Algoritma K-Nearest Neighbors (KNN) untuk Klasifikasi Citra Buah Pisang dengan Ekstraksi Ciri Gray Level Co-Occurrence

Ilhamy, Rifqi Syahrul
Sanjaya, Ucta Pradema

Jurnal Telematika
Vol. 17 No. 2 (2022)

Banana is one type of fruit that is rich in benefits. Bananas have a soft flesh texture. There are various sizes of bananas based on the type. The color and shape of bananas differentiate one type of banana from another. This research recognizes and classifies bananas based on their skin color by using digital image processing. The gray level co-occurrence matrix (GLCM) feature is an extraction technique commonly used to find features in an image. The classification technique in this study uses the k-nearest neighbors (KNN) algorithm. This study obtained the best accuracy of 76.6% at an angle of 0 degrees with a value of k = 1, 90% accuracy at an angle of 45 degrees with a value of k = 1, 86% accuracy at an angle of 90 degrees with a value of k = 1, and 73% accuracy at an angle of 135 degrees with a value of k = 5, 7, 9.