The skeletonization research of low-quality Chinese characters based on principal curves

The skeleton which keeps the topology information as well as reduces the computational complexity is an excellent and robust structural feature to noise and deformation. This paper focuses on the skeletonization of low-quality free handwritten or printed Chinese characters with fuzziness, deformation, sparseness and discontinuity. We use a simple scheme to optimize the initial Hamilton path obtained from the soft K-segments algorithm and then propose a scheme based on the overall average sparseness to set the threshold, thus to delete the redundant edges. Experimental results show that skeletons obtained from this proposed scheme closely conform to human perception without any artifact.