The Research And Implementation Of Chinese Calligraphy Tablet Generation Techniques

The art of Chinese calligraphy has a history of thousands of years. Because of its aesthetic values, calligraphy has been widely used in many occasions. Its application in tablets is one of the important usages. In this paper, we mainly focus on the research of the generation of calligraphy tablet using computer technology. With the content of the tablet provided by the user, the system would choose suitable calligraphy word images with the same style and create the calligraphy tablet. The research, which relates to computer technology and calligraphy art, is rarely studied in computer science domain. The key point of the research lies in how to figure out whether calligraphy words resemble in style, thus how to retrieve calligraphy word features and how to use these features to measure calligraphy word similarities are the focuses of the paper.Usually calligraphy style is difficult to tell. In this paper, we propose a method to represent calligraphy style using calligraphy image features, which solves the difficulty of mapping the underling image features to the upper style of the calligraphy image.We present an overview of the background, motivation, and basic approaches of our research in the beginning of this paper.In chapter 2, we present a review of the research related to our research, including the technology used in signature verification and technology concerning information retrieval. We also briefly introduce and discuss the main work flow of the research.In chapter 3, we propose two types of features to represent calligraphy style; one is related to the word shape and the other related to the strokes of the word. Based on the features, we present the methods of how to measure the similarities between different calligraphy words.As users' feedbacks help to make the system more accordant to the users' need, we introduce the relevance feedback technology used in image retrieval area, including long term feedback and similarity propagation. Through continually learning of users' feedback, we establish the similarity relation between calligraphy works and calligraphist. With these feedback technologies, we could retrieval more suitable calligraphy images to form the tablets.In chapter 5, we present the system of Chinese calligraphy tablet creation and some other related work. We conclude the whole paper in Chapter 6, with a brief discussion of the application prospect and future research directions.