基于过程神经元的手写体汉字仿人识别研究

Humanoid Recognition of Handwritten Chinese Character Based on Procedure Neural Networks

为有效地获取脱机手写体汉字笔划信息,采用过程神经元网络提取手写体汉字基本笔段,分析各类笔段间的拓扑性质,并将手写体汉字图像转化为具有容错表征方式的六种汉字笔划类型在不同位置组成的几何图形。模仿人类汉字形码输入法,统计具有冗余容错形状的笔划类型和相合相交点的数量和位置,建立手写体汉字多维特征知识数据结构表,通过对比和判断仿人容错地识别手写体汉字。对SCUT-IRAC手写体汉字库中汉字进行了实验仿真,该方法具有较强的"认知"手写体汉字的能力。

For acquiring the information of off-line handwritten Chinese character strokes,procedure neural networks are applied to extract basic stroke segments of handwritten Chinese character,and the topological property among stroke segments is analyzed.Image of handwritten Chinese character is transferred to geometric graphics,and the graphics is composed of six styles of Chinese character strokes with fault tolerance in different positions.Chinese character font coding used by human is imitated,and the number and position of strokes,which have redundant fault tolerant shapes,and joint and crossover of strokes are accumulated.A kind of characteristic knowledge data-base table of handwritten Chinese characters is constructed,and the characters are recognized by comparing and judging.Handwritten Chinese character in SCUT-IRAC HCCLIB is tested,and this method is proved to have ability of ＂cognizing＂.