MACHINE LEARNING AND DEEP LEARNING: A COMPARATIVE ANALYSIS FOR APPLE LEAF DISEASE DETECTION

Variations in the visual characteristics of leaf diameters allow for the differentiation of ill states, making leaves valuable indicators for the diagnosis of sickness. Accurate disease diagnosis depends on identifying the distinctive patterns that illnesses leave on foliage. Specialists or cultivators have frequently performed plant inspections, which may be costly and time-consuming. Automation of disease diagnosis is therefore crucial, particularly in areas with limited access to specialists. This work employs five classification algorithms Inception V3, Decision Tree, Support Vector Machine (SVM), and Random Forest to create a model for detecting diseases on apple leaves. The study's prime focus is Apple Rust, Apple Spot, and Apple Scab. To detect these illnesses, a relative examination of machine learning and deep learning models is carried out using the "Apple Leaves Disease Dataset."Among all the models tested, VGG19 achieved the highest test accuracy, reaching an impressive 95 percent.