Title :
Fast cross correlation algorithm for optical flow estimation
Author :
Sadykhov, R. Kh ; Lamovsky, D.V.
Author_Institution :
United Inst. of Informatics Problems, Nat. Acad. of Sci. of Belarus, Minsk
Abstract :
This paper describes a new algorithm to calculate cross-correlation function. We have used this algorithm for real time optical flow estimation between frames of video sequence. We combined box filtering technique for calculation of cross correlation coefficients with parallel processing using MMX/SSE technology of modern general purpose processors. Our algorithm was tested on real world video sequences obtained from the cameras of video surveillance system
Keywords :
correlation methods; filtering theory; image sequences; parallel processing; real-time systems; video signal processing; video surveillance; MMX-SSE technology; box filtering technique; cameras; fast cross correlation algorithm; parallel processing; real time optical flow estimation; video sequence; video surveillance system; Cameras; Field programmable gate arrays; Filtering algorithms; Hardware; Image motion analysis; Informatics; Optical filters; Parallel processing; Video sequences; Video surveillance;
Conference_Titel :
Signal Processing Symposium, 2006. NORSIG 2006. Proceedings of the 7th Nordic
Conference_Location :
Rejkjavik
Print_ISBN :
1-4244-0412-6
Electronic_ISBN :
1-4244-0413-4
DOI :
10.1109/NORSIG.2006.275245
