A Novel Fish Optimization Algorithm for Offshore Disease Detection in Fishes using a Smartphone App

• Published in: 2022 International Conference on Automation, Computing and Renewable Systems (ICACRS) pp. 1 - 7
• Main Authors: Sujatha, K., Kalpalatha Reddy, T., Bhavani, N.P. G.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 13.12.2022
• Subjects: Artificial neural networks; Classification algorithms; Extreme Artificial Neural Networks; Filtering algorithms; Fish; Fish disease; Fish optimization Algorithm; Improved canny algorithm; Machine learning algorithms; Median filtering; Oceans; Smart phones
• DOI: 10.1109/ICACRS55517.2022.10029312

The growth of aquaculture is a source of high income in developing countries like India. Fish liver oil is rich in nutrients and absorption of these nutrients provides high immunity to fight against various harmful diseases that are affecting the human community. Breeding, growing, and gathering fishes together are called as aquaculture. Fundamentally, it also a type of farming carried out on the sea waters which facilitate a hazard free ocean environment. Such a pollution free environment is accountable for serving as source of food and commercial products. Recognizing the infectious fishes is a exigent task on the beds of the ocean which demands sophisticated infrastructure. The increase in spread rate of the contiguous diseases among fishes can be detected at the early stage using the proposed smart phone App. This smart phone App is developed using robust machine learning algorithms associated with Extreme Artificial Neural Networks (EANN) to detect the fish disease at early stage. This paper proposes a novel hybrid algorithm; called the Fish Optimization Algorithm (FOA) with the conventional Back Propagation Algorithm (BPA) called the EANN, for solving the challenges related to global search. In this paper, an effort is made to develop a EANN based diagnostic model for early detection of fish diseases in the sea water. To substantiate the superiority in performance for the proposed EANN architecture, the results are compared with the conventional ANN model trained with BPA and therefore the robust model is used for development of a smart phone App.