Research on weighted naive Bayesian classifier in discriminative tracking

DOI

10.2991/meic-14.2014.386How to use a DOI?

Keywords

discriminative tracking naive bayesian classifier Gaussian function object tracking traind sample

Abstract

Aiming at the question that each positive sample in trained sample set has the same weight when training the naive Bayesian classifier in discriminative tracking algorithm, an algorithm is proposed to put different weight on different positive sample according to the distances between the positive samples and the object, the distance is smaller, the similarity between them is bigger, the weight on this positive sample should be bigger, otherwise, the weight on this sample should be smaller. Taking the distribution of positive samples into consideration, and using the distance between the positive sample and the object to measure the weight of this positive sample, and put different weights on different positive samples. The results of experiments show that this change can prompt the performance of the classifier, and further improve the robustness of the tracking algorithm.

Copyright

© 2014, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Yan Wang
AU  - Dawei Yang
AU  - Guangsan Li
PY  - 2014/11
DA  - 2014/11
TI  - Research on weighted naive Bayesian classifier in discriminative tracking
BT  - Proceedings of the 2014 International Conference on Mechatronics, Electronic, Industrial and Control Engineering
PB  - Atlantis Press
SP  - 1714
EP  - 1719
SN  - 2352-5401
UR  - https://doi.org/10.2991/meic-14.2014.386
DO  - 10.2991/meic-14.2014.386
ID  - Wang2014/11
ER  -