Developing AI-Based Solutions for Indonesian Freshwater Fish Breeding: The Case of Tilapia Seed Calculation
Tilapia (Oreochromis niloticus) is a widely consumed freshwater fish in Indonesia. Its cultivation is economically significant. Conventional methods for calculating fish seeds are time-consuming and prone to error. These issues present challenges for farmers. This study presents the development of an Android-based application for tilapia seed calculation. The application utilizes the YOLO (You Only Look Once) object detection algorithm. The dataset comprised 400 images, with 300 used for training and 100 for testing. Images were collected during the seeding process in Soppeng Regency. The trained YOLO model was converted to TFLite format and integrated into the application. Application functionality was evaluated using the Blackbox testing method. Results indicate that the application achieved 86% accuracy in tilapia seed calculation compared to manual data testing. The application offers a more efficient and accurate solution for tilapia farmers.
Authors:
M. Ashar Alhakim, Eddy Tungadi, Tantri Indrabulan, Andrea Stevens Karnyoto, Najia Al Umri, and Bens Pardamean.
ICCSCi 2026