Rapid video copy detection on compressed domain

ZHANG Yong-dong; ZHANG Dong-ming; GUO Jun-bo; TANG Sheng

您当前的位置：

首页 >

文章列表页 >

Rapid video copy detection on compressed domain

更新时间：2024-10-14

- Rapid video copy detection on compressed domain
- Vol. 30, Issue 3, Pages: 135-140(2009)
- 作者机构：
  
  中国科学院计算研究所前瞻研究实验室
- 作者简介：
- 基金信息：
- DOI：
  CLC： TP391.41
- Published：2009
- 稿件说明：
移动端阅览
ZHANG Yong-dong, ZHANG Dong-ming, GUO Jun-bo, et al. Rapid video copy detection on compressed domain[J]. 2009, 30(3): 135-140.
DOI：

ZHANG Yong-dong, ZHANG Dong-ming, GUO Jun-bo, et al. Rapid video copy detection on compressed domain[J]. 2009, 30(3): 135-140. DOI：

摘要

为了在大规模数据下有效提升拷贝检测速度

提出了一种压缩域视频拷贝检测算法

采用2级匹配框架

利用压缩域上DCT系数顺序度量特征进行相似度匹配

显著减少了查询视频解码操作;并为DCT系数顺序度量特征建立倒排索引

以加快第一级匹配的速度。实验结果表明

与已有的算法相比

所提出算法在保持相当的检测精度条件的同时

大幅提升了检测速度。

Abstract

To reduce the detection time efficiency under large scale data environment

a rapid algorithm was proposed on compressed domain using a two-level hierarchical detection scheme.The ordinal measures of DCT coefficients were adopted as visual features for similarity-matching in order to reduce the computational load in video decoding.Inverted indexing structure was used to accelerate the first level detection process.The experiment results show

compared with the previous algorithm

the algorithm can improve the detection speed obviously with the similar detection precision.

关键词

Keywords

references

Views

901

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Hiss noise reduction method used for audio signals based on AAC bit-stream

Video transcoding for H.264 in the compressed domain and key techniques analysis

Related Author

Feng DENG

Chang-chun BAO

Feng BAO

YANG Gao-bo1

YU Sheng-fa2

ZHANG Zhao-yang3

Related Institution

Speech and Audio Signal Processing Lab,School of Electronic Information and Control Engineering,Beijing University of Technology

School of Computer and Communication,Hunan University

School of Electronic Science and Engineering,National University of Defense Technology

AI问答

⁰