Title :
Real-time local texture feature target tracking method
Author :
Song, Hua-Jun ; Wang, Yan-Jiang
Author_Institution :
Coll. of Inf. & Control Eng., China Univ. of Pet.(East China), Dongying, China
Abstract :
Precision and Speed are two conflict indexes in target tracking system. Good feature extracting method is key to precision. Local texture feature has better performance when target rotated or illumination changed. Moreover, its compute time is very short. In order to accelerate tracking algorithm, a new target tracking method is presented by combines the local texture feature and different value pyramid match methods. Experiment results indicate that the precision of this method is more than the traditional methods and the target displacement can be computed less than 20 ms which can meet the requirements of real time target tracking system.
Keywords :
feature extraction; image matching; image texture; object tracking; feature extracting method; illumination; pyramid match methods; real time local texture feature; target tracking method; Algorithm design and analysis; Feature extraction; Histograms; Real time systems; Signal processing algorithms; Target recognition; Target tracking; LBP; model matching; pyramid mehtod; target track;
Conference_Titel :
Image and Signal Processing (CISP), 2011 4th International Congress on
Conference_Location :
Shanghai
Print_ISBN :
978-1-4244-9304-3
DOI :
10.1109/CISP.2011.6099904