Diagnosis of Fish Disease using UKF and Elman Neural Networks
Gujjala Jhansi1, K.Sujatha2, R.S. Ponmagal3, M. Anand4, V.Srividhya5
1Gujjala Jhansi, Research Scholar, Department of ECE, Dr. MGR Educational and Research Institute, Chennai (Tamil Nadu), India.
2K.Sujatha, Professor, Department of EEE, Dr. MGR Educational and Research Institute, Chennai (Tamil Nadu), India.
3R.S. Ponmagal, Professor, Department of CSE, SRM Institute of Science and Technology, Chennai (Tamil Nadu), India.
4M.Anand, Professor, Department of ECE, Dr. MGR Educational and Research Institute, Chennai (Tamil Nadu), India.
5V.Srividhya, Assistant Professor, Department of EEE, Meenakshi College of Engineering, Chennai (Tamil Nadu), India.
Manuscript received on 25 May 2019 | Revised Manuscript received on 03 June 2019 | Manuscript Published on 22 June 2019 | PP: 438-441 | Volume-8 Issue-3S, February 2019 | Retrieval Number: C10920283S19/19©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: At early stages, identification of fish disease is very difficult because it can prevent from spreading disease underwater. The identification of fish disease is a manual process so far. Hence with a vision of contribution to aquaculture, this new scheme to categorize and detect (Epizootic Ulcerative Syndrome) EUS infected and non fishes are proposed here. The need for an image based automated process arises because the manual process of identification is tedious. This work depends on the prior data base obtained from information fusion study of integrated navigation with GPS/INS. To deal with the uncertainty of error covariance and noise in diagnose the fish disease, the article propose a novel fish disease Identification approach where the Unscented Kalman Filter (UKF) with different covariance. The characteristics like GLCM are pulled out for classification of Non-EUS and EUS affected fish by Algorithms pertaining to Machine Learning to obtain classification accuracy with the help of Elman Neural Network (ENN). This testing was done by MATLAB simulation software with real time database containing images of EUS affected fish.
Keywords: Fish Disease, Unscented Kalman Filter, Feed Forward Neural Network And Elman Neural Network.
Scope of the Article: Neural Information Processing