UITI: Urban Individual Tree Inventory Database

This study constructed a joint computing framework for individual trees by integrating convolutional neural networks and the state-of-the-art Large Model for the first time. It can delineate individual tree crowns and heights. Based on the well-trained model, the first national-scale urban individual tree height database, Urban Individual Tree Inventory (UITI), has been successfully constructed. This database covers over 230 cities in China, providing crown area and height information for more than 80 million trees. It allows for precise tree counting in different cities and the calculation of urban three-dimensional green volume, offering unprecedented data support for the comprehensive inventory and fine-grained management of urban tree resources.