Benoît St-Onge

Automated forest structure mapping from high resolution imagery based on directional semivariogram estimates

Abstract A new segmentation approach that allows forest stands identification on high spatial resolution (⪕1 m) optical imagery is presented. Texture information was first derived by measuring the range of the semivariogram of monochrome image values in three different directions using a moving window. The semivariogram ranges were then used to predict, on a per-pixel basis, three stand structure parameters through regression equations developed for crown diameter, stand density, and crown closure. A region growing algorithm was applied to these three regression estimate images to identify the limits of the forest stands. Calibration of the prediction equations was made using artificial images created by a geometrical-optical process. It was found that forest stands boundaries can be adequately identified on artificial images and that average forest structure estimates within each delineated stand are close to the actual values. Preliminary application of the proposed method to real images acquired with the MEIS-II airborne sensor yielded good segmentation and per stand structure estimates. Some errors were generated due to the fact that the moving window sometimes overlapped two different forest stands because of the presence of areas covered by nonforest vegetation or human made structures. The issue of the moving window size and means to increase the precision of the method are discussed.

Estimating forest stand structure from high resolution imagery using the directional variogram

Abstract The height and stocking of forest stands can be estimated with relatively high precision using an empirical model relating parameters extracted from the directional variogram of high resolution images and forest structure parameters. A geometrical-optical model of the forest was first used to generate images of artificial forest stands in order to establish the relation between tree size. tree density and image texture. The resulting equations were then applied on the computer generated images as well as on high resolution MEIS II images to predict the forest structure values. The results show a good concordance between actual and predicted values, even when spatial resolution was degraded from 0·36m to 2·16m.

Mapping the height and above-ground biomass of a mixed forest using lidar and stereo Ikonos images

Our objective was to assess the accuracy of the forest height and biomass estimates derived from an Ikonos stereo pair and a lidar digital terrain model (DTM). After the Ikonos scenes were registered to the DTM with submetric accuracy, tree heights were measured individually by subtracting the photogrammetric elevation of the treetop from the lidar ground‐level elevation of the tree base. The low residual error (1.66 m) of the measurements confirmed the joint geometric accuracy of the combined models. Matched images of the stereo pair were then used to create a digital surface model. The latter was transformed to a canopy height model (CHM) by subtracting the lidar DTM. Plotwise height percentiles were extracted from the Ikonos‐lidar CHM and used to predict the average dominant height and above‐ground biomass. The coefficient of determination reached 0.91 and 0.79 for average height and biomass, respectively. In both cases, the accuracy of the Ikonos‐lidar CHM predictions was slightly lower than that of the all‐lidar reference CHM. Although the CHM heights did not saturate at moderate biomass levels, as do multispectral or radar images, values above 300 Mg ha−1 could not be predicted accurately by the Ikonos‐lidar or by the all‐lidar CHM.

Mapping canopy height using a combination of digital stereo-photogrammetry and lidar

Ranging techniques such as lidar (LIght Detection And Ranging) and digital stereo‐photogrammetry show great promise for mapping forest canopy height. In this study, we combine these techniques to create hybrid photo‐lidar canopy height models (CHMs). First, photogrammetric digital surface models (DSMs) created using automated stereo‐matching were registered to corresponding lidar digital terrain models (DTMs). Photo‐lidar CHMs were then produced by subtracting the lidar DTM from the photogrammetric DSM. This approach opens up the possibility of retrospective mapping of forest structure using archived aerial photographs. The main objective of the study was to evaluate the accuracy of photo‐lidar CHMs by comparing them to reference lidar CHMs. The assessment revealed that stereo‐matching parameters and left–right image dissimilarities caused by sunlight and viewing geometry have a significant influence on the quality of the photo DSMs. Our study showed that photo‐lidar CHMs are well correlated to their lidar counterparts on a pixel‐wise basis (r up to 0.89 in the best stereo‐matching conditions), but have a lower resolution and accuracy. It also demonstrated that plot metrics extracted from the lidar and photo‐lidar CHMs, such as height at the 95th percentile of 20 m×20 m windows, are highly correlated (r up to 0.95 in general matching conditions).

Comparison of forest attributes extracted from fine spatial resolution multispectral and lidar data

Fine spatial resolution multispectral imagery and light detection and ranging (lidar) data capture differing, yet complementary characteristics of forest structure. Using a dataset consisting of fine spatial resolution multispectral imagery, discrete-return lidar data, and detailed ground-based measurements of individual tree attributes, we applied an automatic tree delineation routine (tree identification and delineation algorithm) to compare and contrast remotely sensed predictions with field observations. The results indicate the automatically extracted crowns derived from lidar data matched tree crown area (coefficient of determination r2 = 0.46, n = 36) and height (r2 = 0.88, n = 36) better than spatial clusters defined in the multispectral imagery (crown area r2 = 0.26, n = 36) for individual trees that were identifiable in both the lidar and multispectral imagery. Differences between crown delineation characteristics were related to the information content of the lidar and multispectral fine spatial resolution data. Investigation of the spectral characteristics of objects defined in the multispectral imagery revealed strong relationships between the vertical positions derived from the lidar data and the apparent multispectral reflectance, with low-reflectance spatial clusters occurring lower in the forest canopy. The application of lidar and multispectral datasets together, in the context of tree crown delineation, provides information not available from either data source independently.

