A Novel Approach to Fish Disease Diagnostic System based on Machine Learning

Abstract

Real-Time identification automated system diagnoses fish disease i.e. Epizootic Ulcerative syndrome (EUS) which is caused by Aphanomyces invadans , a fungal pathogen. In this paper we propose a Real-Time fish disease diagnose system with better accuracy. In order to improve the accuracy we propose a combination (PCA-FAST-NN) which combine the Principle component analysis (PCA) with Features from Accelerated Segment Test (FAST)feature detector using Machine Learning Algorithm(Neural Network) i.e. (PCA-FAST-NN) .The Experimentation has been done on the real images of Epizootic Ulcerative syndrome (EUS) infected fish database and implemented in MATLAB environment.