Font Size: 

The longan fruit is widely available across Indonesia. There are various varieties of the Longan fruit, and they differ in terms of their morphology, including the traits of their leaves. Based on the form of the leaves, the cultivars of longan fruit are to be categorized in this study. To obtain data for various attributes utilized on each cultivar, the author uses the RGB color extraction feature, the Grey Level Co-occurrence Matrix (GLCM), and the Shape feature. To determine the accuracy value that will be utilized as a determinant of the classification of the Longan leaf image, the value is additionally processed using the Back-Propagation Neural Network (BPNN) algorithm. The algorithm relies on the metric and eccentricity parameters. The BPNN algorithm performed well in testing, achieving an accuracy of 70%, demonstrating its suitability for classifying different types of longan fruit leaves.

Longan; Image Processing; RGB; GLCM; Shape, Back-Propagation Neural Network