Your Cart

Your cart is empty

Home  >  Transactions of NAMP VOL 7

10. PLANT LEAVES CLASSIFICATION USING K-NEAREST NEIGHBOR ALGORITHM by Usman M. A., Olayiwola M.O., Folorunsho S.O., Solanke O. O., Hammed F.A.,Okusaga S.T. and Oyebo, A.B. Transaction Volume 7, (March, 2018), pp63 – 70
Sale price: $5.00
Quantity:

PLANT LEAVES CLASSIFICATION USING K-NEAREST NEIGHBOR ALGORITHM

Usman M. A., Olayiwola M.O., Folorunsho S.O., Solanke O. O., Hammed F.A., Okusaga S.T. and Oyebo, A.B.

Department of Mathematical Sciences, Olabisi Onabanjo University, Ago-Iwoye, Ogun State.

Department of Mathematical Sciences, Osun State University, Osogbo.

Abstract

This paper investigates plant leaf classification using k-Nearest Neighbor Algorithm. Plant classification is important for agriculturist, botanist and medicine, and is especially significant to the biology diversity research. As plants are vitally important for environmental protection, it is more important to identify and classify them accurately. This study is aimed at building a machine learning model to classify plant species using extracted features (shape, margin and texture) of plants leaves. This project applied the k-Nearest Neighbor (k-NN) model to automatically classify and predict plant species. k-Nearest Neighbor model was used as a base learner and also as a bagged ensemble with varying values of k = 3, 5, 7, 9, 11, 13, 17, 19, 21. The classification performance of both k-NN and its ensemble model were compared based on accuracy and log-loss values. It was observed from the results obtained that the k-NN with k=3 outperformed all other models with the accuracy value of 88.89% and the log-loss value of 1.17. It is also observed that as the value of k is increasing the accuracy and Log Loss of the model is decreasing.

Keywords: Plant classification, k-NN, Machine learning, log-loss, plant leaf, Ensemble, Bagging

Close
Loading...