Hybrid Primitive Extraction Method for Efficient Character Recognition

  • Lee, Kyoon Ha

초록

In this paper, we propose HPEM(Hybrid Primitive Extraction Method) for the Hangul character recognition which enhances the primitives made by SPEM(Skeleton-BASED Primitive Extraction Method) with adding CPEM(Contour-based Primitive Extraction Method). The approach is to use CPEM as a complementary method to SPEM for the purpose of reducing the loss of information. Using this method, we improve the recognition rate.

제목
Hybrid Primitive Extraction Method for Efficient Character Recognition
저자
Lee, Kyoon Ha
학회명
International Conference on Computer Processing of Oriental Languages(ICCPOL'94)