作者: Louisa Lam , Ching Y. Suen
DOI: 10.1016/0031-3203(88)90068-4
关键词:
摘要: Abstract A system for recognizing totally unconstrained handwritten numerals is described. It comprises a feature extractor and two classification algorithms. The decomposes the skeleton of character into geometric primitives containing topological information character. These consist convex polygons line segments, features are generated from each primitive. recognition process contains fast structural classifier that identifies majority samples, robust relaxation algorithm which classifies rest data. was trained tested on real-life ZIP codes.