An Object Recognition Method Based on Bag-of-Visual-Words and Fusing Multi-feature

DOI

10.2991/lemcs-14.2014.216How to use a DOI?

Keywords

object recognition;BOV model;fusing multi-feature;SIFT;LBP

Abstract

The traditional bag-of-visual-words(BOV) model only uses one single feature to classify objects, which is difficult to achieve good results when dealing with many object categories. To solve this problem, in this paper, we proposed an object recognition method based on BOV model and fusing multi-feature. First, it extracted Scale Invariant Feature Transform (SIFT) features and Local Binary Pattern(LBP) features in the multi-scale space simultaneously, Second, we utilized SVM classifier to pre-classify separately on these two kinds of features and assigned their weights 0 or 1 according to the scale of pre-classification results. Third, the method fused SIFT and LBP features by introducing their weights, obtaining a fused vector. At last, classified on the fused vector using SVM classifier again, and then achieved the final result of object recognition. The experimental results show that the method proposed in this paper shows good performance and could improve the accuracy of object recognition effectively.

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  - Xueting Qi
AU  - Tianhuang Chen
AU  - Hongxia Wang
PY  - 2014/05
DA  - 2014/05
TI  - An Object Recognition Method Based on Bag-of-Visual-Words and Fusing Multi-feature
BT  - Proceedings of the International Conference on Logistics, Engineering, Management and Computer Science
PB  - Atlantis Press
SP  - 959
EP  - 963
SN  - 1951-6851
UR  - https://doi.org/10.2991/lemcs-14.2014.216
DO  - 10.2991/lemcs-14.2014.216
ID  - Qi2014/05
ER  -