Disease Detection In Plant Leaves Using Deep Learning Models: AlexNet And GoogLeNet

In this work, we created a plant disease detection system with the help of deep learning technology that automatically detects the symptoms displayed on the leaves of the potato plant. The methodology evaluates symptoms present on the leaflet and anticipates the type of sickness encountered on the plant based on the classification stated in the dataset. The latest development of the deep learning based convolutional neural network (CNN) has vastly enhanced the accuracy of image classification. Driven by CNN's success in image classification, this paper is based on the deep learning models for detecting plant diseases like early blight and late blight. The study employs advanced CNN models like AlexNet and GoogLeNet. The AlexNet and GoogLeNet models were trained and tested on data sets collected from the kaggel website. Training, testing, and experimental results show that the proposed technique can realise and achieve the highest accuracy of 98.5185%.