computer Vision for Plant Identification

Abstract

Regarding to problems in plant identification, Botanist need a tool to assist them infinding the plant taxonomy. There are two goals of plant taxonomy are identification and classification. Computer vision (image understanding) is a discipline that studies how to reconstruct, interpret and understand a 3D scene from its 2D images in term of the properties of the structures present ill the scene. We have extensively conducted some research to develop computer vision/or plant identification based on image especially for Indonesian medicinal plants and house plant. 11/ the experiment, Ire used three features .for plant identification i.e. morphology, shape and texture. Also, we have conducted some research to classify the plant. There are several classifier algorithms that we used, i.e., Probabilistic Neural Network, Radial Basis Function Neural Network. and Support VectorMachine. Tohelp userfinding information easily about the plants, we also conducted research to develop search engine system based 011 article and research documents. The experiments show that our :,ystel1lsare vel)' promising. In thefuture we need to explore more extensively mathematical model asfunction to 1I10deithe plant in order to improve the accuracy inplant identification and classification.