|
|
|
Violent Videos Classification Algorithm Based on Bag of Audio Words
|
LI Rongjie1,JIANG Xinghao1,2,SUN Tanfeng1,2
|
（1. School of Information Engineering Security, Shanghai Jiaotong University, Shanghai 200240, China; 2. Shanghai Information Security Management and Technology Research Key Lab, Shanghai 200240, China）
|
|
|
|
Abstract A new method to classify the violent videos by the bag of audio words was introduced. The MPEG7 audio descriptors are firstly extracted, including the low level features such as AudioSpectrumCentroid and AudioSpectrumSpread etc. After that, the audio words are built through the MPEG7 high level descriptor, the AudioSighnature, which is considered as the fingerprint of the audio stream. The support vector machine is used to classify the feature vectors into two genres, which are the violent and nonviolent. There are three experiments in this paper: the research on the different types of the audio words, the different size of words and the classification of the shots detected from the visual features. It is demonstrated from the experiment result that the proposed method achieves good recall accuracy.
|
|
Received: 13 June 2010
Published: 28 February 2011
|
|
|
|
|
|
| [1] |
ZHENG Gang, LU Yao-An, ZHU Li-Min. Interference-Free Tool Positioning for Five-Axis Sculptured Surface Machining Using ThirdOrder Point Contact Approach [J]. J. Shanghai Jiaotong Univ.(Sci.) , 2012, 46(02): 172-177. |
| [2] |
WANG Hui-Lin, RAN Cheng-Xin, HUANG Wei, MA Man-Hao. A Time Step Alterable and Predicative Electronic Reconnaissance Satellite Guiding Scheduling Technique[J]. J. Shanghai Jiaotong Univ.(Sci.) , 2011, 45(07): 954-960. |
| [3] |
LIU Huan-Xi-１, WU Zhe-１, ZHU Jun-2, LI Xiong-1, LIU Yun-Cai-1. Unimodal Biometric System Based on Subspace Learning Method[J]. J. Shanghai Jiaotong Univ.(Sci.) , 2011, 45(07): 970-974. |
| [4] |
Wang-Lin, ZHANG Guo-Zhong, ZHU Hua-Yong, SHEN Lin-Cheng. Adaptive Consensus Fusion Estimation for Mobile Sensor Networks[J]. J. Shanghai Jiaotong Univ.(Sci.) , 2011, 45(03): 383-0387. |
|
|
|
|
