Identification of leaf and plants is an area of research which has gained a lot of attention in these years and is also an important tool in the field of agriculture, crop rotation, cultivation, forestry and much more. The process generally begins with the acquisition of images i.e., enhancement of leaf images, segmentation of leaf, its feature extraction and the classification. Today, Classification of plants using its various categories has been a broad application. In this paper we presentdifferent techniques which can be used for plant leaves classification. The classification method includes some segmentation algorithms and pattern classification techniques. This technique helps in plant-leaf classification. This process and analysis is effective and the performance of the leaf classification system is analyzed using Radial Basis Function Neural Network (RBFNN). RBFNN enables non linear transformation followed by linear transformation to achieve a higher dimension in hidden space. RBFNN is trained and tested for various categories of leaf images using different Grey Level Co-Occurrence Matrix(GLCM) Features. The results show satisfactory performance and the highest accuracy of 93.04% is achieved using Gaussian Kernels.

Keywords : Acquisition; Segmentation Algorithms; GLCM Features; RBFNN;