No abstract

Reconstructing surfaces with data coming from an automatic acquisition technique always entails the problem of mass of data. It leads to a mandatory data reduction process. Applying the process to the whole set of points induces an important risk of surface shrinking so that the initial boundary extraction is an important step permitting a simplification inside it. The global surface shape will then be better kept. It is nevertheless required to simplify the boundary, which can be done on the extracted boundary. In this paper, we present a new method to extract and simplify the boundary of an elevation surface given as voxels in a large 3D volume having the characteristics to be sparse since many data are missing. We first present our boundary definition based on mathematical relations between a point and its square neighborhoods. Second, we introduce algorithms to extract such a boundary. Third, we simplify this boundary.

Terrestrial laser scanners provide accurate and detailed point clouds of forest plots, which can be used as an alternative to destructive measurements during forest inventories. Various specialized algorithms have been developed to provide automatic and objective estimates of forest attributes from point clouds. The STEP (Snakes for Tuboid Extraction from Point cloud) algorithm was developed to estimate both stem diameter at breast height and stem diameters along the bole length. Here, we evaluate the accuracy of this algorithm and compare its performance with two other state-of-the-art algorithms that were designed for the same purpose (i.e., the CompuTree and SimpleTree algorithms). We tested each algorithm against point clouds that incorporated various degrees of noise and occlusion. We applied these algorithms to three contrasting test sites: (1) simulated scenes of coniferous stands in Newfoundland (Canada), (2) test sites of deciduous stands in Phalsbourg (France), and (3) coniferous plantations in Quebec, Canada. In most cases, the STEP algorithm predicted diameter at breast height with higher R2 and lower RMSE than the other two algorithms. The STEP algorithm also achieved greater accuracy when estimating stem diameter in occluded and noisy point clouds, with mean errors in the range of 1.1 cm to 2.28 cm. The CompuTree and SimpleTree algorithms respectively produced errors in the range of 2.62 cm to 6.1 cm and 1.03 cm to 3.34 cm, respectively. Unlike CompuTree or SimpleTree, the STEP algorithm was not able to estimate trunk diameter in the uppermost portions of the trees. Our results show that the STEP algorithm is more adapted to extract DBH and stem diameter automatically from occluded and noisy point clouds. Our study also highlights that SimpleTree and CompuTree require data filtering and results corrections. Conversely, none of these procedures were applied for the implementation of the STEP algorithm.

Segmentation of 3D point clouds is still an open issue in the case of unbalanced and in-homogeneous data-sets. In the application context of the modeling of botanical trees, a fundamental challenge consists in separating the leaves from the wood. Based on deep learning and a class decision process, we propose an innovative method designed to separate leaf points from wood points in terrestrial LiDAR point clouds of trees. Although simple, our approach learns trees characteristic point patterns efficiently and robustly. To train our 3D deep learning model, we constructed a 3D labeled point cloud data-set of different tree species. Experiments show that our 3D deep representation together with our geometric approach leads to significant improvement over the state-of-the-art methods in segmentation task.

No abstract

No abstract

International audienceWith specific flora and fauna, regional landscapes and forests constitute an important part of the cultural heritage. Several natural environments have already been classified as national or regional parks. The UNESCO World Heritage covers 13% of the protected forests in the world. Thus, preserving those sites represents a crucial issue. Such a safeguarding involves a detailed knowledge of the sites and forestry management plans. The management of a natural forest is traditionally based on forest plot inventories in which several features of the trees are measured. The set of data collected during these inventories represents the starting point of forest monitoring, flora preservation and risks prevention. Traditionally, measurements are made manually by operators. However, during the last decade, terrestrial laser scanning has become a new and promising way of measuring such attributes. This instrument provides a fine three dimensional point cloud virtual representation of the scanned scene. Trees location, stem diameter, and stem taper can be extracted from these point clouds using pattern recognition algorithms. In this paper we present a novel two steps way to improve the quality of tree branching detection in a three dimensional point cloud acquired by terrestrial laser scanner. This method was developed in order to enhance the results of a previous study. Our approach is based on the combination of a simplification step (using particle simulation), followed by a shape detection (discrete arcs of circle detection). It identifies the lack of accuracy in tree stem diameter measurements at branching junctions for further more detailed analysis

The presence of vegetation and the terrain topography itself generate strong occlusions causing large gaps in terrestrial laser scanning (TLS) data at the ground level as well as a risk of integrating above-ground objects. This article introduces a surface-approximation algorithm dedicated to extracting digital terrain models (DTMs) from terrestrial TLS data acquired in forest areas. The proposed method is based on the combination of a quadtree subdivision of space guided by the local density and distribution of data together with a surface modeling via radial basis functions, which are used as partitions of unity for merging local quadratic approximating patches.