| 基于音频分析的视频场景检测 |
| |
| 引用本文： | 倪宁,卢刚,卜佳俊. 基于音频分析的视频场景检测[J]. 计算机仿真, 2006, 23(8): 184-187,195 |
| |
| 作者姓名： | 倪宁 卢刚 卜佳俊 |
| |
| 作者单位： | 杭州科技职业技术学院计算机系,浙江,杭州,310022;浙江大学计算机科学与工程系,浙江,杭州,310027 |
| |
| 摘 要： | 目前场景检测的研究，主要是基于图像和视频。但音频同样具有丰富的场景信息，基于音频分析的计算量是比较少的，对自动或者半自动的场景检测，基于音频分析的方法也是更为让用户接受的。可以把基于音频分析的方法作为视频场景检测的辅助手段，以获得更为准确的场景检测和分割。该文提出了一个基于内容的音频分析系统，对视频序列实现基于音频分析的场景检测和分割。该系统能有效的解决许多诸如图像变化了，而实际场景并未变化的情形。且本系统整体运算复杂度较基于视频／图像的场景检测与分割系统要低。
|
| 关 键 词： | 音频分析 场景检测 音频特征 隐马尔可夫模型 |
| 文章编号： | 1006-9348（2006）08-0184-04 |
| 收稿时间： | 2005-07-25 |
| 修稿时间： | 2005-07-25 |
Inspection of Video Frequency Scene Based on Audio Frequency Analysis |
| |
| NI Ning,LU gang,BU Jia-jun. Inspection of Video Frequency Scene Based on Audio Frequency Analysis[J]. Computer Simulation, 2006, 23(8): 184-187,195 |
| |
| Authors: | NI Ning LU gang BU Jia-jun |
| |
| Affiliation: | 1. Department of Computer , Hangzhou Poly Technique College, Hangzhou Zhejiang 310012, China; 2. Department of Computer Science and Engineering, ZheJiang University, Hangzhou Zhejiang 310027,China |
| |
| Abstract: | The present study on scene detection is mainly based on image and video sequence. Audio sequence, however, also has rich information. Therefore, a method based on audio analysis can be applied as an auxiliary, so as to get more accurate scene detection and split. In this paper, a content - based audio data analysis system is proposed, which accepts video sequence and implements scene detection and segmentation from audio part. The system can solve problems such as changing of image without changing of scene, and the complexity of operation in this system is much lower than that of the system based on image and video sequence analysis. |
| |
| Keywords: | Audio analysis Scene detection Audio feature Hidden markov model |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
