I-PEDIA: Aplikasi Mobile untuk Identifikasi Penyakit Tanaman Padi dengan Fuzzy Entropy dan Probabilistic Neural Network

Abstract

This research developed a mobile application for paddy plant disease identification system using fuzzy entropy and classifier probabilistic neural network (PNN) that runs on mobile. Paddy plant diseases is one of the problems on increasing paddy production. Fuzzy entropy is used for image feature extraction. PNN is used because training PNN is fast. Training data and testing data divided by k-fold cross validation with a k value of 5. Before training data and testing data are extracted, the images are enhanced using the Laplacian filter. The accuracy of paddy diseases identification without the Laplacian filter is 91.46% while accuracy of paddy diseases identification using the Laplacian filter is 90.55%.