Offline Handwritten Chinese Character Recognition viaRadical Extraction and Recognition
ICDAR '97 Proceedings of the 4th International Conference on Document Analysis and Recognition
Genetic approaches for topological active nets optimization
Pattern Recognition
Hi-index | 0.00 |
In this paper, a popular snake model is enhanced by considering the guiding image force and speeded up by incorporating Genetic Algorithm. It has been applied to segment the radicals in offline handwritten Chinese characters. Testing results show that the proposed approach can effectively decompose the radicals with overlaps and connections from the characters with various layout structures. The segmentation accuracy reaches 94.91% and the average running time is around 0.05 second per character.