Title :
SIFT-based algorithm for object matching and identification
Author :
Yao, Minghai ; Zhu, Hua ; Gu, Qinlong ; Zhu, Licheng ; Qu, Xinyu
Author_Institution :
Coll. of Inf. Eng., Zhejiang Univ. of Technol., Hangzhou, China
Abstract :
Object Recognition is the base and Guarantee of the robot moving object tracking system, the accuracy of the object recognition is directly effect the evaluation of the tracking system. an algorithm based on combining the SIFT and Kalman filter is proposed, using the SIFT to match the feature points, through Kalman filter algorithm to got the smallest effect of noise, and then simulate the algorithm, using sets of pictures with different types of noise of different intensity for object recognition, verify the high accuracy of the algorithm.
Keywords :
Kalman filters; feature extraction; image matching; image motion analysis; object recognition; object tracking; Kalman filter algorithm; SIFT based algorithm; feature point matching; object identification; object matching; object recognition; robot moving object tracking system; Accuracy; Educational institutions; Kalman filters; Matched filters; Noise; Object recognition; SIFT; kalman filter; object recognition;
Conference_Titel :
Remote Sensing, Environment and Transportation Engineering (RSETE), 2011 International Conference on
Conference_Location :
Nanjing
Print_ISBN :
978-1-4244-9172-8
DOI :
10.1109/RSETE.2011.5965512
