Point cloud registration, as the first step for the use of point cloud data, has attracted increasing attention. In order to obtain the entire point cloud of a scene, the registration of point clouds from multiple views is necessary. In this paper, we propose an automatic method for the coarse registration of point clouds. The 2D lines are first extracted from the two point clouds being matched. Then, the line correspondences are established and the 2D transformation is calculated. Finally, a method is developed to calculate the displacement along the z-axis. With the 2D transformation and displacement, the 3D transformation can be easily achieved. Thus, the two point clouds are aligned together. The experimental results well demonstrate that our method can obtain high-precision registration results and is computationally very efficient. In the experimental results obtained by our method, the biggest rotation error is 0.5219^o, and the biggest horizontal and vertical errors are 0.2319 m and 0.0119 m, respectively. The largest total computation time is only 713.4647 s.

The interpretation of deep learning network is an important part in understanding the convolutional neural networks (CNNs). As an exploratory research, this article explored the interpretation method in 3-D point cloud deep learning networks, for the purpose of evaluating the performance of convolution functions in 3-D point cloud CNNs. Specifically, a 3-D point cloud classification network with two branches is used as the interpretation network in two aspects; 1) information entropy is introduced to diagnose the internal representation in the middle layer of CNN; and 2) the external consistency of convolution function is measured by per-point classification accuracy with class activation mapping technique. Four typical convolution functions are tested by the interpretation network on ModelNet40 dataset and the experimental results demonstrate that the proposed evaluation method is reliable. Feature transformation ability and feature recognition ability of convolution functions are extracted by visualization evaluation and proposed measurable metrics evaluation.

This paper presents a novel method for segmentation of planar feature from unorganized point cloud based on 2D Hough Transform and octree. Given the input point cloud, three steps are performed to segment planar features. Firstly, the original point cloud is sampled and projected to the X-Y plane, and an extended 2D Hough transform algorithm is employed to extract the line segments. The selecting weight iteration method is used to calculate line equations and endpoint coordinates of those line segments. The space geometric equations of the vertical planes are then determined. Secondly, the octree structure of the original point cloud is established, and then the exact endpoint coordinates of the line segment are used to design a cube perpendicular to the X-Y plane and all points held by the cube are extracted. The distance from each of the extracted points inside the cube to its corresponding facade is calculated, if the distance is less than the predefined threshold, the point is regarded as a point inside the facade. Finally, all the facade points are removed from the original point cloud, and remaining point cloud is sampled and projected to the X-Z plane. The above process is repeated to extract horizontal planes. Evaluation experiments are performed by analyzing the performance of our method in four different scenes. The experimental results indicate that the proposed algorithm is suitable for segmentation of building planar features in different scenes. A comparison with competing techniques shows that our approach is considerably faster and scales significantly better than previous ones.

Most of the existing surface area measurement methods suffer from poor efficiency, low precision, and high computational cost, especially for inaccessible, large-scale, rough, and curved surfaces. In this paper, we propose a method to directly measure the surface areas of unorganized point clouds applicable to various scenes. First, an adaptive super-voxel segmentation algorithm is adopted to divide the input point cloud into a collection of facets with multiple scales. For each facet, all points belonging to it are projected onto its corresponding accurately fitted plane. Second, for each projected facet, rigid transform is performed so that its normal vector is parallel to the Z axis. For each 2D facet point cloud, the x coordinates and y coordinates are utilized to abstract its boundary points. Third, the boundary points are sorted in clockwise order so that every two adjacent points and the center point determine a triangle. Next, an improved interpolation method is adopted to interpolate the sparse edge points. The surface area calculation results of different scales can be obtained by counting the sum of the triangular area inside each facet. Finally, the optimum value is determined from these results. The proposed method is tested on various types of point clouds acquired from different ways. Comprehensive experiments demonstrate that the proposed method is efficient and effective, and capable of obtaining good performances in both simple regular planes and complex surfaces. Particularly, compared with traditional reconstruction-based methods, the proposed method significantly outperforms when dealing with large-scale and complex scenes.

This paper redefines point cloud density by utilizing statistical skewness derived from the geometric relationships between points and their local centroids. By comparing with a symmetric uniform reference model, this method can efficiently describe distribution patterns and detect non-uniform regions. Furthermore, a deep learning model trained on these skewness features achieves 85.96% accuracy in automated boundary extraction, significantly reducing omission errors compared to conventional density-based methods. The proposed framework offers an effective solution for automated point cloud segmentation and modeling.

Point cloud semantic segmentation is a key technology for applications such as autonomous driving, robotics, and virtual reality. Current approaches are heavily reliant on local relative coordinates and simplistic attention mechanisms to aggregate neighborhood information. This often leads to an ineffective joint representation of geometric perturbations and feature variations, coupled with a lack of adaptive selection for salient features during context fusion. On this basis, we propose LSSCC-Net, a novel segmentation framework based on LACV-Net. First, the spatial-feature dynamic aggregation module is designed to fuse offset information by symmetric interaction between spatial positions and feature channels, thus supplementing local structural information. Second, a dual-dimensional attention mechanism (spatial and channel) is introduced to symmetrically deploy attention modules in both the encoder and decoder, prioritizing salient information extraction. Finally, Lovász-Softmax Loss is used as an auxiliary loss to optimize the training objective. The proposed method is evaluated on two public benchmark datasets. The mIoU on the Toronto3D and S3DIS datasets is 83.6% and 65.2%, respectively. Compared with the baseline LACV-Net, LSSCC-Net showed notable improvements in challenging categories: the IoU for “road mark” and “fence” on Toronto3D increased by 3.6% and 8.1%, respectively. These results indicate that LSSCC-Net more accurately characterizes complex boundaries and fine-grained structures, enhancing segmentation capabilities for small-scale targets and category boundaries.