中国邮电高校学报(英文) ›› 2012, Vol. 19 ›› Issue (3): 100-106.doi: 10.1016/S1005-8885(11)60271-9

• Others • 上一篇    下一篇

Combining SVM and CHMM classifiers for porno video recognition

赵志诚   

  1. Beijing University of Posts and Telecommunications
  • 收稿日期:2011-12-27 修回日期:2012-03-22 出版日期:2012-06-30 发布日期:2012-06-08
  • 通讯作者: 赵志诚 E-mail:zhaozc@bupt.edu.cn
  • 基金资助:

    视频选择性注意机理与语义特征提取;视觉注意模型在语义视频搜索中的应用

Combining SVM and CHMM classifiers for porno video recognition

Zhi-Cheng ZHAO   

  1. Beijing University of Posts and Telecommunications
  • Received:2011-12-27 Revised:2012-03-22 Online:2012-06-30 Published:2012-06-08
  • Contact: Zhi-Cheng ZHAO E-mail:zhaozc@bupt.edu.cn

摘要:

Porno video recognition is important for Internet content monitoring. In this paper, a novel porno video recognition method by fusing the audio and video cues is proposed. Firstly, global color and texture features and local scale-invariant feature transform (SIFT) are extracted to train multiple support vector machine (SVM) classifiers for different erotic categories of image frames. And then, two continuous density hidden Markov models (CHMM) are built to recognize porno sounds. Finally, a fusion method based on Bayes rule is employed to combine the classification results by video and audio cues. The experimental results show that our model is better than six state-of-the-art methods.

关键词:

ioporno video recognition, SVM, keyframe, CHMM, aud

Abstract:

Porno video recognition is important for Internet content monitoring. In this paper, a novel porno video recognition method by fusing the audio and video cues is proposed. Firstly, global color and texture features and local scale-invariant feature transform (SIFT) are extracted to train multiple support vector machine (SVM) classifiers for different erotic categories of image frames. And then, two continuous density hidden Markov models (CHMM) are built to recognize porno sounds. Finally, a fusion method based on Bayes rule is employed to combine the classification results by video and audio cues. The experimental results show that our model is better than six state-of-the-art methods.

Key words:

ioporno video recognition, SVM, keyframe, CHMM, aud