POLIBITS, Vol. 62, pp. 13-19, 2020.
Abstract: Huge loss in crop production occurs every year due to late identification of pant diseases in developing countries like Nepal. Timely and correct identification of such diseases with less dependency in related field expert can be more effective solution to the problem. Plants suffer from various diseases and correctly identifying them by observing the leaves is major challenge especially if they have similar texture. Consideration of plant leaf color and various texture features is extremely important to correctly predict the defect in plant. The aim of this work is to classify and predict given disease for plant images using different machine learning models like Support Vector Machine(SVM), k-Nearest Neighbors (KNN), Random forest Classifier (RFC), Convolutional Neural Network and compare the results. Image features like contrast, correlation, entropy, inverse difference moments are extracted using Haralick texture features algorithm which are fed to SVM, KNN and Random Forest Algorithms whereas CNN directly feeds upon images as input. Among the used models CNN produced highest level of accuracy of 97.89% and RFC, SVM and KNN had accuracy of 87.43%, 78.61% and 76.96% respectively for sixteen different image categories used.
Keywords: Plant disease, haralick texture, support vector machine, k nearest neighbor, random forest classifier, convolutional neural network
PDF: Plant Leaf Disease Recognition Using Random Forest, KNN, SVM and CNN
PDF: Plant Leaf Disease Recognition Using Random Forest, KNN, SVM and CNN
https://doi.org/10.17562/PB-62-2
See table of contents of POLIBITS 62.