Proposed CNN Model for the Classification of Plant Leaf Disease

Agriculture plays a vital role in global food security and economic stability. Early and accurate detection of plant diseases is essential for minimizing crop losses and ensuring agricultural sustainability. This study proposes a Convolutional Neural Network (CNN)-based model for plant leaf disease classification, utilizing a dataset of 4,502 images spanning 22 different plant disease categories. The developed CNN model automates the disease identification process, providing a scalable, cost-efficient, and highly precise alternative to traditional manual inspection methods. By leveraging deep learning, the model extracts distinctive features from leaf images to effectively differentiate between healthy and diseased samples. The dataset includes plant species such as Alstonia Scholaris, Arjun, Bael, Basil, Chinar, Guava, Jamun, Jatropha, Lemon, Mango, Pomegranate, and Pongamia Pinnata, ensuring a broad scope of application. The model architecture integrates convolutional layers, pooling layers, batch normalization, and dropout mechanisms to enhance generalization and minimize overfitting. It was trained for 15 epochs using the Adam optimizer, achieving a peak classification accuracy of 98.52%. Experimental results highlight a notable improvement in disease recognition, reinforcing the potential of this approach for intelligent agricultural monitoring and disease prevention. This study advances precision agriculture by incorporating deep learning for early disease detection, enabling proactive intervention and fostering sustainable farming practices.