Spatial-frequency approaches to texture analysis

• Main Author: Mo, Xiaoran.
• Other Authors: Chan, Kap Luk
• Format: Theses and Dissertations
• Language: English
• Published: 2008
• Subjects: DRNTU::Engineering::Electrical and electronic engineering::Computer hardware, software and systems; DRNTU::Engineering::Electrical and electronic engineering::Electronic systems::Signal processing
• Online Access: http://hdl.handle.net/10356/13253
• Institution: Nanyang Technological University

Spatial-frequency methods have been extensively and successfully employed by many computer vision researchers to texture analysis in the last two decades. The focus of this thesis is on the research work carried out based on such approaches. First, application of Gabor filters to texture analysis is investigated. A filter selection algorithm for texture recognition has been developed to select a small subset of Gabor filters from a pre-defined Gabor filter bank. The filter selection is based on the discriminative power of each individual Gabor filter in regard to the recognition of all the textures in an image database. The proposed filter se-lection algorithm is demonstrated to be capable of selecting more discriminative filter through texture classification and retrieval experiments.