ProgramsView all programs
From the news
- Chancellor Amma Addresses the Parliament of World’s Religions
- Amrita Students Qualify for the European Mars Rover Challenge
Publication Type : Journal Article
Publisher : Journal of Intelligent and Fuzzy Systems, IOS Press
Source : Journal of Intelligent and Fuzzy Systems, IOS Press, Volume 38, Issue 5, p.6311-6323 (2020)
Keywords : Computer vision, fuzzy system, regions of interest, Saliency, Visual Attention
Campus : Bengaluru, Coimbatore
School : Department of Computer Science and Engineering, School of Engineering
Department : Computer Science
Year : 2020
Abstract : Top–down influences play a major role in the primate’s visual attention mechanism. Design of top-down influences for target search problems is the recommended approach to develop better computational models. Existing top down computational visual attention models mainly exploit three factors namely the context information, target information and task demands. Here in this paper we propose a Fuzzy based System for Target Search (FSTS) which makes use of target information as the top-down factor. The system uses Fuzzy logic to predict the salient locations in an image based on the prior information about a target object to be detected in a scene or frame. The performance of the system was analysed using multiple evaluation parameters and is found to have a better average hit number, number of first hits and elapsed CPU time than the existing system. The saliency map comparison is performed with human eye fixation map and is found to predict the human fixations with better accuracy than existing systems.
Cite this Research Publication : Amudha J., K.V, D., and Aarthi, R., “A fuzzy based system for target search using top-down visual attention”, Journal of Intelligent and Fuzzy Systems, vol. 38, no. 5, pp. 6311-6323, 2020.