Segmentation of handwritten Chinese characters based on connected components amalgamation and penetration times algorithm
WANG Lin-wan, YANG Yang, XIE Bin, YANG Yi (Information Engineering Institute, Beijing University of Science and Technology, Beijing 100083, China)
This paper consults basic structure features of Chinese characters and uses connected components as basic processing units. Connected components are merged according to their topological relations. An improved penetration times algorithm is used to choose optimized cut points in touching characters. The experimental results show that the method has satisfactory performance for segmenting single-line free-format handwritten Chinese characters.
CAJViewer7.0 supports all the CNKI file formats; AdobeReader only supports the PDF format.