Title :
A Dynamic Tracing Algorithm Based High Speed Camera
Author :
Wang, Baobao ; Chen, Lingzhi ; Xiao, Yang
Author_Institution :
Dept. of Comput. Sci. & Technol., Xidian Univ., Xian, China
fDate :
March 31 2009-April 2 2009
Abstract :
A dynamic tracing Algorithms for the detection of industrial sewing machines on images sequences from a high-speed camera is presented. First, feature extraction is completed with the Harris Corner Detection Algorithms based on the change of intensity; Then the prediction of the move tendency of the corners with Extended Kalman Filter is accomplished; By using of the Neighborhood Matching Algorithms the paths of the component tracked is obtainable finally . The Harris Corner Detection Algorithms has the properties of translation, rotation and scaling invariance and high precision, the Neighborhood Matching Algorithms has the advantage of low computational complexity , the combination with Extended Kalman Filter increase the efficiency and the accuracy of the system. Experimental results show that it have a good performances in the problem proposed.
Keywords :
Kalman filters; cameras; computational complexity; feature extraction; nonlinear filters; object detection; sewing machines; Harris corner detection algorithms; computational complexity; dynamic tracing algorithm; extended Kalman filter; feature extraction; high speed camera; image sequences; industrial sewing machine detection; neighborhood matching algorithms; rotation; scaling invariance; translation; Cameras; Change detection algorithms; Computer science; Detection algorithms; Eigenvalues and eigenfunctions; Feature extraction; Heuristic algorithms; Image sequences; Particle filters; Trajectory; Dynamic Tracing Algorithms; Extended Kalman Filter; Harris Corner Detection; Neighborhood Matching Algorithms;
Conference_Titel :
Computer Science and Information Engineering, 2009 WRI World Congress on
Conference_Location :
Los Angeles, CA
Print_ISBN :
978-0-7695-3507-4
DOI :
10.1109/CSIE.2009.572