一种优化的手写字符自动分割算法  

An Optimized Automatic Segmentation Algorithm for Handwritten Characters

作　　者：黄一琦 郑佳春[2] 曹长玉 HUANG Yiqi;ZHENG Jiachun;CAO Changyu(Navigation College,Jimei University,Xiamen 361021,China;School of Information Engineering,Jimei University,Xiamen 361021,China)

机构地区：[1]集美大学航海学院,福建厦门361021 [2]集美大学信息工程学院,福建厦门361021

出　　处：《集美大学学报（自然科学版）》2021年第2期152-159,共8页Journal of Jimei University：Natural Science

基　　金：福建省科技计划重点项目(2017H0028);福建省自然科学基金项目(2015J01265);集美大学校基金项目(2P2020042)。

摘　　要：在手写字符自动识别时,由于手写字符中存在字符大小间距不一、粘连、断点,以及不连贯等情况,给字符自动分割识别带来极大的困难。针对该问题,提出了一种优化的手写字符自动分割算法。该方法依据滴水算法的原理,结合CFS(color filling segmentation)做初步分割;再根据分割字符的连续黑色像素点的宽度判断是否为粘连字符,若为粘连字符,则在分割字符图片0.2倍宽度与0.8倍宽度之间扫描黑色像素位置;结合分割图片中间位置来确定滴水算法起始滴落点,解决特殊情况下的起始滴落点的定位不精准问题。经手写字符识别实验结果表明,优化后手写字符分割准确率比传统方法分割准确率提高了11.6%,且有良好的通用性,可提高手写字符的单个识别率。In the automatic recognition of handwritten characters,due to the differences in character size and spacing,adhesion,breakpoints,and inconsistencies in handwritten characters,it has brought great difficulties to the automatic segmentation of characters during recognition.Aiming at this problem,an optimized automatic segmentation algorithm for handwritten characters is proposed.This method is based on the principle of the drip algorithm,combined with color filling segmentation(CFS)for preliminary segmentation;and then judges whether the characters are sticky characters based on the width of the continuous black pixels of the segmented characters.The black pixel position is scanned between 0.8 times the width and combined with the middle position of the segmented picture to determine the initial drip point of the drip algorithm,which solves the inaccurate positioning of the initial drip point in special cases.The experimental results of handwritten characters show that the segmentation accuracy of optimized handwritten characters is 11.6%higher than that of traditional methods.After optimization,it has good generality and can improve the single recognition rate of handwritten characters.

关 键 词：手写字符分割 滴水算法 粘连字符 起始滴落点 

分 类 号：TP312[自动化与计算机技术—计算机软件与理论]