Three-dimensional forest reconstruction and structural parameter retrievals using a terrestrial full-waveform lidar instrument (Echidna®) (original) (raw)
Related papers
• This study was designed to explore the utility of using a full-waveform terrestrial lidar for three-dimensional reconstructions of forest stands. • Reconstructions were assembled from multiple scans of full waveform terrestrial lidar using the Echidna® Validation Instrument (EVI). • The study area included four forest stands in Sierra National Forest, CA and two in Harvard Forest, ME, each 50 m by 50 m in size, with varying canopy structure and species composition, and using data acquired in 2008 and 2009. • Each lidar pulse return was processed to identify one or multiple "hits" and their associated peak return power, converted peak power to apparent reflectance, located hits in Cartesian coordinate space, and stored them as points in a point cloud with associated attributes. • In addition, five (Sierra) or nine (Harvard) overlapping scans were registered and merged into a single point cloud, then the ground plane was identified and ground hits classified; a local digital elevation model was produced; non-groundhits were identified as trunk/branch or foliage hits; and commercial software was used to display, manipulate and interact with the point cloud to make direct measurements of trees in the virtual space of the reconstruction. • The typical size of an EVI scan is 1 hectare and the time required to perform a single scan was about 20 minutes.
PLOS ONE
Terrestrial laser scanning is a powerful technology for capturing the three-dimensional structure of forests with a high level of detail and accuracy. Over the last decade, many algorithms have been developed to extract various tree parameters from terrestrial laser scanning data. Here we present 3D Forest, an open-source non-platform-specific software application with an easy-to-use graphical user interface with the compilation of algorithms focused on the forest environment and extraction of tree parameters. The current version (0.42) extracts important parameters of forest structure from the terrestrial laser scanning data, such as stem positions (X, Y, Z), tree heights, diameters at breast height (DBH), as well as more advanced parameters such as tree planar projections, stem profiles or detailed crown parameters including convex and concave crown surface and volume. Moreover, 3D Forest provides quantitative measures of between-crown interactions and their real arrangement in 3D space. 3D Forest also includes an original algorithm of automatic tree segmentation and crown segmentation. Comparison with field data measurements showed no significant difference in measuring DBH or tree height using 3D Forest, although for DBH only the Randomized Hough Transform algorithm proved to be sufficiently resistant to noise and provided results comparable to traditional field measurements.
Retrieval of forest structural parameters using a ground-based lidar instrument (Echidna
2008
In this paper, a literature overview is presented on the use of laser rangefinder techniques for the retrieval of forest inventory parameters and structural characteristics. The existing techniques are ordered with respect to their scale of application (i.e. spaceborne, airborne, and terrestrial laser scanning) and a discussion is provided on the efficiency, precision, and accuracy with which the retrieval of structural parameters at the respective scales has been attained. The paper further elaborates on the potential of LiDAR (Light Detection and Ranging) data to be fused with other types of remote sensing data and it concludes with recommendations for future research and potential gains in the application of LiDAR for the characterization of forests.
LiDAR remote sensing of forest structure
Progress in physical …, 2003
Light detection and ranging (LiDAR) technology provides horizontal and vertical information at high spatial resolutions and vertical accuracies. Forest attributes such as canopy height can be directly retrieved from LiDAR data. Direct retrieval of canopy height provides opportunities to model above-ground biomass and canopy volume. Access to the vertical nature of forest ecosystems also offers new opportunities for enhanced forest monitoring, management and planning.
