Terrestrial laser scanning (TLS) has been proposed to be an effective tool to serve as the basis for rapid, precise forestry measurements such as diameter-at-breast-height (DBH), tree height, and stem volume. Automated techniques have proved effective at extracting structural measurements from single-tree point clouds using techniques such as quantitative structural modeling (Raumonen et al. 2013). However, the automatic segmentation of single-point tree clouds from scans of forest stands remains challenging, and many approaches involve a significant amount of manual input to generate usable single-tree point clouds. In this study, we investigate the use of Hough transforms to automatically generate stem maps for a range of forest conditions across the northern Sierra Nevada Mountains. We used a combination of Hough circle transforms in the horizontal plane combined with Hough line transforms in the vertical plane to provide additional parameterization on which to optimize filtering to choose which circles correctly identified the presence and size of a tree (Chmielewski et.al 2010, Yuen et. al 1989). Using an independent dataset of tree presence and DBH combined with a machine learning model, we determined the ideal parameter set to maximize the accuracy of estimates of tree presence/absence and tree size.
Works Cited:
Chmielewski, L. J., Bator, M., Zasada, M., Stereńczak, K., & Strzeliński, P. (2010). Fuzzy Hough Transform-Based Methods for Extraction and Measurements of Single Trees in Large-Volume 3D Terrestrial LIDAR Data. Computer Vision and Graphics Lecture Notes in Computer Science, 265-274. doi:10.1007/978-3-642-15910-7_30
Raumonen, P., Kaasalainen, M., Åkerblom, M., Kaasalainen, S., Kaartinen, H., Vastaranta, M., . . . Lewis, P. (2013). Fast Automatic Precision Tree Models from Terrestrial Laser Scanner Data. Remote Sensing, 5(2), 491-520. doi:10.3390/rs5020491
Yuen, H. K., Princen, J., Dlingworth, J., & Kittler, J. (1989). A comparative study of Hough Transform methods for circle finding. Proceedings of the Alvey Vision Conference 1989. doi: 10.5244/c.3.29
