The Complex Background Of Video Text Localization And Segmentation

Text information in the video is an important clue for helping to understand videocontent, extracting the text information effectively is a key technology for application ofautomatic understanding of video content and retrieval. Commercial OCR also can noteffectively identify the video text images with background is complex and strong interference,looking for a universal method of text positioning and segmentation to improve the efficiencyof identification is one of the hot issues in the present study.This paper focuses on the method of video text localization and segmentation. Accordingto the characteristics of the video text under complex background, text positioning method ofcorner extraction combined with morphological is adopted, aiming at the problems ofdifferent size of characters, the multi-scale corner detection method is proposed, combinedwith characteristics of the characters with the same color, the method of color clustering onthe text area is used to get a more precise positioning; utilizing the characteristics of thewavelet-domain Hidden Markov Tree models widely used, the character and backgroundlocated in the text area is modeled in the wavelet-domain, combining with the scale fusiontheory and making full use of the class logo features on different scales, the problem ofsegmentation of the characters in the text is solved better, and images can be get.The experimental results show that the improved text positioning algorithm caneffectively detect text messages of complex video, not only detect the text messages of ballgame scene and video of news type, but also can be used in natural scene of the text of thedetection positioning, The segmentation results of the wavelet domain HMT model is betterthan the Otsu. In maximum keep text information cases, it emphasizes the text features, and topromote the quality of the segmentation words strokes.