AI-Driven Crop Health Monitoring: A Deep Learning Framework for Automated Plant Leaf Disease Detection

Plant Diseases have also a great impact on the productivity of crops, so that, for sustainable agriculture, early and accurate detection of these diseases is an essential. In this research, a deep learning network for detecting leaf diseases of plants in smart agriculture, based on Convolutional Neural Networks (CNN) and AlexNet architecture, is presented. In the proposed framework, image processing, feature extraction and classification techniques are used for detection of various plant leaf diseases using digital images. CNN learns discriminative features automatically from the images of leaves, while the deep layered structure of AlexNet and optimized learning process enhance its classification performance. The plant diseases dataset used for the experiments are public domain and include healthy and diseased leaves for experimental evaluation. The classification accuracy proved to result high classification accuracy of 98.4%, precision of 97.9%, recall of 98.1% and F1 score of 98.0% with more than good classification accuracy of several traditional machine learning approaches. The proposed model will also have the capability to diagnose the disease in real-time, enabling appropriate preventive measures taken by the farmers and ensuring agriculture sustainability, yield, and productivity.