Abstract
This paper describes a new algorithm to calculate cross-correlation function. We combined box filtering technique for calculation of cross correlation coefficients with parallel processing using MMX/SSE technology of modern general purpose processors. We have used this algorithm for real time optical flow estimation between frames of video sequence. Our algorithm was tested on real world video sequences obtained from the cameras of video surveillance system.
Highlights
Feature matching is an important task in the area of computer vision
Three classes of metrics are commonly applied for area matching: cross correlation (CC), intensity differences (sum of absolute differences (SAD), sum of squared differences (SSD)), and rank [3] metrics
Te most computationally expensive stage of cross correlation function calculation is computing of equation (1)
Summary
Feature matching is an important task in the area of computer vision. There are several different approaches for image correspondence estimation [1]. The main problem of most correlation based techniques is that they are computationally expensive This fact doesn’t allow using them in real time video processing. Description of the algorithms that compute SAD (sum of absolute differences) metrics using box filtering technique can be found in [8, 9] This technique in [10, 11] is used for fast NCC (normalize cross-correlation) metrics. In our paper we use MMX/SSE technology for developing fast box filtering technique based algorithm to perform cross correlation calculation and employ this algorithm for real time optical flow estimation. Vsumi(n) is the sum of products of pixel intensity computational costs for calculation cross correlation values from column n of window with center in (i,j) function. The main idea of the technique is that for in this method for fast cross correlation between calculation Vsumj(n) it is necessary to update Vsumjframes
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
