Classification of Healthy and Diseased Plant Leaves Using Deep Learning Model

In today's world, the combination of climate change, sustainable agriculture, and globalization highlights the vital need for plant disease preventive techniques. The connection between plant disease control and agriculture is discussed in this paper. It highlights how crucial early disease detection is becoming and how deep learning technology can be used to identify plant diseases. The methodology focuses on identifying plant species and detecting the health conditions of plants by utilizing deep learning techniques for plant disease classification. Moreover, The following procedures must be followed in order to use deep learning techniques to discriminate between healthy and sick plants: data collection, data preprocessing, model selection, model training, model assessment, and deployment. In this paper, the authors assess four primary metrics—accuracy, recall, precision, and F1-score—for three distinct models: ResNet50, MobileNetV3, and EfficientNet. In experimentation work, the authors have analyzed model performance and classification accuracy, with EfficientNetB5 proving to be the most accurate method having 98.52% accuracy. This categorization has great promise for transforming farming methods and guaranteeing food security for all people on the planet.