作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2010, Vol. 36 ›› Issue (3): 195-196. doi: 10.3969/j.issn.1000-3428.2010.03.065

• 人工智能及识别技术 • 上一篇    下一篇

基于相邻尺度小波变换乘积的镜头渐变检测

黄 敏,姜 静   

  1. (郑州轻工业学院计算机与通信工程学院,郑州 450002)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2010-02-05 发布日期:2010-02-05

Shot Gradual Changes Detection Based on Adjacent Scale Wavelet Transformation Product

HUANG Min, JIANG Jing   

  1. (School of Computer and Communication Engineering, Zhengzhou University of Light Industry, Zhengzhou 450002)
  • Received:1900-01-01 Revised:1900-01-01 Online:2010-02-05 Published:2010-02-05

摘要: 镜头边界检测是进行视频内容分析的首要步骤。提出一种基于相邻尺度小波变换乘积的镜头边缘检测方法。对镜头内的每帧视频图像进行小波变换,求出各帧图像相邻尺度小波变换尺度积的模和方向。计算连续帧图像尺度积与模之间的绝对距离,得到2帧图像间的不连续值。通过自适应阈值分割,检测出镜头边界。

关键词: 镜头渐变, 边缘检测, 小波变换, 尺度积

Abstract: Shot boundary detection is the chiefly step of video content analysis. This paper proposes a method of shot boundary detection based on the product of adjacent scale wavelet transform. Each video frame image in the shot is carried on wavelet transform. The modulus and direction of adjacent scale wavelet transform scale product of every frame image are obtained. It calculates absolute distance among consecutive frames image scale plot and modulus, and gets discrete value between two frame images. The shot boundary is detected by using adaptive threshold segmentation.

Key words: shot gradual changes, edge detection, wavelet transformation, scale product

中图分类号: