东巴象形文字基础分类算法研究  

Study on the Basic Classification Algorithm for Dongba Hieroglyphs

作　　者：杨玉婷 康厚良 YANG Yu-ting;KANG Hou-liang(School of Computer Engineering,Suzhou Vocational University;Sports Department,Suzhou Vocational University,Suzhou 215000,China)

机构地区：[1]苏州市职业大学计算机工程学院 [2]苏州市职业大学体育部,江苏苏州215000

出　　处：《软件导刊》2019年第11期149-151,156,共4页Software Guide

基　　金：云南省科学研究基金项目(2018JS748)

摘　　要：从构字结构要素分析东巴象形文字,可分为单素字和复素字两类,而单素字又可进一步分为轮廓型和结构型两种。使用分类算法实现不同类型东巴文字的自动分离,有利于独立研究轮廓型、结构型单素字及复素字,提取同类文字的本质特征及不同类型文字间的特征差异。从东巴文字的构字结构要素出发,给出了适用于东巴象形文字自动分类的预处理和分类算法。该算法能实现单素字和复素字的分离,还能实现对单素字中轮廓型和结构型单素字的完全分离。通过实验对算法的准确性进行了验证。By analyzing the structural features of characters,Dongba hieroglyphs can be divided into two types:single graphemes and compound graphemes.The single graphemes can be further divided into contour type and structure type.The automatic separation of different types in Dongba hieroglyphs is conducted with the classification algorithm,which is conducive to the independent study of single and compound graphemes and the extraction of the commonalities among hieroglyphs of the same type and the differences in different types.Therefore,by studying the structure features of Dongba hieroglyphs,we give a preprocessing and classification algorithm.The algorithm can achieve complete separation of single graphemes and compound graphemes,and even includes the separation of contour type and structure type in single graphemes.Finally,we verified the accuracy of the algorithm through experiments.

关 键 词：东巴象形文字 自动分类算法 文字构字结构 文字形态 文字结构 

分 类 号：TP317.4[自动化与计算机技术—计算机软件与理论]