Stroke Reasoning for Robotic Chinese Calligraphy Based on Complete Feature Sets

Zhe Ma, Jian-Bo Su

Abstract: There are a lot of functional and artistic elements in Chinese calligraphy, involving brush trajectories, velocity and force in stroke writing and so on. These features make Chinese calligraphy a charming art and meanwhile bring difficulties in writing. This paper investigates the approach in stroke reasoning for Chinese calligraphy with robots. The complete feature set, which is defined as the one containing all the information required to fulfill corresponding task, is first introduced to characterize Chinese calligraphy. A stroke reasoning strategy is proposed from the complete feature set point of view according to some ad hoc strategies in Chinese writing. Stroke characteristics are described by several indexes, which are selected by C4.5 decision tree to reason out stroke directions. A hierarchical reasoning pyramid is proposed to obtain stroke order in Chinese characters. The proposed stroke reasoning methodology is tested on two databases consisting of strokes from Ouyang Xun’s tablet inscriptions.