A new simplex growing endmember extraction algorithm and its fast implementation

摘要：

　　The simplex growing algorithm (SGA) suffers from two issues in its practical implementation.One is that it needs to reduce the original data’s dimensionality in advance.The other one is its excessive computation.In order to solve the first issue,a new simplex growing algorithm (N-SGA) which uses a new simplex volume formula with no need for dimension reduction is proposed in this paper.Furthermore,a developed approach called Fast N-SGA is put forward to reduce the computation complexity.The conducted experiments demonstrate that the proposed algorithms N-SGA and Fast N-SGA perform better than the tradition SGA.

关键词： photogrammetry and remote Sensing endmember extraction N-SGA fast N-SGA dimension reduction computation complexity

作者: Lijiao Wang Xiaorun Li Jiantao Cui Liaoying Zhao

作者单位: Zhejiang University,College of Electrical Engineering,Hangzhou,310027,China Institute of Computer Application Technology,Hangzhou Dianzi University,Hangzhou,310027,China

会议类型: 国内会议

会议名称: 第二届高分辨率对地观测学术年会

会议地点: 北京

会议语种:英文

页码: 1-14

在线出版日期: 2013-12-01（万方平台首次上网日期，不代表论文的发表时间）