This research proposes a leaf identification system with features fusion of leafmorphology, convex hulls (shape features) and ð¿ðµðð,ð
ððð¢2 (texture features). Probabilistic Neural Network (PNN) is used as classifier. The experimental results of leaf identification system, average accuracy of combining all the features is 87.5%. Accuracy by combining three features higher than using morphological features (58.125%) or texture features (68.125%). In this research showed that the texture features more influence than morphological features for recognition of plants.
Copyrights © 2016