Early Screening Protozoan White Spot Fish Disease using Convolutional Neural Network

Authors

  • Amiera Syazlin Binti Md Azhar Data Science Research Lab, School of Computing, Universiti Utara Malaysia, 06010 Sintok, Kedah, Malaysia
  • Nor Hazlyna Binti Harun Data Science Research Lab, School of Computing, Universiti Utara Malaysia, 06010 Sintok, Kedah, Malaysia
  • Mohamad Ghozali Bin Hassan School of Technology Management and Logistics, Universiti Utara Malaysia, 06010 Sintok, Kedah, Malaysia
  • Nooraini Binti Yusoff Department of Data Science, Universiti Malaysia Kelantan, 16100 Kota Bharu, Kelantan, Malaysia
  • Siti Naquiah Binti Md Pauzi School of Technology Management and Logistics, Universiti Utara Malaysia, 06010 Sintok, Kedah, Malaysia
  • Nurul Nadiah Yusuf Data Science Research Lab, School of Computing, Universiti Utara Malaysia, 06010 Sintok, Kedah, Malaysia
  • Kua Beng Chu National Fish Health Research Centre, Fisheries Research Institute, Department of Fisheries Malaysia, 11960 Batu Maung, Penang, Malaysia

DOI:

https://doi.org/10.37934/araset.37.1.4955

Keywords:

Aquaculture, Protozoan white spot disease, Underwater image, Disease screening, Convolutional neural network

Abstract

Aquaculture is in critical need of both intelligence and automation control in order to maintain a sustainable level of production. Historically, the accuracy of the disease diagnosis is determined by a person’s abilities, experiences and length of time spent. Due to the high level of expertise, time, and effort necessary to obtain an accurate diagnosis through manual inspection, inadequate early treatment could result in the rapid spread of the disease. As a result, there needs to be much focus on early-stage fish disease screening due to the rapid spread of infectious diseases in the vast fish system. This research focused specifically on Protozoan white spot disease, an infectious disease caused by Cryptocaryon irritans in saltwater considering the fact that the infection is contagious. Consequently, this research aims to create an intelligent system utilizing a convolutional neural network (CNN) algorithm, namely GoogleNet to detect infected fish based on raw underwater images taken. 90% accuracy achieved showed that the innovation could ease the process of fish disease screening. This effort could be a contributor to the aquaculture industry since humans rely on fish for survival in modern times for fisheries and livestock.

Downloads

Download data is not yet available.

Downloads

Published

2024-01-09

How to Cite

Amiera Syazlin Binti Md Azhar, Nor Hazlyna Binti Harun, Mohamad Ghozali Bin Hassan, Nooraini Binti Yusoff, Siti Naquiah Binti Md Pauzi, Nurul Nadiah Yusuf, & Kua Beng Chu. (2024). Early Screening Protozoan White Spot Fish Disease using Convolutional Neural Network . Journal of Advanced Research in Applied Sciences and Engineering Technology, 37(1), 49–55. https://doi.org/10.37934/araset.37.1.4955

Issue

Section

Articles