Application of Adaptive Sparse Representation in Multi - spectral Remote Image Classification ⋆

To improve the accurate rate of classification multi-spectral remote sensing images, in this paper we construct a classification algorithm based on adaptive sparse representation (ASP). The geometrical explanation of ASP algorithm is to approximate the hyper-spherical cap with least hyper-planes, clustered the error vectors of each step, and signed the cluster center as new atoms which made the dictionary more suitable for spare representation of samples. we made testing samples as a linear combination of a few training samples of structured dictionary, this make the dictionary more suitable for spare representation of samples. A sample of remote sensing image of Dadukou areas Chongqing containing forest, land, water, building, green is used to examine the proposed approach. The accurate recognition rate reaching over 97% demonstrates that the proposed approach is capable of dealing with image classification.