Fast searching Chinese calligraphic information based on character recognition

The easy acquisition of various information of a Chinese calligraphic character can greatly enhance comprehension and learning efficiency of Chinese calligraphy. This paper presents a handy tool for fast searching all kinds of information of a written calligraphic character. Taking the characters written in regular style for example, which are the popular and representative style in calligraphy, first we establish an information library of over 6,700 standard Chinese calligraphy including Pinyin, structure, components and explanation. Then in order to fast identify a given image of a written calligraphic character in the library, we build a neural network with a feature set that largely determines the shape and spatial structure of a character and for the character that is provided more than one identifying results, we turn to image-based matching algorithm to find out the one with the highest similarity. Finally, users can easily obtain the whole information of the input character.