28 February 2014 Automatic generation of Chinese character using features fusion from calligraphy and font
Author Affiliations +
Abstract
A spatial statistic based contour feature representation is proposed to achieve extraction of local contour feature from Chinese calligraphy character, and a features fusion strategy is designed to automatically generate new hybrid character, making well use of contour feature of calligraphy and structural feature of font. The features fusion strategy employs dilation and erosion operations iteratively to inject the extracted contour feature from Chinese calligraphy into font, which are similar to “pad” and “cut” in a sculpture progress. Experimental results demonstrate that the generated new hybrid character hold both contour feature of calligraphy and structural feature of font. Especially, two kinds of Chinese calligraphy skills called “Fei Bai” and “Zhang Mo” are imitated in the hybrid character. “Fei Bai” depicts a phenomenon that part of a stroke fade out due to the fast movement of hair brush or the lack of ink, and “Zhang Mo” describes a condition that hair brush holds so much ink that strokes overlap.
© (2014) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Cao Shi, Jianguo Xiao, Canhui Xu, Wenhua Jia, "Automatic generation of Chinese character using features fusion from calligraphy and font", Proc. SPIE 9012, The Engineering Reality of Virtual Reality 2014, 90120N (28 February 2014); doi: 10.1117/12.2038945; https://doi.org/10.1117/12.2038945
PROCEEDINGS
7 PAGES


SHARE
Back to Top