A FAST 2-STAGE CLASSIFICATION METHOD FOR HIGH-DIMENSIONAL REMOTE-SENSING DATA

Classification for high-dimensional remotely sensed data generally req uires a large set of data samples and enormous processing time, partic ularly for hyperspectral image data, Hat this paper, we present a fast two-stage classification method composed of a band selection (BS) alg orithm with feature extraction/selection (FSE) followed by a recursive maximum likelihood classifier (MLC). The first stage is to develop a BS algorithm coupled with FSE for data dimensionality reduction. The s econd siege is to design a fast recursive MLC (RMLC) so as to achieve computational efficiency, The experimental results shelf that the prop osed recursive MLC, in conjunction with BS and FSE, reduces computing time significantly by a factor ranging from 30 to 145, as compared to the conventional MLC.