Open AccessHYPERSPECTRAL IMAGE CLASSIFICATION BASED ON MANIFOLD DATA ANALYSIS AND SPARSE SUBSPACE PROJECTION
Author(s) -
Zedong Zheng,
Yanbin Peng
Publication year - 2021
Publication title -
international journal of engineering technologies and management research
Language(s) - English
Resource type - Journals
ISSN - 2454-1907
DOI - 10.29121/ijetmr.v8.i9.2021.1040
Subject(s) - subspace topology , dimensionality reduction , hyperspectral imaging , projection (relational algebra) , pattern recognition (psychology) , manifold (fluid mechanics) , artificial intelligence , manifold alignment , dimension (graph theory) , nonlinear dimensionality reduction , neural coding , sparse approximation , mathematics , computer science , sparse matrix , algorithm , combinatorics , physics , mechanical engineering , quantum mechanics , engineering , gaussian
Aiming at the problem of "dimension disaster" in hyperspectral image classification, a method of dimension reduction based on manifold data analysis and sparse subspace projection (MDASSP) is proposed. The sparse coefficient matrix is established by the new method, and the sparse subspace projection is carried out by the optimization method. To keep the geometric structure of the manifold, the objective function is regularized by the manifold learning method. The new method combines sparse coding and manifold learning to generate features with better classification ability. The experimental results show that the new method is better than other methods in the case of small samples.