Naxi Dongba hieroglyphs of China are the only living hieroglyphs world widely which still in use. There are thousands of manuscripts written in Dongba hieroglyphs scattering in different counties for history reason. For culture protection and inheritance, those manuscripts are in urgent need to be recognized and organized quickly. This paper focuses on the recognition of Naxi Dongba hieroglyphs by using coarse grid method to extract features and using support vector machine to classify. The designed Experiment shows that the method performs better than the commonly used clustering method in recognition accuracy in recognition of Naxi Dongba hieroglyphs. This method also provides some experience for recognition of other hieroglyphs.