Estimating canopy structure of Douglas-fir forest stands from discrete-return LiDAR
Trees-structure and Function, 2007
Variations in vertical and horizontal forest structure are often difficult to quantify as field-based methods are labour intensive and passive optical remote sensing techniques are limited in their capacity to distinguish structural changes occurring below the top of the canopy. In this study the capacity of small footprint (0.19 cm), discrete return, densely spaced (0.7 hits/m -2 ), multiple return, Light Detection and Ranging (LiDAR) technology, to measure foliage height and to estimate several stand and canopy structure attributes is investigated. The study focused on six Douglas-fir [Pseudotsuga menziesii spp. menziesii (Mirb.) Franco] and western hemlock [Tsuga heterophylla (Raf.) Sarg.] stands located on the east coast of Vancouver Island, British Columbia, Canada, with each stand representing a different structural stage of stand development for forests within this biogeoclimatic zone. Tree height, crown dimensions, cover, and vertical foliage distributions were measured in 20 m · 20 m plots and correlated to the LiDAR data. Foliage profiles were then fitted, using the Weibull probability density function, to the field measured crown dimensions, vertical foliage density distributions and the LiDAR data at each plot. A modified canopy volume approach, based on methods developed for full waveform LiDAR observations, was developed and used to examine the vertical and horizontal variation in stand structure. Results indicate that measured stand attributes such as mean stand height, and basal area were significantly correlated with LiDAR estimates (r 2 = 0.85, P < 0.001, SE = 1.8 m and r 2 = 0.65, P < 0.05, SE = 14.8 m 2 ha -1 , respectively). Significant relationships were also found between the LiDAR data and the field estimated vertical foliage profiles indicating that models of vertical foliage distribution may be robust and transferable between both field and LiDAR datasets. This study demonstrates that small footprint, discrete return, LiDAR observations can provide quantitative information on stand and tree height, as well as information on foliage profiles, which can be successfully modelled, providing detailed descriptions of canopy structure.
Assessment of forest structure with airborne LiDAR and the effects of platform altitude
Remote Sensing of Environment, 2006
Airborne scanning LiDAR is a spatial technology increasingly used for forestry and environmental applications. However, the accuracy and coverage of LiDAR observations is highly dependent on both the extrinsic specifications of the LiDAR survey as well as the intrinsic effects such as the underlying forest structure. Extrinsic parameters which are set as part of the LiDAR survey include platform altitude, scan angle (half max. angle off nadir), and beam cross sectional diameter at the reflecting surface (referred to as footprint size). In this paper we investigate the effect of a number of these extrinsic parameters, including three different platform altitudes (1000, 2000, and 3000 m), two scan angles at 1000 m (10°and 15°h alf max. angle off nadir), and three footprint sizes (0.2, 0.4, and 0.6 m). The comparison was undertaken in eucalypt forests at three sites, varying in vegetation structure and topography within the Wedding Bells State Forest, Coffs Harbour, Australia. Results at the plot scale (40 × 90 m areas) indicate that tree heights computed from the 1000 m LiDAR data set (10°half max. angle off nadir) are well correlated with maximum plot heights (difference < 3 m) and field measured canopy volume (r 2 > 0.75, p < 0.001). Using normalised canopy height profiles (CHP) derived for sites, from data recorded at each altitude, we observed no significant difference between the relative distribution of LiDAR returns, indicating that platform altitude and footprint size have not had a major influence on CHP estimation. Interestingly, comparisons of first and last returns for individual pulses at increasing altitudes identified progressively fewer discrete first/last pulse combinations with more than 70% of pulses recorded as a single return at the highest altitude (3000 m). A possible hypothesis is that greater platform altitude and footprint size reduces the intensity of laser beam incident on a given surface area thus decreasing the probability of recording a last return above the noise threshold. Furthermore, tree scale analysis found a positive relationship between platform altitude and the underestimation of crown area and crown volume. The implications of this work for forest management are: (i) platform altitudes as high as 3000 m can be used to quantify the vertical distribution of phyto-elements, (ii) higher platform altitudes record a lower proportion of first/last return combinations that will further reduce the number of points available for forest structural assessment and development of digital elevation models, and (iii) for discrete LiDAR data, increasing platform altitude will record a lower frequency of returns per crown, resulting in larger underestimates of individual tree crown area and volume if standard algorithms are applied.
LiDAR remote sensing of forest strucuture
Light detection and ranging (LiDAR) technology provides horizontal and vertical information at high spatial resolutions and vertical accuracies. Forest attributes such as canopy height can be directly retrieved from LiDAR data. Direct retrieval of canopy height provides opportunities to model above-ground biomass and canopy volume. Access to the vertical nature of forest ecosystems also offers new opportunities for enhanced forest monitoring, management and planning.
We examine the accuracy of tree height estimates obtained via light detection and ranging (LiDAR) in a temperate rainforest characterized by complex terrain, steep slopes, and high canopy cover. The evaluation was based on precise top and base locations for Ͼ1,000 trees in 45 plots distributed across three forest types, a dense network of ground elevation recordings obtained with survey grade equipment, and LiDAR data from high return density acquisitions at leaf-on and leaf-off conditions. Overall, LiDAR error exceeded 10% of tree height for 60% of the trees and 43% of the plots at leaf-on and 55% of the trees and 38% of the plots at leaf-off. Total error was decomposed into contributions from errors in the estimates of tree top height, ground elevation model, and tree lean, and the relationships between those errors and stand-and site-level variables were explored. The magnitude of tree height error was much higher than those documented in other studies. These findings, coupled with observations that indicate suboptimal performance of standard algorithms for data preprocessing, suggest that obtaining accurate estimates of tree height via LiDAR in conditions similar to those in the US Pacific Northwest may require substantial investments in laser analysis techniques research and reevaluation of laser data acquisition specifications. FOR. SCI. 56(2):139-155.
Automated Methods for Measuring DBH and Tree Heights with a Commercial Scanning Lidar
Accurate forest structural parameters are crucial to forest inventory, and modeling of the carbon cycle and wildlife habitat. Lidar (Light Detection and Ranging) is particularly suitable to the measurement of forest structural parameters. In this paper, we describe a pilot study to extract forest structural parameters, such as tree height, diameter at breast height (DBH), and position of individual tree using a terrestrial lidar (LMS-Z360i; Riegel, Inc.). The lidar was operated to acquire both vertical and horizontal scanning in the field in order to obtain a point cloud of the whole scene. An Iterative Closet Point (ICP) algorithm was introduced to obtain the transformation matrix of each range image and to mosaic multiple range images together. Based on the mosaiced data set, a variable scale and threshold filtering method was used to separate ground from the vegetation. Meanwhile, a Digital Elevation Model (DEM) and a Canopy Height Model (CHM) were generated from the classified point cloud. A stem detection algorithm was used to extract the location of individual trees. A slice above 1.3 m from the ground was extracted and rasterized. A circle fitting algorithm combined with the Hough transform was used to retrieve the DBH based on the rasterized grid. Tree heights were calculated using the height difference between the minimum and maximum Z values within the position of each individual tree with a 1 m buffer. All of the 26 trees were detected correctly, tree height and DBH were determined with a precision of 0.76 m and 3.4 cm, respectively, comparing with those visually measured in the lidar data. Our methods and results confirm that terrestrial lidar can provide nondestructive, high-resolution, and automatic determination of parameters required in forest inventory. PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING M a r c h 2 0 1 1 219
Forest structural parameter extraction using terrestrial LiDAR
Accurate forest structural parameters are crucial to forest inventory, and modeling of carbon cycle and wildlife habit. LiDAR (Light Detection and Ranging) is a novel technique to the measurement of forest structural parameters. In this paper we describe a pilot study to extract forest structural parameters, such as tree height, DBH (diameter at breast height), and the position of individual tree using a terrestrial LiDAR. Raw data were acquired using a terrestrial LiDAR (Riegel, LMS-Z360i). The LiDAR was operated to acquire both vertical and horizontal scanning in order to obtain point cloud of the whole scene. An ICP (Iterative Closet Point) algorithm was introduced to obtain the transformation matrix of each range image and to mosaic multiple range images together. Based on the whole data sets, a variable scale and threshold filtering method was used to separate ground from the vegetation. Meanwhile, a DEM (Digital Elevation Model) and a CHM (Canopy Height Model) were generated from the classified point cloud. A stem detection algorithm was used to extract the location of individual trees. A slice above 1.3 m from the ground was extracted and rasterized. A circle fitting algorithm was used to retrieve the DBH based on the rasterized grid. Tree heights were calculated using the height difference between the minimum and maximum Z value within the position of individual tree with 1 m buffer. Our methods and results confirm that terrestrial LiDAR can provide nondestructive, high-resolution and automatic determination of parameters required in forest inventory.