A genetic approach to the normalization of distorted character images

Character normalization recovers distortions occurring in character images. There are two kinds of character distortions: local and global. Local distortion has been discussed in the literature, but the global distortion that is usually produced by geometric transformation is not usually considered. In this paper, a novel method for solving the global character distortion problem is proposed. The global character distortion problem is regarded as a constrained geometric transformation problem, and an adaptive optimization approach using genetic algorithms is proposed to solve the problem. Experiments on Chinese characters with six kinds of distortions show satisfactory results