Building boundaries play an essential role in many applications such as urban planning and production of 3D realistic views. In this context, airborne LiDAR data have been explored for the generation of digital building models. Despite the many developed strategies, there is no method capable of encompassing all the complexities in an urban environment. In general, the vast majority of existing regularization methods are based on building boundaries that are made up of straight lines. Therefore, the development of a strategy able to model building boundaries, regardless of their degree of complexity is of high importance. To overcome the limitations of existing strategies, an iterative CD-spline (changeable degree spline) regularization method is proposed. The main contribution is the automated selection of the polynomial function that best models each segment of the building roof boundaries. Conducted experiments with real data verified the ability of the proposed approach in modeling boundaries with different levels of complexities, including buildings composed of complex curved segments and point cloud with different densities, presenting F_score and PoLiS around 95% and 0.30 m, respectively.

LiDAR data acquired by various platforms provide unprecedented data for forest inventory and management. Among its applications, individual tree detection and segmentation are critical and prerequisite steps for deriving forest structural metrics, especially at the stand level. Although there are various tree detection and localization approaches, a comparative analysis of their performance on LiDAR data with different characteristics remains to be explored. In this study, a new trunk-based tree detection and localization approach (namely, height-difference-based) is proposed and compared to two state-of-the-art strategies—DBSCAN-based and height/density-based approaches. Leaf-off LiDAR data from two unmanned aerial vehicles (UAVs) and Geiger mode system with different point densities, geometric accuracies, and environmental complexities were used to evaluate the performance of these approaches in a forest plantation. The results from the UAV datasets suggest that DBSCAN-based and height/density-based approaches perform well in tree detection (F1 score > 0.99) and localization (with an accuracy of 0.1 m for point clouds with high geometric accuracy) after fine-tuning the model thresholds; however, the processing time of the latter is much shorter. Even though our new height-difference-based approach introduces more false positives, it obtains a high tree detection rate from UAV datasets without fine-tuning model thresholds. However, due to the limitations of the algorithm, the tree localization accuracy is worse than that of the other two approaches. On the other hand, the results from the Geiger mode dataset with low point density show that the performance of all approaches dramatically deteriorates. Among them, the proposed height-difference-based approach results in the greatest number of true positives and highest F1 score, making it the most suitable approach for low-density point clouds without the need for parameter/threshold fine-tuning.

The alpha-shape ( $\alpha $ -shape) concept, which has its origin in computational geometry, is usually applied in building boundary extraction from airborne LiDAR data. However, the results depend on the appropriate choice of the parameter $\alpha $ . Despite several studies in the literature, the adaptive choice of the parameter $\alpha $ persists a challenge in boundary extraction, especially when abrupt density variations occur. To overcome this limitation, this letter proposes a new approach combining five estimation strategies. In the proposed method, these strategies are tested sequentially, prioritizing the one that provides greater level of details. The experiments were conducted considering buildings with different characteristics, which were selected from two LiDAR data sets with the average point densities of 12 points/m ² and 4 points/m ² . The obtained results, presenting $\boldsymbol {F} _{{\text {score}}}$ and PoLiS around 98% and 0.32 m, respectively, indicate the robustness of the proposed method even when abrupt density variation occurs.

The classification is an important step in the extraction of geometric primitives from LiDAR data. Normally, it is applied for the identification of points sampled on geometric primitives of interest. In the literature there are several studies that have explored the use of eigenvalues to classify LiDAR points into different classes or structures, such as corner, edge, and plane. However, in some works the classes are defined considering an ideal geometry, which can be affected by the inadequate sampling and/or by the presence of noise when using real data. To overcome this limitation, in this paper is proposed the use of metrics based on eigenvalues and the k-means method to carry out the classification. So, the concept of principal component analysis is used to obtain the eigenvalues and the derived metrics, while the k-means is applied to cluster the roof points in two classes: edge and non-edge. To evaluate the proposed method four test areas with different levels of complexity were selected. From the qualitative and quantitative analyses, it could be concluded that the proposed classification procedure gave satisfactory results, resulting in completeness and correctness above 92% for the non-edge class, and between 61% to 98% for the edge class.