A Study on Species Identification Based on Leaf Contours of Taiwan Lauraceae and Fagaceae Plants

Abstract

In this paper, a leaf species identification platform for Taiwan Lauraceae and Fagaceae plants is developed by using a variety of morphological features of leaf shape in combination with fuzzy theory and template matching technology. Firstly, the binary leaf contour is extracted by normalized sampling of leaf length and width through image preprocessing, and then the special geometric features of leaves, such as morphological convex hull, centroid-contour distance and serrated shape segmentation, are extracted respectively. Finally, the sample feature trainings and template comparison are carried out by fuzzy theory to judge the species identification of leaves. In this study, 54 species of mixed Lauraceae and Fagaceae were used to analyze the effect of leaf identification by the well-known algorithm k-NN and a method proposed in this paper.