Abstract
Individual tree crown detection and morphological parameter estimation can be used to quantify the social, ecological, and landscape value of urban trees, which play increasingly important roles in densely built cities. In this study, a novel architecture based on deep learning was developed to automatically detect tree crowns and estimate crown sizes and tree heights from a set of red‐green‐blue (RGB) images. The feasibility of the architecture was verified based on high‐resolution unmanned aerial vehicle (UAV) images using a neural network called FPN‐Faster R‐CNN, which is a unified network combining a feature pyramid network (FPN) and a faster region‐based convolutional neural network (Faster R‐CNN). Among more than 400 tree crowns, including 213 crowns of Ginkgo biloba, in 7 complex test scenes, 174 ginkgo tree crowns were correctly identified, yielding a recall level of 0.82. The precision and F‐score were 0.96 and 0.88, respectively. The mean absolute error (MAE) and mean absolute percentage error (MAPE) of crown width estimation were 0.37 m and 8.71%, respectively. The MAE and MAPE of tree height estimation were 0.68 m and 7.33%, respectively. The results showed that the architecture is practical and can be applied to many complex urban scenes to meet the needs of urban green space inventory management.
Highlights
Urban trees play important roles in densely built cities, with activities that include reducing atmospheric carbon dioxide, alleviating the urban heat island effect [1, 2], isolating noise [3], alleviating urban flood risk [4], and providing shelters for wildlife [5, 6]
Individual tree crown detection (ITCD) technology has traditionally been used in forest monitoring and consists of 2 phases: (1) locating and delineating individual tree crowns and (2) classifying tree species and estimating morphological parameters such as crown size, tree height, and diameter at breast height (DBH) ([8])
Considering the lack of an integrated architecture for ITCD with practical capabilities, this study proposes an automated urban canopy detection architecture based on deep learning that can obtain the number, location, crown size, and tree height of a given tree species from a set of RGB images
Summary
Urban trees play important roles in densely built cities, with activities that include reducing atmospheric carbon dioxide, alleviating the urban heat island effect [1, 2], isolating noise [3], alleviating urban flood risk [4], and providing shelters for wildlife [5, 6]. Detailed data on urban trees, such as species, location, number, diameter at breast height (DBH), tree height, and crown size, are essential for quantifying these benefits. Individual tree crown detection (ITCD) technology based on remote sensing, which has the advantage of providing spatially explicit data, potentially with fine temporal resolution and low cost [6], can facilitate urban green space inventory development and monitoring. ITCD technology has traditionally been used in forest monitoring and consists of 2 phases: (1) locating and delineating individual tree crowns and (2) classifying tree species and estimating morphological parameters such as crown size, tree height, and DBH ([8]). In the location and delineation phase, the data source and computational methods are the two major factors dominating the results. Crown delineation has been carried out based on multispectral imagery involving wavelength bands crucial for the identification of vegetation characteristics [16]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
