Development of an Automatic Feeding Machine for Aquatic Animals Considering Key Factors Affecting Proper Feeding

Abstract

This research focuses on developing an automatic feeder that optimizes aquatic animal feeding by incorporating key environmental factors. Using image processing technology and IoT connectivity, the system ensures that aquatic animals receive the appropriate amount of food while monitoring their behavior. The system accounts for critical environmental factors, including water temperature, oxygen levels, turbidity, pH, and light, all of which influence feeding behavior. The system development is divided into three parts, with the web application serving as a tool to manage the operation of the automatic feeder for aquatic animals based on key feeding factors and to access various data. The image processing component utilizes real-time object detection modeling technology to identify food in the pond. The automatic feeder is designed to accommodate key feeding factors and is adaptable for use in various environments.

          The experimental study of the automatic aquatic animal feeder, which operates based on key factors affecting food intake and is controlled via a web application, the system successfully dispenses food automatically according to the scheduled time and user-specified environmental conditions. Water quality is monitored through sensors, and the system tracks food intake by detecting leftover food in the pond. The image processing method achieved an accuracy of 66.09%, with detection errors primarily caused by uncontrollable light variations, which make air bubbles appear similar to food pellets. Additionally, users can monitor various data and the machine's operating status through the web application.