An Efficient Fragmented Plant Leaf Classification Using Color Edge Directivity Descriptor

Plant species identification is one of the most important research branches of botanical science. The current work proposes an efficient methodology for recognition of plant species from whole as well as fragmented digital leaf images. The situation becomes challenging when only a partial portion of the leaf can be obtained. Since leaves are fragile and prone to be fragmentation due to various environmental and biological factors, the paper studies how recognition of fragmented leaves can be effectively done. In this study the combination of texture and color based method (Color Edge Directivity Descriptor) is used to extract the feature and Euclidean distance is used as the classifier for the classification of the fragmented as well as whole leaf images.