首页> 外国专利> METHOD FOR CODING LOSS OF BINARY IMAGE BY USING PATTERN INTERRELATIONSHIP EFFECTIVELY

METHOD FOR CODING LOSS OF BINARY IMAGE BY USING PATTERN INTERRELATIONSHIP EFFECTIVELY

机译:有效地利用模式相互关系编码二进制图像损失的方法

摘要

PURPOSE: A method for coding a loss of a binary image is provided to increase a matching rate by separating a consonant and a vowel in case of Korean alphabet and performing a partial search to characters which are combined with two characters in case of the Roman alphabet. CONSTITUTION: A symbol is extracted in a binary image(201). It is investigates whether the extracted symbol is a ligature(a combined character)(202). In case of a ligature of the Korean alphabet, a ligature separation of an objected symbol is processed(203). In case of a ligature of the Roman alphabet, a partial search of an objected symbol is processed(205). A dictionary is searched using the separated symbol(207). If a matching symbol is existed in the dictionary(208), a refined bit map and position information of components comprising a ligature in the Roman alphabet character are stored in the dictionary as coding data. Otherwise, the symbol is rotated and searched in the dictionary. In case that a matching symbol is existed in the dictionary, index information of the matching symbol is coded, and the central points of the current symbol and the matching symbol are accorded(213). Otherwise, a direction having the highest compressing rate is decided and a four-direction scanning is performed and a direct bit map of the symbol is coded, and new symbol is registered in the dictionary at preparing a direct bit map of the symbol(214,215). In case that a matching symbol is existed in the dictionary, a direction having the highest compressing rate is decided and a four-direction scanning is performed and a refined bit map of the matching symbol is coded, and matching symbol is registered in the dictionary at preparing a refined bit map of the matching symbol(216).
机译:目的:提供了一种编码二进制图像丢失的方法,以通过在朝鲜语字母的情况下分离辅音和元音并在罗马字母情况下对与两个字符组合的字符执行部分搜索来提高匹配率。构成:在二进制图像中提取一个符号(201)。研究所提取的符号是否为连字(组合字符)(202)。在韩文字母的连字的情况下,处理对象符号的连字分隔(203)。在罗马字母的连字的情况下,处理对象符号的部分搜索(205)。使用分离的符号搜索字典(207)。如果在字典中存在匹配的符号(208),则将包含罗马字母字符中的连字的成分的精制位图和位置信息作为编码数据存储在字典中。否则,将旋转符号并在词典中进行搜索。在字典中存在匹配符号的情况下,对匹配符号的索引信息进行编码,并且使当前符号和匹配符号的中心点一致(213)。否则,确定压缩率最高的方向,并执行四向扫描,并对符号的直接​​位图进行编码,并在准备符号的直接​​位图时将新符号注册到字典中(214,215) 。在字典中存在匹配符号的情况下,确定具有最高压缩率的方向,并执行四向扫描,并对匹配符号的精炼位图进行编码,并且将匹配符号注册到字典中。准备匹配符号的精炼位图(216)。

著录项

  • 公开/公告号KR100312810B1

    专利类型

  • 公开/公告日2001-10-12

    原文格式PDF

  • 申请/专利权人 KO HYUNG HWA;

    申请/专利号KR19970068093

  • 发明设计人 KIM HYEON MIN;KO HYUNG HWA;KIM YEONG TAE;

    申请日1997-12-12

  • 分类号G06T1/00;

  • 国家 KR

  • 入库时间 2022-08-22 01:12:01

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号