Real-Time Mobile Application for Apple Leaf Disease Detection and Classification Using AFD-NET

Agriculture serves as a vital livelihood source, offering extensive employment opportunities in developing countries like India, where around 70% of the population depends on it. Despite this, many Indian farmers employ manual cultivation methods due to limited technical knowledge, leading to uncertainty regarding crop suitability. Disease outbreaks affecting plant leaves significantly impact agricultural production and profitability, necessitating efficient disease detection techniques. Farmers currently rely on pesticide spraying, which poses risks to human health and economic sustainability. But the advancements in computer vision methods utilizing deep learning (DL) have opened up avenues for identifying and understanding these diseases at an early stage, directly on the leaves. The goal of this study is to tackle these issues by developing a mobile application that can predict apple leaf diseases, i.e., Alternaria, Leaf Spot, Marssonina Blotch, Powdery Mildew, Apple Scab and Apple Black Rot, in the early stages, facilitating timely intervention and improved crop management. AFD-Net model (modified efficient net-B0) is utilized for classifying apple leaf diseases. A dataset of 9,000 images is employed for training and testing of AFD-NET model.