摘要
字幕分割是指对检测定位到的视频字幕图像进行分割,使其字符像素与本底背景像素分离,二值化为可供OCR软件识别的字幕图像。为了克服字幕图像分割中容易出现的过分割及欠分割现象,提出一种基于字幕区域和外扩区域"白像素"数量增量比判决的字幕图像分割算法,该算法通过逐步改变图像分割阈值,以分析图像分割结果作为反馈来判决当前分割效果的好坏,从而确定最优分割阈值。大量结果表明,该算法性能良好,其分割效果远优于传统的OTSU算法、K均值聚类等经典算法。
Video caption segmentation is the process of extracting binary text pixels from caption image for inputting OCR recognition software. In order to overcome the shortcomings of the over-segmentation and under-segmentation problems of the traditional segmentation methods, a method is proposed based on ＂white pixels＂ increment ratio. This method finds the optimal threshold using a feed-back model by gradually changing the segmenting thresh- old. In order to verify the performance of the proposed segmentation method, several experiments under different complex conditions are designed. The experiment results show that the proposed method can resolve the over-segmenting and under-segmenting problems of the traditional segmentation meth-ods such as the OTSU algorithm and the K-means clustering algorithm.
出处
《电视技术》
北大核心
2014年第5期190-193,共4页
Video Engineering
关键词
视频字幕
图像分割
白像素增量比
二值化
video caption
image segmentation
white pixels increment ratio
binarization
