Method for positioning and extracting video frequency caption by supporting vector computer

Abstract

The method includes following steps: first the pixel feature of image waiting for test is extracted, and then it will be divided into 11X11 image sub-block; the pixel gray value of each image sub-block is extracted by using a mode as a Chinese character 'mi' to take it as input of support vector machine; then the tool of support vector machine LINSVM is used to construct sort learning machine of support vector machine; mass samples are used to train the support vector machine for making it correctly recognize subtitle block and non subtitle block; then pyramid type model is used to successively reduce the original image resolution, and the subtitle is checked by support vector machine at each stage; the checking results of all stages are restored to original image resolution for composition, and ultimate checking result is generated; finally all tested image sub-blocks are made post-treatment, clearing noise to get combined subtitle area.

Description

technical field [0001] The invention relates to a method for locating and extracting video subtitles by using a support vector machine, which is used for video image processing and computer vision. It belongs to the technical field of computer information image processing. Background technique [0002] The subtitles contained in the video stream express rich semantics and can play an effective role in the analysis and understanding of the original video stream. For example, subtitles in video news reports generally summarize important information such as the time, place, person and main event of the reported news. Video subtitles provide highly generalized semantics for corresponding video streams, and can segment and mark corresponding video streams after automatic positioning, extraction and recognition, thereby realizing browsing and retrieval based on semantic video streams. Video stream subtitles can be divided into two categories, one is the artificial subtitles synt...

AI Technical Summary

Problems solved by technology

[0003] However, only utilizing the properties of the subtitle itself to locate the video stream subtitles faces many difficulties: (1) the size of the video stream subtitles often changes, and in the same scene video image sequence, large-size and small-size subtitles will appear simultaneously; ( 2) The fonts of the subtitles of the video stream are diverse, and the character forms of different languages ​​are different. Even for the same language, there are characters with various shapes; (3) The colors of the video subtitles can be various, and the video background color It can also be diverse, and the subtitles are embedded in the video background, so the color information of the subtitles is unpredictable and complex; (4) In some cases, the subtitles are translated left and right or vertically moved up and down, and the location of the subtitles should be considered when extracting exercise state

Images