Detection of Soybean Pest and Disease through Machine Learning Techniques
plant disease, SVM, KNN, Random Forest, Machine learning
Plants diseases have been a major setback in agriculture, causing the plants to produce lesser thereby reducing agricultural productivity.Autonomous detection of diseases and pathogens has helped to significantly improve plant productivity. These different automatic detection methods has achieved accuracies bellow the 89% and also very few has been accessible techniques and data-sets to the end users. This study aims to compare the results of different machine learning models used for image classification, then deploy the model with the best accuracy, precision, AUC score and recall in detecting the presence of a caterpillar pest and Frog-eye leaf spot disease in soybeans. Neo 2017 has compared the performances of SNM, KNN and Random Forest for classifying land cover imagery and SVM was found to have the highest classification accuracy. Also compared to
Jad hav 2019 soybean leaf diseases were detected using multi-class SVM classifier and K-NN, in which the SVM gave an accuracy 88.38% and K-NN gave 84.64%. Thus, we pretend to compare the performance of these main classifiers on plant leaf image and projecting the one with the highest accuracy, working the main features for a better plant disease classification. In this work, we will analyze the performance metrics of accuracy, recall, precision, and F-measure with the aim of comparison to the related works.