Multisensor Data Fusion for Improved Segmentation of Individual Tree Crowns in Dense Tropical Forests
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 multispec...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2021-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9387530/ |
| _version_ | 1818573995295899648 |
|---|---|
| author | Melaine Aubry-Kientz Anthony Laybros Ben Weinstein James Ball Toby Jackson David Coomes Gregoire Vincent |
| author_facet | Melaine Aubry-Kientz Anthony Laybros Ben Weinstein James Ball Toby Jackson David Coomes Gregoire Vincent |
| author_sort | Melaine Aubry-Kientz |
| collection | DOAJ |
| description | 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. |
| first_indexed | 2024-12-15T00:20:31Z |
| format | Article |
| id | doaj.art-d64d54fa2f0544dba4d3794bc754cdf8 |
| institution | Directory Open Access Journal |
| issn | 2151-1535 |
| language | English |
| last_indexed | 2024-12-15T00:20:31Z |
| publishDate | 2021-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj.art-d64d54fa2f0544dba4d3794bc754cdf82022-12-21T22:42:20ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing2151-15352021-01-01143927393610.1109/JSTARS.2021.30691599387530Multisensor Data Fusion for Improved Segmentation of Individual Tree Crowns in Dense Tropical ForestsMelaine Aubry-Kientz0https://orcid.org/0000-0002-4439-8626Anthony Laybros1https://orcid.org/0000-0002-5111-5225Ben Weinstein2James Ball3Toby Jackson4David Coomes5https://orcid.org/0000-0002-8261-2582Gregoire Vincent6AgroParisTech, UMR EcoFoG (CNRS, Cirad, INRAE, Université des Antilles, Université de la Guyane), Kourou, French GuianaAMAP, University of Montpellier, CIRAD, CNRS, INRAE, IRD, Montpellier, FranceDepartment of Wildlife Ecology and Conservation, University of Florida, Gainesville, FL, USADepartment of Plant Sciences, University of Cambridge, Cambridge, U.K.Department of Plant Sciences, University of Cambridge, Cambridge, U.K.University of Cambridge Conservation Research Institute, Cambridge, U.K.AMAP, University of Montpellier, CIRAD, CNRS, INRAE, IRD, Montpellier, FranceAutomatic 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.https://ieeexplore.ieee.org/document/9387530/Airborne laser scanning (ALS)data fusiondeepforesthigh-resolution imageryhyperspectral3-D adaptive mean-shift (AMS3D) |
| spellingShingle | Melaine Aubry-Kientz Anthony Laybros Ben Weinstein James Ball Toby Jackson David Coomes Gregoire Vincent Multisensor Data Fusion for Improved Segmentation of Individual Tree Crowns in Dense Tropical Forests IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Airborne laser scanning (ALS) data fusion deepforest high-resolution imagery hyperspectral 3-D adaptive mean-shift (AMS3D) |
| title | Multisensor Data Fusion for Improved Segmentation of Individual Tree Crowns in Dense Tropical Forests |
| title_full | Multisensor Data Fusion for Improved Segmentation of Individual Tree Crowns in Dense Tropical Forests |
| title_fullStr | Multisensor Data Fusion for Improved Segmentation of Individual Tree Crowns in Dense Tropical Forests |
| title_full_unstemmed | Multisensor Data Fusion for Improved Segmentation of Individual Tree Crowns in Dense Tropical Forests |
| title_short | Multisensor Data Fusion for Improved Segmentation of Individual Tree Crowns in Dense Tropical Forests |
| title_sort | multisensor data fusion for improved segmentation of individual tree crowns in dense tropical forests |
| topic | Airborne laser scanning (ALS) data fusion deepforest high-resolution imagery hyperspectral 3-D adaptive mean-shift (AMS3D) |
| url | https://ieeexplore.ieee.org/document/9387530/ |
| work_keys_str_mv | AT melaineaubrykientz multisensordatafusionforimprovedsegmentationofindividualtreecrownsindensetropicalforests AT anthonylaybros multisensordatafusionforimprovedsegmentationofindividualtreecrownsindensetropicalforests AT benweinstein multisensordatafusionforimprovedsegmentationofindividualtreecrownsindensetropicalforests AT jamesball multisensordatafusionforimprovedsegmentationofindividualtreecrownsindensetropicalforests AT tobyjackson multisensordatafusionforimprovedsegmentationofindividualtreecrownsindensetropicalforests AT davidcoomes multisensordatafusionforimprovedsegmentationofindividualtreecrownsindensetropicalforests AT gregoirevincent multisensordatafusionforimprovedsegmentationofindividualtreecrownsindensetropicalforests |