Artificial Neural Network for Determining the Freshness of Nile Tilapia Based on Gill Images Using the LVQ (Learning Vector Quantization) Method

Abstract

Determining the freshness of Nile tilapia is essential in the fisheries industry to ensure quality and safety. This study develops an Artificial Neural Network (ANN) system to classify Nile tilapia freshness based on gill images using the Learning Vector Quantization (LVQ) method, known for its effective data clustering. The process involves collecting fresh and non-fresh gill images, extracting features like color, texture, and shape, and training the ANN with the LVQ algorithm. The system achieved an accuracy of 86.67% and an F1 Score of 88.74%, demonstrating its effectiveness in identifying freshness patterns. This approach provides a reliable solution for assessing fish quality, reducing the risk of consuming non-fresh fish, and contributes to ANN-based advancements in the fisheries sector