {10.1117/12.832880},
}

Loading related content...

Figure