Image Classification Using Sparse Representation

Please enable javascript in your browser.

Haghiri, Siyavash | 2013

968 Viewed

Type of Document: M.Sc. Thesis
Language: Farsi
Document No: 44200 (19)
University: Sharif University of Technology
Department: Computer Engineering
Advisor(s): Rabiee, Hamid Reza
Abstract:
In this thesis, we have discussed image classification by sparse representation. Sparse representation is used in two different ways for image classification. The first goal of sparse representation is to make an efficient classifier, that can learn the subspace, in which the data lies. In this field we have surveyed various methods. We also proposed a method, called ”Locality Preserving Dictionary Learning” that works approximately better than state of the art similar methods, specially when training data is limited. We have reported the result of lassification on four datasets including MNIST, USPS, COIL2 and ISOLET. Another use of sparse representation, is to extract local features from natural images. The use of sparse coding and other codings are discussed in related work. Semi-supervised classification of natural images is a new and challenging problem that is discussed. Also, we propose a method
to extract features in order to semi-supervised classification. Methods in this section are compared with Caltech101 and COREL10 datasets
Keywords:
Classification ; Sparse Representation ; Machine Learning ; Dictionary Learning