KLASIFIKASI KESEGARAN IKAN TONGKOL BERDASARKAN CITRA MATA BERBASIS CONVOLUTIONAL NEURAL NETWORK (CNN)

Fitria Ningsih (Universitas Teknologi Yogyakarta)
Agus Suhendar (Universitas Teknologi Yogyakarta)

Scientific Journal of Informatics
Vol. 10 No. 2 (2025): Jurnal Ilmiah Informatika

It Fish freshness is a crucial factor in ensuring food quality and safety. However, the conventional assessment process still relies on human observation, which is subjective supporting system, the risk of distributing non-fresh fish to consumers remains high, potentially affecting public health and consumer trust in fishery products. To address this issue, a fish freshness classification system based on eye image analysis using the Convolutional Neural Network (CNN) method was developed. The system development stages include collecting fihs eye image data, labeling, image preprocessing, CNN model training, and implementing the system in an convolution and pooling layers to extract visual features from the images. The initial testing results show that the system can classify fish freshness into two categories, Fresh and Not Fresh, with a high level of accuracy. This system is expected ti assist the public and fishery industry practitioners in evaluating fish quality more accurately ang efficiencly.