PENERAPAN ALGORITMA K-NEAREST NEIGHBOR UNTUK KLASIFIKASI KEMATANGAN BUAH JERUK NIPIS

Fahmi, Yusril
Qomaruzzaman
Agustin, Soffiana

Jurnal TEFSIN ( Jurnal Teknik Informatika dan Sistem Informasi)
Vol. 3 No. 2 (2025): November 2025

Classification of fruit ripeness has become an important topic in agriculture, as this process often requires considerable time and effort. This study aims to develop an automatic classification system that can identify the ripeness level of lime (unripe, half-ripe, ripe) based on RGB and HSV color features using the K-Nearest Neighbor (KNN) algorithm. A total of 83 datasets were collected using a Poco M3 Pro 5G camera, followed by preprocessing, feature extraction, and classification with the KNN algorithm. Using 16 test data in classification, the highest accuracy achieved was 75% with k=5. The implementation of this method demonstrates that KNN is quite effective in classifying color features.