Plant Leaf Recognition using Neural Networks (LeafRApp)

Abstract

Plant Leaf Recognition using Neural Networks (LeafRApp) is a desktop application which recognizes a plant from an input image file using the plant leaf’s shape. A hybrid of two modelling techniques is used to extract features from the leaf. The Moment-Invariant method is used to extract the first four moments of the image while Centroid-Radii method is used to extract 36 radii with respect to the image’s centroid. An optional reduction feature is then used to reduce the list of matches of the unidentified plant species inputted by the user.