Response of a boreal forest to canopy opening: assessing vertical and lateral tree growth with multi-temporal lidar data

Fine-scale height-growth response of boreal trees to canopy openings is difficult to measure from the ground, and there are important limitations in using stereophotogrammetry in defining gaps and determining individual crowns and height. However, precise knowledge on height growth response to different openings is critical for refining partial harvesting techniques. In this study, we question whether conifers and hardwoods respond equally in terms of sapling growth or lateral growth to openings. We also ask to what distance gaps affect tree growth into the forest. We use multi-temporal lidar to characterize tree/sapling height and lateral growth responses over five years to canopy openings and high resolution images to identify conifers and hardwoods. Species-class-wise height-growth patterns of trees/saplings in various neighborhood contexts were determined across a 6-km matrix of Canadian boreal mixed deciduous coniferous forests. We then use statistical techniques to probe how these growth responses vary by spatial location with respect to the gap edge. Results confirm that both mechanisms of gap closure contribute to the closing of canopies at a rate of 1.2% per annum. Evidence also shows that both hardwood and conifer gap edge trees have a similar lateral growth (average of 22 cm/yr) and similar rates of height growth irrespective of their location and initial height. Height growth of all saplings, however, was strongly dependent on their position within the gap and the size of the gap. Results suggest that hardwood and softwood saplings in gaps have greatest growth rates at distances of 0.5-2 m and 1.5-4 m from the gap edge and in openings smaller than 800 m2 and 250 m2, respectively. Gap effects on the height growth of trees in the intact forest were evident up to 30 m and 20 m from gap edges for hardwood and softwood overstory trees, respectively. Our results thus suggest that foresters should consider silvicultural techniques that create many small openings in mixed coniferous deciduous boreal forests to maximize the growth response of both residual and regenerating trees.

Development of a pit filling algorithm for LiDAR canopy height models

LiDAR canopy height models (CHMs) can exhibit unnatural looking holes or pits, i.e., pixels with a much lower digital number than their immediate neighbors. These artifacts may be caused by a combination of factors, from data acquisition to post-processing, that not only result in a noisy appearance to the CHM but may also limit semi-automated tree-crown delineation and lead to errors in biomass estimates. We present a highly effective semi-automated pit filling algorithm that interactively detects data pits based on a simple user-defined threshold, and then fills them with a value derived from their neighborhood. We briefly describe this algorithm and its graphical user interface, and show its result in a LiDAR CHM populated with data pits. This method can be rapidly applied to any CHM with minimal user interaction. Visualization confirms that our method effectively and quickly removes data pits.

A multiscale geographic object-based image analysis to estimate lidar-measured forest canopy height using Quickbird imagery

Lidar (light detection and ranging) has demonstrated the ability to provide highly accurate information on forest vertical structure; however, lidar data collection and processing are still expensive. Very high spatial resolution optical remotely sensed data have also shown promising results to delineate various forest biophysical properties. In this study, our main objective is to examine the potential of Quickbird (QB) imagery to accurately estimate forest canopy heights measured from small-footprint lidar data. To achieve this, we have developed multiscale geographic object-based image analysis (GEOBIA) models from QB data for both deciduous and conifer stands. In addition to the spectral information, these models also included (1) image-texture [i.e., an internal-object variability measure and a new dynamic geographic object-based texture (GEOTEX) measure that quantifies forest variability within neighboring objects] and (2) a canopy shadow fraction measure that acts as a proxy of vertical forest structure. A novel object area-weighted error calculation approach was used to evaluate model performance by considering the importance of object size. To determine the best object scale [i.e., mean object size (MOS)] for defining the most accurate canopy height estimates, we introduce a new perspective, which considers height variability both between- and within-objects at all scales. To better evaluate the improvements resulting from our GEOBIA models, we compared their performance with a traditional pixel-based approach. Our results show that (1) the addition of image-texture and shadow fraction variables increases the model performance versus using spectral information only, especially for deciduous trees, where the average increase of R 2 is approximately 23% with a further 1.47 m decrease of Root Mean Squared Error (RMSE) at all scales using the GEOBIA approach; (2) the best object scale for our study site corresponds to an MOS of 4.00 ha; (3) at most scales, GEOBIA models achieve more accurate results than pixel-based models; however, we note that inappropriately selected object scales may result in poorer height accuracies than those derived from the applied pixel-based approach.

Tree and Canopy Height Estimation with Scanning Lidar

A large part of the research efforts concerning the remote sensing of forests has been devoted to the development of repeatable methods for the extraction of information from monoscopic, two-dimensional images. Emphasis has been on spectral pattern recognition. Although appropriate for species or health characterisation, this approach comes with several limitations when detailed information on forest structure, e.g. three-dimensional aspects of forest canopies, is sought (Wulder 1998). Accurate measurements of height, density, volume, stratification, etc. at local scales, which are of prime interest for foresters and forest ecologists, and which have a geometric rather than radiometric nature, are still beyond the capabilities of two-dimensional remote sensing and image processing.

Moving Toward Consistent ALS Monitoring of Forest Attributes across Canada

As airborne laser scanning (ALS) gains wider adoption to support forest operations in Canada, the consistency and quality of derivative products that support long-term monitoring and planning are becoming a key issues for managers. The Canadian Consortium for Lidar Environmental Applications Research (C-CLEAR) has supported almost 200 projects across Canada since 2000, with forest-related studies being a dominant theme. In 2010 and 2011, field operations were mobilized to support 13 ALS projects spanning almost the full longitudinal gradient of Canada’s forests. This paper presents case studies for seven plus an overview of some best practices and data processing workflow tools that have resulted from these consortium activities. Although the projects and research teams are spread across Canada, the coordination and decade of experience provided through C-CLEAR have brought common methodological elements to all. It is clear that operational, analytical and reporting guidelines that adhere to community accepted standards are required if the benefits promised by ALS forestry are to be realized. A national Lidar Institute that builds upon the C-CLEAR model and focuses on developing standards, guidelines, and certified training would address this need.