Consolidated Dataset of 38 Studies on Urban Tree Detection, Segmentation, and Classification using Remote Sensing and Machine Learning (2014–2026)

This dataset consolidates into a single spreadsheet the data extracted from 38 peer-reviewed articles selected through a Systematic Literature Review (PRISMA protocol) on automated urban tree detection, segmentation, and classification using remote sensing and machine learning. Each row represents one study and integrates 14 columns: reference, country, objective, platform(s), sensor(s), primary algorithm, approach type, pre-processing, post-processing, evaluation metrics, and best reported performance. The articles span from 2014 to 2026, allowing direct cross-correlations between infrastructure, methodology, and accuracy. The dataset summarizes key trends (predominance of YOLO, RGB/LiDAR, and aerial platforms) while highlighting critical gaps, such as poor cross-city generalization and the lack of public benchmarks. It serves as a strategic resource for researchers planning experiments or conducting meta-analyses in automated urban tree inventory.