Automated Multiclass Classification of Groundnut Leaf Diseases Using Fine-Tuned InceptionV3 Model

In agriculture, the specific identification and differentiation of varieties of plant disease is a very important factor in preserving the health of subsistence crops and resulting in optimal crop yields. The mentioned research paper is about an original technique for the multiclass classification of groundnut disease using fine-tuned InceptionV3 model. Groundnut, a key crop for both nutrition and economic aspects, is vulnerable to multiple diseases such as Early Leaf Spot, Late Leaf Spot, Rust, and Groundnut Rosette Disease; each of which has an intense effect on crop yields. Using groundnut plant leaf images as a base for a comprehensive dataset, this study finetunes the InceptionV3 model, pre-fed on ImageNet, to perform more precise differentiations between healthy and diseased leaves. The methodology that has been used includes dataset preparation, model fine-tuning, and model training by using training and validation datasets and relying on standard metrics including accuracy, precision, recall, and F1 score. Data confirms the model's accuracy in disease categorization and that it is much more efficient than routine and manual approaches. This study pragmatically points out the prospects of deep learning in improving agricultural management systems and also encourages extensive research of different forms of AI for diagnostics applications.