Shape retrieval and recognition

Abstract

Shape recognition has been an attractive research area with both theoretical and practical importance in the fields of computer vision, pattern analysis and object classification. However, to the best of our knowledge, there is as yet not one unique technique of general validity to measure shape similarity and robust to noise. In this research, we aim to propose and investigate one shape recognition system which can tackle different types of input data, such as logos, fingerprints, work tools and characters.