Abstract—Selection of leaf features that are appropriate for identification is very important. Local Binary Patterns (LBP) is one of texture feature that is efficiency and robustness for plant identification. Meanwhile, in LBP we have to define good size of sampling point. In this research we propose fusion of LBP features, which incorporates different size of sampling point. There are two ways for fusion of LBP. First, we perform a straightforward fusion by calculating histogram of multiple LBP features separately using varying the size of sampling points and radius, then concatenate the multiple histograms together. Second, each histogram of LBP features is classified, and the feature fusion can be accomplished by classifier combination. For leaf classification, we used probabilistic neural network (PNN) to classify LBP features. The experiment performed on tropical medicinal plants and house plants. According to experimental results, the fusion of LBP features can improve accuracy in plant identification. This system is very promising to help people identify medicinal plant automatically and for conservation and utilization of medicinal plants. I. INTRODUCTION NDONESIA has many plentiful and potential biodiversity of tropical medicinal plants for nation’s health development. In 2001 the Laboratory of Plant Conservation, Faculty of Forestry Bogor Agricultural University (IPB) has recorded 2039 tropical medicinal plant species from Indonesia forest ecosystems [1]. Unfortunately only 20-22% of medicinal plant is cultivated by people [2]. One of the conservation and utilization of medicinal plants using the technology is developing a medicinal plants identification system automatically [1]-[2]. Identification of house plants has been done by [3] using texture features with the method of Local Binary Patterns (LBP). Ojala et al. [4] combines LBP operators with a variety of sampling points and different radius to achieve a robust texture features. Herdiyeni et al. [5] combines morphology, texture and shape features to identify the leaf of plants using classifier combination. In this research, we use medicinal plants image by applying fusion of LBP features. The features also tested on house plants image which has complex background. Fusion of LBP features was done by fusion operators of LBP such as conducted by [4]. In addition, the fusion of LBP features also performed using classifier combination methods as conducted by [5]. In this research, we used Probabilistic Neural Network (PNN) as a classifier [11]. II. LOCAL BINARY PATTERNS Local Binary Patterns (LBP) proposed by [4] for rotation invariant texture classification. To obtain LBP value, thresholding performed on the neighborhood circular pixels using the central pixel, then multiply by binary weighting. As an example for the sampling points 8 P  and radius 1 R  , the calculation of LBP value is illustrated in Fig. 1. Fig. 1. Calculation of LBP. LBP can be formulated as:     ｦ      1 0 , 2 , P p p c p c c R P g g s y x LBP (1) ｯ ｮ ｭ  t  0 0 0 1 ) ( x x x s (2) where c x and c y are the coordinate of center pixel, p is circular sampling points, P is number of sampling points or neighborhood pixels placed on a circle of radius R , p g is gray scale value of p , c g is center pixel, and s or sign is threshold function. For Fusion of Local Binary Patterns Features for Tropical Medicinal Plants Identification Yeni Herdiyeni, Iyos Kusmana Department of Computer Science Faculty of Mathematics and Natural Sciences, Bogor Agricultural University (IPB) West Java, Indonesia Email: yeni.herdiyeni@ipb.ac.id I ICACSIS 2013 ISBN: 978-979-1421-19-5 353 /13/$13.00 ©2013 IEEE