Bacterial Disease Detection for Pepper Plant by Utilizing Deep Features Acquiredfrom DarkNet-19 CNN Model

Abstract

In recent years, computer-aided agriculture applications have been developing rapidly as a prominentresearch area. In parallel with the developments in technology, the use of automatic systems, sensor fusion,the internet of things, and artificial intelligence-based systems is becoming widespread in agriculture. Theuse of these systems allows for safer, faster, and more cost-effective operations based on human factors inagricultural applications. Among these applications, there are artificial intelligence applications developedbased on image processing and machine learning. Plant disease detection systems are also among theseartificial intelligence studies. Within the scope of this study: I. It has been ensured that the leaf images ofthe pepper plant have been segmented and their features have been extracted from the pre-trainedconvolutional neural network. II. These obtained features have been classified through the classifiermethods in order to detect bacterial disease. In the study, a total of 2475 images of pepper leaves with1478 healthy and 997 bacterial diseases, which are among the PlantVillage data sets, have been used. Toextract the features, the DarkNet-19 network model has been used as a pre-trained convolutional network.The SoftMax classifier in the last layer of the convolutional network model has been removed from thenetwork and SVM, KNN, and Decision-Tree-based classifiers are used instead of it. According to theresults, the level of performance achieved using the DarkNet-19 network and SVM classifier is quitesatisfactory.