A Porn Video Detecting Method Based on Motion Features Using HMM

Author

Qu, Zhiyi ; Liu, Yanmin ; Liu, Ying ; Jiu, Kang ; Chen, Yong

Author_Institution

Sch. of Inf. Sci. & Eng., Lanzhou Univ., Lanzhou, China

Volume

2

fYear

2009

fDate

12-14 Dec. 2009

Firstpage

461

Lastpage

464

Abstract

This paper proposes a method of identified reciprocating motion in pornographic video from other human action using Hidden Markov Model (HMM). The motion vectors are obtained by decoding the compressed MPEG video. Then the feature vectors are extracted by calculating the direction and the magnitude of the motion vectors. The feature vectors are fed to Hidden Markov Model for training and classification of actions. Six actions were trained with distinct HMM for classification. The correct recognition result is up to 90%.

Keywords

feature extraction; hidden Markov models; motion estimation; video coding; HMM; action classification; compressed MPEG video; feature vector extraction; hidden Markov model; human action; motion features; motion vectors; porn video detecting method; pornographic video; reciprocating motion identification; Feature extraction; Filters; Government; Hidden Markov models; Humans; Internet; Motion detection; Skin; Transform coding; Video compression; HMM; Motion features; Motion vectors; Porn video; Reciprocating motion detection;

fLanguage

English

Publisher

ieee

Conference_Titel

Computational Intelligence and Design, 2009. ISCID '09. Second International Symposium on

Conference_Location

Changsha

Print_ISBN

978-0-7695-3865-5

Type

conf

DOI

10.1109/ISCID.2009.261

Filename

5368715