For text locating it is unrealistic to expect a system to agree exactly with the bounding rectangle for a word identified by a human tagger.
对文本定位来说,期望识别系统能够像人一样准确的分辨文本区域的边界是不现实的,对系统的衡量也需要综合各方面的因素。
参考来源 - 基于边缘和纹理的文本定位算法的研究·2,447,543篇论文数据，部分数据来源于NoteExpress
In their simplest form, regular expressions are the search criteria used for locating text in a file.
在最简单的形式中，正则表达式(regular expression)是用来在文件中定位文本的一些搜索标准。
Without solutions integrated into enterprise systems, text analytics often becomes a last-resort process of locating and reading pertinent text records when problems arise.
如果没有进入企业系统，当出现问题时，文本分析的综合解决方案往往成为一个定位和阅读相关文字记录的最后的过程。
Compared with other methods nowadays, it's robust in locating single characters, skewed and even vertical text lines.
与现有的其他方法相比，在定位单个字符、倾斜甚至竖直的文字行方面有更好的鲁棒性。
应用推荐