Chinese Calligraphy Image Analysis Project

It has been long for the literates in China and the region in the cultural sphere of East Asia to use the hairy bristle brush as a tool of writing and drawing. Numerous calligraphic masterpieces have been produced by using this simple tool which is made of a bamboo stem and a bundle of starched bristle hairlines only. However, due to the complex interaction between the hairlines in a bundle, the special expressive ability of this simple tool is so strong that makes it capable to produce a number of aesthetic effects such as flying-white, dry brush and ink diffusion which can hardly be seen, if not lacked in western artwork. To be preserved in digital format, Chinese calligraphic masterpieces can be scanned or photographed. The resulting static image representation is often annotated with a text description.

Our projects focus on various issues in analyzing Chinese calligraphic images. We first apply model-based analysis for Chinese calligraphic images to derive a compact electronic representation by estimating the parameters given a brush model. Then we will derive the dynamic writing information and determine the writing style from the analysis of an input Chinese calligraphic image. To obtain the dynamic writing information automatically, it is necessary to extract the underlying strokes from a Chinese character image and then identify the writing direction of each stroke and the order of the strokes. The set of geometric parameters can be used for both representing the extracted strokes and deriving features for the style analysis. In addition, we will exploit the Chinese writing characteristics under different styles to make our approach robust to handle complex strokes written in a particular style. With our proposed work, the writing process of an ancient Chinese calligraphic masterpiece can be animated automatically. It is also possible to determine whether a calligraphic masterpiece falls into a certain brush or calligrapher style so that it can help in labeling a non-annotated calligraphy.

• Obtain a vector-based representation of Chinese calligraphy

• Enable dynamic reproduction of calligraphic writings through analysis of static Chinese calligraphic images

• Extract the underlying strokes given a Chinese calligraphy character written in a particular brush style

• Determine automatically the order of the extracted strokes and the direction of each stroke

• Derive robust features from brush-written Chinese character images for identifying Chinese brush styles and writer styles
  　

Stroke Extraction for Chinese Calligraphy Characters		Regeneration of Strokes from Chinese Calligraphic Images

　

• Sam T.S. Wong, Howard Leung and Horace H.S. Ip, "Automatic Estimation of Ink Depositing Model Parameters from Analysis of Chinese Calligraphic Images", Pacific Graphics 2006 (PG 2006), Taipei, Taiwan, October 2006.
• Sam T.S. Wong, Howard Leung and Horace H.S. Ip, "Fitting Ellipses to a Region with Application in Calligraphic Stroke Reconstruction", International Conference on Image Processing (ICIP 2006), Atlanta, USA, October 2006.
• Sam T.S. Wong, Howard Leung and Horace H.S. Ip, "Brush Writing Style Classification from Individual Chinese Characters", International Conference on Pattern Recognition (ICPR 2006), Hong Kong, August 2006.
• Howard Leung and Hui Kang, "Stroke Extraction for Chinese Calligraphy Images", Asia Pacific Workshop on Visual Information Processing (VIP2005), Hong Kong, December 2005.
• Sam T.S. Wong, Howard Leung and Horace H.S. Ip, "Model-based Analysis of Chinese Calligraphy Images", 9th Intl. Conf. on Information Visualisation (IV05), London, England, July 2005.

All the projects are supervised by Dr. Howard Leung. He is an Assistant Professor at City University of Hong Kong. Any suggestions or comments are welcome. Please feel free to send them to Dr. Howard Leung.

We thank the Chinese Civilisation Centre for allowing us to use the wall-paper in this webpage.