الفهرس | Only 14 pages are availabe for public view |
Abstract Hyperspectral image Classification is one of the most active areas of research and development in the field of hyperspectral image analysis. Recently, many approaches have been extensively studied to improve the classification performance, by integrating the spectral and the spatial information contained in the original hyperspectral image data in a simple and effective way. The goal of this thesis is to develop an algorithm for hyperspectral image classification. Given a set of observations (i.e., pixel vectors in a hyperspectral image), the goal of classification is to assign a distinct class label to every pixel in the image. It can be classified as supervised and unsupervised classification according to the availability of class prototypes. The supervised classification based on spectral-spatial hyperspectral image issue will be studied. It is a difficult and challenging low level vision problem with important applications in vision-guided autonomous robotics, product quality inspection, and medical diagnosis and in the analysis of remotely sensed images. In this thesis, a novel spectral-spatial hyperspectral image classification method is proposed. In the preprocessing stag, the principle component analysis (PCA)/Randomized Singular Value Decomposition (RSVD) is used for dimensionality reduction before classification. Then, the 3-dimensional discrete wavelet transform (3DDWT) is applied to extract the spectral-spatial feature in the Feature Extraction stage |