Individual tree detection in large-scale urban environments using high-resolution multispectral imagery

• Published in: International journal of applied earth observation and geoinformation Vol. 130; p. 103848
• Main Authors: Ventura, Jonathan, Pawlak, Camille, Honsberger, Milo, Gonsalves, Cameron, Rice, Julian, Love, Natalie L.R., Han, Skyler, Nguyen, Viet, Sugano, Keilana, Doremus, Jacqueline, Fricker, G. Andrew, Yost, Jenn, Ritter, Matt
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.06.2024; Elsevier
• Subjects: Aerial imagery; Computer vision; Convolutional neural network; Deep learning; Multispectral imagery; Object detection; Tree detection; Urban forests; Aerial imagery; Deep learning; Computer vision; Tree detection; Convolutional neural network; Object detection; Multispectral imagery; Urban forests
• ISSN: 1569-8432
• DOI: 10.1016/j.jag.2024.103848

Systematic maps of urban forests are useful for regional planners and ecologists to understand the spatial distribution of trees in cities. However, manually-created urban forest inventories are expensive and time-consuming to create and typically do not provide coverage of private land. Toward the goal of automating urban forest inventory through machine learning techniques, we performed a comparative study of methods for automatically detecting and localizing trees in multispectral aerial imagery of urban environments, and introduce a novel method based on convolutional neural network regression. Our evaluation is supported by a new dataset of over 1,500 images and almost 100,000 tree annotations, covering eight cities, six climate zones, and three image capture years. Our method outperforms previous methods, achieving 73.6% precision and 73.3% recall when trained and tested in Southern California, and 76.5% precision 72.0% recall when trained and tested across the entire state. To demonstrate the scalability of the technique, we produced the first map of trees across the entire urban forest of California. The map we produced provides important data for the planning and management of California’s urban forest, and establishes a proven methodology for potentially producing similar maps nationally and globally in the future. •We perform a comparative study of tree detection methods in urban environments.•We introduce a new dataset of over 1,500 images and almost 100,000 tree annotations.•Our novel neural network regression approach outperformed previous methods.•We applied our method to produce a complete map of the urban forest of California.