Abstract—PC-SVM is a new developed support vector machine classifier with probabilistic constrains which presence of samples probability in each class is determined based on a distribution function. The presence of noise causes incorrect calculation of support vectors thereupon margin can not be maximized. In the Pc-SVM, constraints boundaries and constraints occurrence have probability density functions which it helps for achieving maximum margin. The main target of this paper is introducing a robust visual object recognition based on PC- SVM. Human detection is used as benchmark problem for the proposed algorithm. Experimental results show superiority of the probabilistic constraints support vector machine (PC-SVM) relative to standard SVM in human detection.
Index Terms—pc-svm, human detection, histograms of oriented gradients.
S. M. Hosseini is with the Islamic Azad University Of Firoozkooh, electrical engineering group, Iran; e-mail: M.hosseini@ Birjand.ac.ir).
H. Farsi was with University Of Birjand, Iran. He is now with the Department of Electrical Engineering, Birjand, Iran (e-mail: HFarsi@Birjand.ac.ir).
[PDF]
Cite: Seyyed Meysam Hosseini, Hasan Farsi, "A Robust Method Applied to Human Detection,"
International
Journal of Computer Theory and Engineering vol. 2, no. 5, pp. 692-694, 2010.
