At IBM, I developed an AI-based crop disease detection system using ConvNeXt and Vision Transformer for image classification.
For AdaLeafNet, I developed a hybrid deep learning model for crop disease classification, combining ConvNeXt Tiny and Vision Transformer. It achieved 99.83% accuracy across 38 crop disease classes, with Grad-CAM++ and SHAP for explainable predictions.
I also built deep learning projects for lemon leaf and skin cancer disease detection. My work includes CNN-based classification, transfer learning, and a Streamlit interface for real-time prediction.
United States
United Kingdom