Title :
The Detection of TV Commercial Based on Multi-Feature Fusion
Author :
Fei, Ge ; Ping, Shi
Author_Institution :
Sch. of Inf. Eng., Commun. Univ. of China, Beijing, China
Abstract :
A new method for TV commercial detection based on multi-feature fusion is proposed. Firstly, the video and audio features of TV commercials are analyzed and extracted. Then, TV commercial detection is performed by using these video and audio features as well as a statistical analysis method. Finally, experiments are implemented and results are analyzed. Experimental results show that TV commercials can be detected effectively by using this method.
Keywords :
audio signal processing; feature extraction; image fusion; statistical analysis; television; video signal processing; TV commercial detection; audio feature extraction; multifeature fusion; statistical analysis method; video feature extraction; Data mining; Detection algorithms; Feature extraction; Histograms; Image edge detection; Statistical analysis; TV;
Conference_Titel :
Multimedia Technology (ICMT), 2010 International Conference on
Conference_Location :
Ningbo
Print_ISBN :
978-1-4244-7871-2
DOI :
10.1109/ICMULT.2010.5630899
