“Plant Leaf Disease Detector And Solution Recommendation”

Abstract: Agricultural productivity and food security are significantly affected by plant diseases, which cause substantial economic losses and threaten global food supply chains. Traditional methods of disease detection rely heavily on manual inspection by experts, which are time-consuming, subjective, and often infeasible for large-scale monitoring. This project presents a mobile application designed for plant leaf disease detection and solution recommendation, with the key advantage of functioning in offline mode. This research proposes an intelligent, automated system for plant leaf disease detection and solution recommendation using computer vision and machine learning techniques.

High-resolution images of plant leaves are preprocessed using contrast enhancement and noise reduction techniques. A Convolutional Neural Network (CNN) model, trained on datasets such as PlantVillage, is used to extract features and classify diseases including bacterial blight, powdery mildew, and leaf spot. The model achieves an average accuracy exceeding 85%.

After detection, the system provides solution recommendations including chemical, biological, and cultural practices. The proposed system supports farmers through a mobile-based interface and promotes sustainable agriculture. Future work includes integrating environmental factors like humidity and temperature for better predictions.

Keywords: Mobile Application, Convolutional Neural Network, Image Processing, Crop Disease Classification.