Plant Identification based on Fractal Refinement Technique (FRT)

We propose here a new algorithm for plant classification and identification based on fractal dimension. It is a simple and efficient technique for identifying plants using three levels of fractal refinement on leaf images. Contour, Contour-Nervure and Nervure fractal dimensions are computed and are used in the first, second and third level of refinement respectively. A 50 set species with each set containing 10 samples are used for training the algorithm. The performance of the algorithm was examined with a test set of 500 leaves arbitrarily selected from different groups of species. The fault acceptance rate (FAR), the fault rejection rate (FRR) and the classification accuracy of the algorithm were analyzed experimentally and demonstrated that the proposed method has an accuracy rate of 84%.