欢迎访问《图学学报》 分享到:

图学学报

• 视觉与图像 • 上一篇    下一篇

基于自动确定特征点数的镜头分割方法

  

  • 出版日期:2013-06-29 发布日期:2015-06-11

Approach to Shot Segmentation Based on Automatic Decision of Feature-Point Number

  • Online:2013-06-29 Published:2015-06-11

摘要: :镜头分割对实现基于图像内容的视频检索具有重要作用。为了提高视频镜
头分割的效果,采用Harris 算子进行像素点的能量评估,并抑制伪特征点。通过图像帧分块
和边缘检测,获取最小化特征点集,并使特征点分布相对均匀合理,然后根据特征点计算图
像帧统计量,进一步形成图像帧的特征向量。在此基础上,结合图像帧渐变检测和闪光检测
实现镜头分割方法,取得很好镜头分割效果。

关键词: Harris 算子, 特征提取, 闪光检测, 渐变检测, 镜头分割

Abstract: Shot segmentation is very important for video retrieval based on image content. In
order to improve the performance of shot segmentation, Harris operator is adopted to evaluate the
energy of image pix, and exclude pseudo feature points. By means of partition of frame in the
video into blocks and edge detection of the blocks, the minimal feature-point set is obtained and
the feature points are made relatively well-distributed. And then according to the feature points,
statistical quantity of frame in video is computed to construct feature vector to express the frame.
Based on above, along with gradual transition detection and flash detection of frame, a method of
shot segmentation is presented. Experiments of shot segmentation are efficient.

Key words: Harris operator, feature extraction, flash detection, gradual transition detection;
shot segmentation