Extreme learning machine based fast object recognition
Date of Issue2012
International Conference on Information Fusion (FUSION) (15th : 2012)
School of Electrical and Electronic Engineering
Extreme Learning Machine (ELM) as a type of generalized single-hidden layer feed-forward networks (SLFNs) has demonstrated its good generalization performance with extreme fast learning speed in many benchmark and real applications. This paper further studies the performance of ELM and its variants in object recognition using two different feature extraction methods. The first method extracts texture features, intensity features from Histogram and features from two types of color space: HSV & RGB. The second method extracts shape features based on Radon transform. The classification performances of ELM and its variants are compared with the performance of Support Vector Machines (SVMs). As verified by simulation results, ELM achieves better testing accuracy with much less training time on majority cases than SVM for both feature extraction methods. Besides, the parameter tuning process for ELM is much easier than SVM as well.
DRNTU::Engineering::Electrical and electronic engineering
© 2012 International Society of Information Fusion. This paper was published in 2012 15th International Conference on Information Fusion (FUSION) and is made available as an electronic reprint (preprint) with permission of International Society of Information Fusion. The paper can be found at the following official URL: http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=6289984&url=http%3A%2F%2Fieeexplore.ieee.org%2Fxpls%2Fabs_all.jsp%3Farnumber%3D6289984. One print or electronic copy may be made for personal use only. Systematic or multiple reproduction, distribution to multiple locations via electronic or other means, duplication of any material in this paper for a fee or for commercial purposes, or modification of the content of the paper is prohibited and is subject to penalties under law.