Filtering stems and branches from terrestrial laser scanning point clouds using deep 3-D fully convolutional networks

Simple item page
dc.contributor.authorXi, Zhouxin
dc.contributor.authorHopkinson, Christopher
dc.contributor.authorChasmer, Laura
dc.date.accessioned2024-08-28T22:25:36Z
dc.date.available2024-08-28T22:25:36Z
dc.date.issued2018
dc.descriptionOpen access article. Creative Commons Attribution 4.0 International license (CC BY 4.0) applies
dc.description.abstractTerrestrial laser scanning (TLS) can produce precise and detailed point clouds of forest environment, thus enabling quantitative structure modeling (QSM) for accurate tree morphology and wood volume allocation. Applying QSM to plot-scale wood delineation is highly dependent on wood visibility from forest scans. A common problem is to filter wood point from noisy leafy points in the crowns and understory. This study proposed a deep 3-D fully convolution network (FCN) to filter both stem and branch points from complex plot scans. To train the 3-D FCN, reference stem and branch points were delineated semi-automatically for 14 sampled areas and three common species. Among seven testing areas, agreements between reference and model prediction, measured by intersection over union (IoU) and overall accuracy (OA), were 0.89 (stem IoU), 0.54 (branch IoU), 0.79 (mean IoU), and 0.94 (OA). Wood filtering results were further incorporated to a plot-scale QSM to extract individual tree forms, isolated wood, and understory wood from three plot scans with visual assessment. The wood filtering experiment provides evidence that deep learning is a powerful tool in 3-D point cloud processing and parsing.
dc.description.peer-reviewYes
dc.identifier.citationXi, Z., Hopkinson, C., & Chasmer, L. (2018). Filtering stems and branches from terrestrial laser scanning point clouds using deep 3-D fully convolutional networks . Remote Sensing, 10(8), Article 1215. https://doi.org/10.3390/rs10081215
dc.identifier.urihttps://hdl.handle.net/10133/6894
dc.language.isoen
dc.publisherMDPI
dc.publisher.departmentDepartment of Chemistry and Biochemistry
dc.publisher.facultyArts and Science
dc.publisher.institutionUniversity of Lethbridge
dc.publisher.urlhttps://doi.org/10.3390/rs10081215
dc.subjectTLS
dc.subjectQSM
dc.subjectCNN
dc.subjectFCN
dc.subjectDeep learning
dc.subject3-D point clouds
dc.subjectForest
dc.subjectSegmentation
dc.subjectTree structure
dc.subjectStem points
dc.subjectBranch points
dc.subjectWood filtering
dc.subjectTerrestrial laser scanning
dc.subject.lcshDeep learning (Machine learning)
dc.titleFiltering stems and branches from terrestrial laser scanning point clouds using deep 3-D fully convolutional networks
dc.typeArticle
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Hopkinson-filtering-stems.pdf
Size:
7.53 MB
Format:
Adobe Portable Document Format
Description:
Download
License bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
license.txt
Size:
1.75 KB
Format:
Item-specific license agreed upon to submission
Description:
Download
Collections
Hopkinson, Christopher
Chasmer, Laura