Research Of Handwritten Chinese Character Recognition

Chinese Character Recognition is an integrated technology, it involves pattern recognition,image processing, digital signal processing, and other disciplines, it also has a significantapplication value and implications of performance theory for Chinese information processing,office automation, artificial Smart, and other fields. Handwritten Chinese character recognitionhas maintained a rapid and sound development during the 1990s, and it makes the producibilityand maturation of handwritten Chinese character recognition system perfection furthermore.Because of the non-standardability of the handwritten Chinese character written arbitrarily and thelow quality and resolution of handwritten Chinese character specimen, the recognition rate of thesystem is still relatively low, a great gap there between production phase and current state.Feature extraction is the most important section of the Chinese character recognition processto quickly extract stable and reliable and can fully describe the characteristics of Chinese characterpatterns, and it is the key to handwritten Chinese character recognition. Mathematical morphologybased-on the collection on the basis of algebra is the analysis of geometric shape and structure ofthe mathematical method, and the set theory describes the collection of quantitative methods ofscience.This article researches on the foundation Chinese character feature extraction method usingthe mathematical morphology as the fundamental operation. It improves a new Chinese characterfeature extraction method based-on mathematical morphology, using the the mathematicalmorphology as the fundamental operation, making preprocessing and thinning to the originalcharacters image that performs preprocessing to the origina handwritten Chinese character image,and carrying on the stable and available extraction to the feature of Chinese character stroke. Thefeature extraction method based-on this article has an advantage of over the others inoperational speed, and so on.