This paper described an implementation of plant identification using neural networks. As preliminary research, three kinds of plants, Cissus discolor, Piper crocatum, and Zamio zamioculcas, had been explored using PNN and backpropagation. Features such as aspect ratio, invariant moments, and color averages were incorporated. The result showed that any improvements should be accomplished by adding features such as textural features and color features
                        
                        
                        
                        
                            
                                Copyrights © 2011