Author : Subba Rao K 1
Date of Publication :7th March 2017
Abstract: Classification is the process of arranging the pixels into groups, called clusters that have some common characteristics. In this paper a Hybrid, and yet powerful classification method is proposed, which can be used to classify the textured and nontextured images. Traditional classification methods such as statistical classifiers, knowledge-based systems, and neural networks have number of limitations in classifying the images because of strict assumptions, particularly in the presence of the coarse pixels. The Ant Colony Optimization (ACO) is used to generate classification rules from the training set. Due to feedback property of the ACO, it considers all the changes into account in constructing the rules. These rules are then used in the process of classifying test set of the image. An entropy based fuzzy partitioning along with ACO is used to generate rules. ACO enables to construct simple rules to obtain better performance.
Reference :
-
[1] David Martens and Manu De Backer, "Classification With Ant Colony Optimization," IEEE Trans on Evolutionary Computing, Oct 2007, pp. 651-665.
[2] Xiaoping Liu and Xia Li, "An Innovative Method to Classify Remote Sensing Images Using Ant Colony Optimization," IEEE Trans. on Geosciences, Dec 2008, vol. 46, no. 12, pp. 4198-4208.
[3] D.LU and Q. WENG, "A survey of image classification methods and techniques," International Journal of Remote Sensing, March 2007,pp. 823-870.
[4] Jia Li and James Ze Wang, "Classification Of Textured And Non-Textured Images Using Region Segmentation," Proc. 7th Int. Conf. on Image Processing, 2000.
[5] T. Chattoopadhyay, B. Bhowmic and A. Sinha, "Application of image processing in industries," CSI Communication, July 2012, pp. 8-11.
[6] A. Colorni, M Dorigo et al. "Distributed Optimization by Ant Colonies," in Proc. 1st Eur. Conf. Artif. life, 1991, pp. 134-142.
[7] Kwang Mong Sim and Weng Hong Sun, "Multiple ant colony optimization for network routing," 1st Int.Symp Cyber Worlds, 2002 pp.277-281.
[8] R S Parpinelli, H S Lopes And A A Freitas, " An ant colony algorithm for classification rule discovery," in Data Mining: A Heuristic Approach, H.A. Abbas, R.A. Sarker, and C.S Newton, Eds, London, UK.: Idea Group Publishing, 2002.
[9] R S Parpinelli, H S Lopes And A A Freitas, "Data Mining with Ant Colony Optimization" IEEE Tran. Evol. Comput. vol. 6, no.4, pp321-332, Aug 2002.