Perbandingan Local Binary Pattern dan Fuzzy Local Binary Pattern untuk Ekstraksi Citra Tumbuhan Obat

Abstract

This research implements Fuzzy Local Binary Pattern (FLBP) method for plant images feature extraction. FLBP is used to handle uncertainty on images with various patterns. FLBP approach is based on the assumption that a local image neighbourhood may be characterized by more than a single binary pattern. In order to improve the speed on image searcing, this research used Probabilistic Neural Network (PNN). This approach was experimentally evaluated and compared with the original LBP on a dataset of medicinal plant for images non-background and house plant for images with background. The database contains 1440 medicinal plant leaf images and 300 tree images belonging to 30 different types and is obtained from Biofarmaka IPB, Cikabayan Farm, Green house Center Ex-Situ Conservation of Medicinal Plant Indonesia Tropical Forest and Gunung Leutik. Experimental results show that FLBP is superior to LBP based on the increased accuracy in medicinal plant identification (FLBP: 66.33% vs LBP: 34.46% ). It can be concluded that this approach is capable to identify medicinal plants species efficiently and accurately