Title :
A Single-class Support Vector Machine Translation Algorithm To Compensate For Non-stationary Data In Heterogeneous Vision-based Sensor Networks
Author :
Rhinelander, Jason ; Liu, Peter X.
Author_Institution :
Dept. of Syst. & Comput. Eng., Carleton Univ., Ottawa, ON
Abstract :
This paper develops a translation algorithm that adapts an existing support vector machine (SVM) to observations that have a different probability distribution than originally trained with. The primary advantage of this algorithm is that the re-training can be avoided. The support vector translation algorithm can be used in a fully distributed vision-based sensor network for target classification and tracking. Preliminary results are discussed and planned future work is briefly outlined.
Keywords :
image classification; image sensors; probability; support vector machines; heterogeneous vision-based sensor networks; machine learning; nonstationary data; pattern recognition; probability distribution; single-class support vector machine translation algorithm; Cameras; Computational intelligence; Distributed computing; Equations; Intelligent sensors; Layout; Quadratic programming; Sensor systems; Support vector machine classification; Support vector machines; Support vector machine; heterogeneous vision-based sensor network; machine-learning; pattern recognition;
Conference_Titel :
Instrumentation and Measurement Technology Conference Proceedings, 2008. IMTC 2008. IEEE
Conference_Location :
Victoria, BC
Print_ISBN :
978-1-4244-1540-3
Electronic_ISBN :
1091-5281
DOI :
10.1109/IMTC.2008.4547203