Multi-classifier system for robust pattern recognition

• Main Author: Ng, Geok See.
• Other Authors: Singh, Harcharan
• Format: Theses and Dissertations
• Language: English
• Published: 2008
• Subjects: DRNTU::Engineering::Computer science and engineering::Computing methodologies::Pattern recognition; DRNTU::Engineering::Computer science and engineering::Theory of computation::Analysis of algorithms and problem complexity
• Online Access: http://hdl.handle.net/10356/13584
• Institution: Nanyang Technological University

Despite the success of many pattern recognition problems in a constrained domain, the task of pattern recognition is "ill-defined" and difficult due to the noise and large variations in input data. A promising approach is to use several classifiers simultaneously, such that they can complement each other in correctness. This thesis tackles the recognition problem in two aspects: (1) propose a new classifier called Contender Network (CN) and (2) propose a combining classifier called Combined Classifier (CC) which aggregates the outputs of a number of pattern classifiers using a new evidence combination method. So the primary objective of this work is to propose an effective framework of multiple classifier system that takes advantage of the strength of the individual classifier. This framework is then applied to the task of recognition of hand-written numeric digits.