Machine learning-based detection and classification of leaf diseases in rooftop gardens

Urban Farm DB summary

This research proposed a plant species recognition system based on deep learning technology to guide plant selection and management in rooftop gardening and urban agriculture. The system comprises four steps: image acquisition, preprocessing, feature extraction, and classification. The model was trained on a dataset of 80,000 images of 14 plant species, utilizing transfer learning on a ResNet9 architecture. The model achieved a score of approximately 92 out of 100.