Intelligent technique for scene cut detection from MPEG video

Author

Al-Hammadi, A.S. ; Dawood, Ali M.

Author_Institution

Multimedia Commun. & Signal Process. Res. Group, Etisalat Coll. of Eng., Sharjah, United Arab Emirates

fYear

2003

fDate

27-29 July 2003

Firstpage

73

Lastpage

77

Abstract

In this paper we propose a neuro-fuzzy (NF) algorithm that automatically detects the type of MPEG frame at which a clean scene cut occurred. The algorithm is based on detection of the majority of intra-macroblocks and directionality of the two neighboring B-frame motion vector types from the bit stream, it is shown that the method is able to identify the location of the scene cut I-frame, P-frame and each B-frame with a very high success rate.

Keywords

block codes; feature extraction; fuzzy logic; image motion analysis; multimedia databases; neural nets; video coding; video databases; I-frame; MPEG frame type detection; MPEG video; MPEG-7; P-frame; bit stream; feature extraction; high-level video processing; intramacroblock; motion vector type; neighboring B-frame directionality; neuro-fuzzy algorithm; scene cut detection; scene cut location; Bit rate; Image coding; Layout; Multimedia communication; Multimedia databases; Signal processing algorithms; Streaming media; Transform coding; Video compression; Video signal processing;

fLanguage

English

Publisher

ieee

Conference_Titel

Virtual Environments, Human-Computer Interfaces and Measurement Systems, 2003. VECIMS '03. 2003 IEEE International Symposium on

Print_ISBN

0-7803-7785-0

Type

conf

DOI

10.1109/VECIMS.2003.1227033

Filename

1227033