AI-driven grape leaf disease detection using optimal image extraction local binary pattern (LBP)

The agriculture sector has demonstrated growing interest in using machine learning to detect and classify crop diseases.Viticulturists and grape growers confront many problems, but grape leaf disease identification and management are critical to crop productivity and quality.Machine learning, mostly in computer vision and pattern recognition, has opened new avenues for sickness of grape leaves diagnosis and classification.By offering a quick and correct finding method based on bonded bottomless structures and a lightweight Deep Convolutional Neural Network (DCNN) model for finding high-level representations of hidden data, this study determined how to identify illnesses in grape leaves.By merging the bottomless structures with the traditionally derived Local Binary Pattern (LBP) features, the local texture information is recovered from plant leaf pictures.The photos were initially resized to satisfy the DCNN network's input requirements.The deep learning from the input photos was extracted using the DCNN network's customized bottomless structures layer.The significant categorization feature information was then obtained by fusing two different kinds of bottomless structures from separate networks together utilizing a variety of fusion approaches.This research afterward recognized a classifier using the combined bottomless structures.On Kaggle datasets (Grape Leaf), the recommended model is trained and assessed.With an accuracy of 97.76% in diagnosing grape and apple leaf diseases, the recommended DCNN-LBP provides outstanding consequences.The materials of the experiments mention that the technique presented in this work may classify grape leaf diseases rapidly and exactly while also satisfying the demands of actual yield from agriculture.