Abstract
Automatic tree crown segmentation from remote sensing data is especially challenging in dense, diverse, and multilayered tropical forest canopies, and tracking mortality by this approach is even more difficult. Here, we examine the potential for combining airborne laser scanning (ALS) with multispectral and hyperspectral data to improve the accuracy of tree crown segmentation at a study site in French Guiana. We combined an ALS point cloud clustering method with a spectral deep learning model to achieve 83% accuracy at recognizing manually segmented reference crowns (with congruence >0.5). This method outperformed a two-step process that involved clustering the ALS point cloud and then using the logistic regression of hyperspectral distances to correct oversegmentation. We used this approach to map tree mortality from repeat surveys and show that the number of crowns identified in the first that intersected with height loss clusters was a good estimator of the number of dead trees in these areas. Our results demonstrate that multisensor data fusion improves the automatic segmentation of individual tree crowns and presents a promising avenue to study forest demography with repeated remote sensing acquisitions.
Highlights
INTRODUCTIONA IRBORNE laser scanning (ALS) is a powerful technology for mapping forest biomass and tracking forest dynamics
A IRBORNE laser scanning (ALS) is a powerful technology for mapping forest biomass and tracking forest dynamicsManuscript received November 26, 2020; revised March 15, 2021; accepted March 19, 2021
We develop a method for complementing airborne laser scanning (ALS) with RGB and hyperspectral imagery to improve individual tree crown (ITC) segmentation
Summary
A IRBORNE laser scanning (ALS) is a powerful technology for mapping forest biomass and tracking forest dynamics. More complex individual tree crown (ITC) approaches seek to recognize ITC within ALS point clouds, predict the biomass of these trees using allometric functions, and by summation, calculate biomass per unit area [6], [19], [20] These ITC approaches are a little better than area-based approaches at mapping forest biomass, but their true value lies in tracking tree-level responses to environmental stressors, such as drought events [20], [21] and disease [22]. We work with the mean-shift algorithm—amongst the most effective approach—currently available for segmenting ALS point clouds of tropical rainforests [23] This algorithm draws polygons around each predicted tree crown; we merge neighboring spectrally similar segments to reduce oversegmentation. We demonstrate how this approach can be used to track individual tree mortality over time
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 callMore From: IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Disclaimer: 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.
