Year: 2024
Type: article
Abstract: The purpose of this study was to suggest a way for segmenting orchard trees from aerial images using several techniques. The goal was to automatically identify and recognize the orchard tree canopy in a variety of settings, including varying seasons, tree ages, and weed cover levels. Three distinct walnut orchards' worth of photos made up the implemented dataset. The dataset's attained variety led to the acquisition of photos that fit various use cases. Accuracy for training, validation, and testing were 91%, 90%, and 87%, respectively, for the best-trained model. To address problems with out-of-the-field boundary transparent pixels from the image, the trained model was additionally evaluated on previously unseen orthomosaic images of orchards using two techniques (oversampling and undersampling).
Institution: D. Y. Patil Agriculture and Technical University
Language: English
Topic: Smart Agriculture and AI
Subfield: Plant Science
Domain: Life Sciences
Open Access status: diamond
OpenAlex ID: W4405642444