Text in images and videos carries plenty of semantic information useful for understanding thecontent of images and videos. Thus, it makes text recognition very significant for understanding andretrieval of images and video. However, text is usually embedded in complex background of images,which makes direct optical character recognition almost impossible. Therefore, it becomes necessary toextract text from complex background before recognition. In the last decades, many efforts have beendevoted to developin...
| 
点击下列相关研究机构和相关文献作者，可以直接查到这些机构和作者被《中国知识资源总库》收录的其它文献，使您全面了解该机构和该作者的研究动态和历史。
