Stroke extraction based on ambiguous zone detection: a preprocessing step to recover dynamic information from handwritten Chinese characters

A method of stroke extraction based on ambiguous zone detection is presented to facilitate the recovery of dynamic information from static handwritten Chinese character images. First, ambiguous zones are detected using feature points of the skeleton and the contour information around them. Then, a graph is built to represent each character, and the continuity of sub-strokes is analyzed using Bayesian classification. Several constraint conditions are proposed to search stroke paths in the graph and two criteria are also utilized to deal with multi-traced sub-strokes. Finally, strokes are reconstructed by B-spline interpolation. Experimental results show that the proposed method can detect the ambiguous zones accurately, and is feasible and effective for stroke extraction.