22 May 2014 Applying region growing algorithm to hyperspectral image for oil segmentation
Author Affiliations +
Region growing is one of the popular segmentation algorithms for 2-D image, which comes up a continuous interested region. How to extent this method to hyperspectral image processing effectively is a problem needs to be discussed deeply. Here in this paper, three ways of using region growing in hyperspectral scenario are explored to separate oil from sea water. Furthermore, in order to release the influence of sunlight, a modification to growing rule is prompted, considering the property of local region. At last, a normalized ATGP is used to obtain more potential target. The experiment results show that combining unmixing techniques with region growing is better than other methods.
© (2014) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Mei-ping Song, Mei-ping Song, Xing-wei Xu, Xing-wei Xu, Shuangyang Lu, Shuangyang Lu, Wei Xu, Wei Xu, Hai-mo Bao, Hai-mo Bao, "Applying region growing algorithm to hyperspectral image for oil segmentation", Proc. SPIE 9124, Satellite Data Compression, Communications, and Processing X, 912404 (22 May 2014); doi: 10.1117/12.2053333; https://doi.org/10.1117/12.2053333

Back to Top