FPTA快速细化算法在脱机手写体汉字识别中的应用  

The Application of the FPTA Fast Thinning Algorithm in Offline Chi nese Handwritten Characters Recognition

作　　者：邓刚[1] 童学锋[1] 

机构地区：[1]同济大学计算机科学与工程系,上海200092

出　　处：《计算机工程与应用》2002年第1期135-136,共2页Computer Engineering and Applications

摘　　要：阐述了将FPTA快速细化算法应用于脱机手写体汉字识别中,该算法提出了一种11象素支撑窗口,能有效地抑制样本中的“毛刺“,快速提取保持原始汉字样本连通性和拓扑性的骨架,它的处理速度和细化质量明显优于现有的一些细化算法,根据实验结果,该算法是一种有效的汉字样本细化算法。This article describes the application of the FPTA fast thinning algorithm in offline Chinese handwritten characters recognition.This algorithm puts forword a window that is supported by eleven pixels,it can restrain the'burr'of the swatch and take out the shape of Chinese handwritten characters.What is more,its processing speed is fast and its thinning quality is good.Based on the results of the experiment ,this algorithm is an effective thinning algorithm in Chinese characters thinning.

关 键 词：细化 识别 脱机 手写体 

分 类 号：TP391.43[自动化与计算机技术—计算机应用技术]