一种结合边缘和角点特征的视频字幕检测方法.pdf

Scientific Journal of Information Engineering October 2013, Volume 3, Issue 5, PP.85-91 A Video Caption Detection Method with Edge and Corner Features 1 1 2 Yunlan Zhao †, Yonghong Song , Yi Zhou 1. Software Engineering School, Xi’an Jiaotong University, Xi’an Shanxi 710049, China 2. Broadcast and TV Center, propaganda department, Xi’an Jiaotong University, Xi’an Shanxi 710049, China †Email: 617732720@ Abstract Video caption contains important clue of video content, so it plays a significant role in the content based video retrieval. In this paper, a video text detection approach which combines edge and corner point features is proposed. Firstly, the gradient map is obtained by applying Roberts edge operator on the three color components from a single video frame. Then the Harris corner point is extracted on the gradient maps. Thirdly, the candidate text regions are acquired by morphological operations. Finally, the captions are located using the proposed text features and connected component analysis. The experiment results show the proposed approach can detect captions with high speed and high accuracy. Keywords: Caption Detection; Roberts Gradient Operator ; Corner Detection; Video Text Features 一种结合边缘和角点特征的视频字幕检测方法 1 1 2 赵云兰 ，宋永红 ，周颐 1. 西安交通大学 软件学院，陕西 西安 710049 2. 西安交通大学 宣传部广电中心，陕西 西安 710049 摘 要：视频中的字幕提供了描述视频内容的有用信息，对于基于关键字的视频检索具有重要作用。对视频中字幕进行了 研究，提出了一种结合视频帧中的边缘和角点特征进行字幕检测的方法。该方法首先提取彩色视频帧在RGB 3 个颜色分量 上的彩色Roberts 梯度算子，得到凸显边缘区域的灰度图像，然后在得到的灰度图像上提取Harris 角点，再通过形态学处理 得到候选字幕连通区域，最后结合本文提出的一些视频文本特征进行连通区域分析得到最终字幕区域。实验证明，该算法 能够快速定位字幕区域，定位准确性较