Early-Stage Apple Leaf Disease Prediction Using Deep Learning

Abstract

Premature leaf fall, scab, Alternaria leaf spot, brown spot, mosaic, grey spot, and rust are common types of apple leaf diseases. Due to the arrival of monsoon, there is excess moisture in the air because of an outbreak of diseases in plants that is being witnessed in the hilly region. Farmers from these regions are always worried about the health of Apple plants. The scientists working in various departments, KrishiVigyanKendra's, and regional research stations have given the required inputs to control the problems but that is not useful to identify the problem in the early stage. Also, the current disease diagnosis based on human scouting is time-consuming and expensive. Our proposed system identifies various apple leaf diseases in an early stage that will alert the farmers and nearby research institutes to take appropriate action to control it. The dataset contains 1821 images of apple leaves which has normal leaves, scab, rust, and other disease infected leaves. The proposed regional convolutional neural network-based approach is capable of localizing and classifying the disease with 90% accuracy.