Abstract
Image classification plays a significant role in pattern recognition. In the recent past, due to the advancements in imaging technology, massive data are being generated through various image acquisition techniques. Classifying these massive images is a challenging task among the researchers. This paper presents a novel feature selection method to improve the performance of image classification. The performance of the proposed method is tested on the publically available real image dataset and compared with various state-of-the-art feature selection methods. The experimental results show that the proposed method outperforms the other state-of-the-art methods..
Keywords
Image classification, Feature selection, Image recognition, Clustering, Image acquisition techniques.
Citation
D. ASIR ANTONY GNANA SINGH, S. APPAVU ALIAS BALAMURUGAN, E. JEBAMALAR LEAVLINE, A novel feature selection method for image classification, Optoelectronics and Advanced Materials - Rapid Communications, 9, 11-12, November-December 2015, pp.1362-1368 (2015).
Submitted at: Sept. 27, 2015
Accepted at: Oct. 28, 2015