首页> 外国专利> CHARACTER RECOGNITION METHOD AND DEVICE BASED ON SUPPORT VECTOR MACHINE, AND COMPUTER DEVICE

CHARACTER RECOGNITION METHOD AND DEVICE BASED ON SUPPORT VECTOR MACHINE, AND COMPUTER DEVICE

机译:基于支持向量机的字符识别方法,装置及计算机装置

摘要

Disclosed are a character recognition method and device based on support vector machine, a computer device and a storage medium. The method comprises: obtaining a specified picture to be classified; calculating a normalized vector Gi of the specified picture so as to obtain n-dimensional vectors (G1, G2,... Gn); inputting the n-dimensional vectors (G1, G2,... Gn) into multiple preset and trained support vector machines for calculation; recording the category of the specified picture to be classified as a preliminary classification result corresponding to the return value of max (an output value corresponding to the multiple preliminary classification results); obtaining a character recognition mode corresponding to the specified picture; recognizing a character recognition area as character text and storing the character text. The method reduces the waste of computing power, improves the classification efficiency and the recognition efficiency, and is adaptive to specified pictures which are stretched or tilted.
机译:公开了一种基于支持向量机的字符识别方法和装置,计算机装置和存储介质。该方法包括:获得要分类的指定图片;以及计算指定图片的归一化向量Gi,以获得n维向量(G1,G2,...,Gn);将n维向量(G1,G2,... Gn)输入到多个预设且训练有素的支持向量机中进行计算;将要分类的指定图片的类别记录为与max的返回值相对应的初步分类结果(与多个初步分类结果相对应的输出值);获取与指定图片相对应的字符识别模式;将字符识别区域识别为字符文本并存储该字符文本。该方法减少了计算能力的浪费,提高了分类效率和识别效率,并且适用于被拉伸或倾斜的指定图片。

著录项

  • 公开/公告号WO2020155484A1

    专利类型

  • 公开/公告日2020-08-06

    原文格式PDF

  • 申请/专利权人 PING AN TECHNOLOGY (SHENZHEN) CO. LTD.;

    申请/专利号WO2019CN89057

  • 发明设计人 ZHENG JIA;

    申请日2019-05-29

  • 分类号G06K9/62;

  • 国家 WO

  • 入库时间 2022-08-21 11:10:01

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