一种新的视频镜头边界检测及关键帧提取方法

A New Method of Video Shot Boundary Detection and Key Frame Extraction

根据视频邻域片段变化的特点提出了镜头边界系数模型,镜头边界系数具有良好的抗噪能力,且对视频帧的时域变化有较好的描述能力,可用于镜头边界检测与关键帧提取.它可以单独检测镜头边界,也可以与传统的基于帧差的镜头边界检测方法相结合检测镜头边界.在提取关键帧时,根据镜头边界系数的分布,自适应地确定镜头内的关键帧数,用非极大值抑制方法与基于镜头边界系数的优先级方法确定关键帧的位置.实验结果表明,该方法在镜头边界检测性能上要优于已有的基于帧差的镜头边界检测方法,提取的关键帧对镜头的视觉内容具有较好的表达能力,且可在视频回放时实时执行.

In this paper, a model of shot boundary coefficients (SBCs) is proposed by analyzing the characteristics of video neighboring slice changes. SBC is of excellent anti-noise ability and can describe the time-domain change of frames well. So it can be employed to detect shot boundaries and extract key frames. It detects shot boundaries in two ways: the direct use of SBC and the combination of the SBC and the traditional frame difference-based methods. For the extraction of key frames, the number of key frames is self-determined, subject to the distribution feature of SBCs, and the replacements of key frames are determined by means maximum suppression and SBC-based prior levels. Experimental results show that the performance of shot boundary detection is superior to the traditional frame difference-based methods, and the extracted key frames can well express the video contents.