Applications of B-splines in fingerprint classification and stereo matching

• Main Author: Chong, Men Soon.
• Other Authors: Gay, Robert Kheng Leng
• Format: Theses and Dissertations
• Published: 2010
• Subjects: DRNTU::Engineering::Electrical and electronic engineering::Electronic systems::Signal processing
• Online Access: http://hdl.handle.net/10356/38964
• Institution: Nanyang Technological University

In a world of order and organization, there is always an abundance of structured objects, either man-made or natural. One of the most interesting facts about the human visual system is its ability to derive such structures as collinearity, parallelism and symmetry without prior knowledge of the objects or scène on which these structures are found. This phenomena is called perceptual organization. Besides having a high likelihood of corresponding to object or scène structures, such structures are also invariant to viewpoint transformations. These significant characteristics suggest the utility of perceptual organization to image interpretation in general. To effectively apply the principles of perceptual organization for computer vision and pattern recognition, we require a unified approach that can facilitate the extraction of such structures in images, as well as provide the methodology for utilizing the structures for various image interpretation tasks.