...
【24h】

Realtime box-free on-line handwriting string recognition using layer-delayed segmentation method

机译:使用分层延迟分割方法的实时无盒在线手写字符串识别

获取原文
获取原文并翻译 | 示例
           

摘要

For a purpose of inputting text naturally, pen-input handwriting recognition is recently needed especially in mobile situations with PDA or pen-PC. In this paper; we will describe a box-free on-line handwritten string recognition method with which users can get recognition results immediately. As the conventional box-free on-line string recognition methods require much more time than the box-based character recognition methods, box-free methods might be not very comfortable for the user to input handwriting text. To solve this problem about recognition time, we have developed a layer-delayed segmentation method which provides real-time box-free string recognition and is also robust to some irregular inputs such as overwritten strokes or stroke deletions. We will also explain some experimental results which describe how this method works well, and talk about our plans in the future.
机译:为了自然地输入文本,最近需要笔输入手写识别,特别是在具有PDA或笔式PC的移动情况下。本文我们将介绍一种无框在线手写字符串识别方法,用户可以通过该方法立即获得识别结果。由于常规的无框在线字符串识别方法比基于框的字符识别方法需要更多的时间,因此无框方法对于用户输入手写文本可能不太舒服。为了解决有关识别时间的问题,我们开发了一种分层延迟的分割方法,该方法可提供实时的无框字符串识别,并且对于某些不规则的输入(例如笔画覆盖或笔画删除)也具有较强的鲁棒性。我们还将解释一些实验结果,这些结果描述了该方法的工作原理,并讨论了我们将来的计划。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

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