期刊名称：International Journal of Multimedia and Ubiquitous Engineering
印刷版ISSN：1975-0080
出版年度：2016
卷号：11
期号：10
页码：407-420
出版社：SERSC
摘要：Overlay text in video carries important semantic clues for video information retrieval and summarization. In this paper, we propose a robust method that is able to accurately locate text lines and extract text even in complex video scene. In thetext localization stage, this paper adopts the method based on corner point. First, corner detection is used to extract cornersas text features from video frames. Then multi-layer filtering mechanism(MLFM)is used to locate the text lines, which consists of corners clustering, corners horizontal projection, background filteringand heuristic rules. This MLFMcan effectively remove the isolated corners, locate the text linesaccurately and remove the background orpseudo text linesautomatically. In the text extraction stage, this paper proposed a twice binarization method that combines with polarity judgment on image. The polarity judgment was used as a guide to adjust the first binarization threshold when we perform the first binarization. After the first binarization, a mainproportion of the image has been processed, and the rest will be processed by the second binarization. Experimental results show that this approach can fast and robustly locatetext linesand extract text in video evenundercomplex background.
关键词：;Text localization; Text extrac;tion; Polarity judgment; Corner; Twice ;binari;zation
