Classification of airborne laser scanning (ALS) point clouds is needed in digital cities and 3-D modeling. To efficiently recognize objects in ALS point clouds, we propose a novel hierarchical aggregated deep feature representation method, which can adequately employ spatial association of multilevel structures and deep feature discrimination. In our method, a 3-D deep learning model is constructed to represent the discriminative feature of each point cluster in a hierarchical structure by decreasing the within-class distance and increasing the between-class distance. Our method aggregates the discriminative deep features in different levels into a hierarchical aggregated deep feature that considers the spatial hierarchy and feature distinctiveness. Lastly, we build a multichannel 1-D convolutional neural network to classify the unknown points. Our tests demonstrate that the proposed hierarchical aggregated deep feature method can enhance point cloud classification results. Comparing with seven state-of-the-art methods, those results also verified the superior performance of our method.

Accurate and efficient semantic segmentation of buildings in high spatial resolution (HSR) remote sensing images is the basis for applications such as fine urban management, high-precision mapping, land resource utilization investigation, and human settlement suitability evaluation. The current building extraction methods based on deep learning can obtain high-level abstract features of images. However, due to the limitation of convolution kernel size and the vanishing gradient, the extraction of some buildings is inaccurate, and some small-volume buildings are missing as the network deepens. In this regard, we design a horizontally connected residual blocks-based multiscale attention network to achieve high-quality extraction of buildings in HSR remote sensing image. In this network, we subdivide each residual block by channel grouping and feature horizontal connection to consider the difference and saliency of feature information between channels, and then combine the output features with multiscale attention module to consider the contextual semantic relationship of different regions and integrate multilevel local and global information of buildings. A stepwise up-sampling mechanism is designed in the decoding process to finally achieve precise semantic segmentation of buildings. We conduct experiments on two public datasets and compare the proposed method with state-of-the-art semantic segmentation methods. The experiments show that our method could achieve better building extraction results in HSR remote sensing image, which proves the effectiveness of our proposed method.

This letter introduces an efficient bundle adjustment (BA) method based on the sparse Broyden-Fletcher-Goldfarb-Shanno (sparse BFGS, sBFGS) solution, which efficiently estimates camera poses and 3-D points. The Levenberg-Marquardt (LM) solution, which is widely applied in BA problems, requires more linear equation solution and more iterations. A gain matrix calculated using both the Jacobian matrix and residual vector replaces the simple diagonal matrix of the LM solution, which results in a better estimation of the descent direction and step size when finding a path to a local minimum in a sparse BFGS solution. Four datasets were verified, and the results demonstrate that the proposed method requires fewer linear equation solutions to converge to a minimum compared with that of the LM-based BA method.

The point cloud registration is important and necessary for the applications of changing detection, deformation monitoring, and so on, which is also challenging due to the vast clustered points, irregular, and complex structures of spatial objects, and quality effects of the labeled corresponding points. Aiming at this problem, we design an end-to-end 3-D graph deep learning framework of point cloud registration, which can simultaneously learn the detector (graph attention expression) and the descriptor (graph deep feature) for point cloud registration in a weakly supervised way, so that the learned detector and descriptor promote each other in the process of model optimization. Then, the detector is used to automatically extract the keypoints, and the descriptor describes the deep feature of each keypoint. In the framework, we innovatively propose a new module (named MLP_GCN), which fuses multilayer perceptron (MLP) and graph convolutional network (GCN). The MLP_GCN module is further integrated into the detector branch and descriptor branch to fully express the detector and descriptor of the point cloud. In the training process of the framework, we rotate and translate the point cloud randomly to form the training data in a weakly supervised way, which can save plenty of manually labeling time of corresponding points. In the experiments, our method can achieve better results of point cloud registration in comparison with other methods, which verifies the advantages of the proposed method.

In close-range or unmanned aerial vehicle (UAV) photogrammetry, Schneider concentric circular coded targets (SCTs), which are public, are widely used for image matching and as ground control points. GSI point-distributed coded targets (GCTs), which are only mainly applied in a video-simultaneous triangulation and resection system (V-STARS), are non-public and rarely applied in UAV photogrammetry. In this paper, we present our innovative detailed solution to identify GCTs. First, we analyze the structure of a GCT. Then, a special 2D P²-invariant of five coplanar points derived from cross ratios is adopted in template point registration and identification. Finally, the affine transformation is used for decoding. Experiments indoors—including different viewing angles ranging from 0° to 80° based on 6 mm-diameter GCTs, smaller 3 mm-diameter GCTs, and different sizes mixed—and outdoors with challenging scenes were carried out. Compared with V-STARS, the results show that the proposed method can preserve the robustness and achieves a high accuracy rate in identification when the viewing angle is not larger than 65° through indoor experiments, and the proposed method can achieve approximate or slightly weaker effectiveness than V-STARS on the whole. Finally, we attempted to extend and apply the designed GCTs in UAV photogrammetry for a preliminary experiment. This paper demonstrates that GCTs can be designed, printed, and identified easily through our method. It is expected that the proposed method may be helpful when applied to image matching, camera calibration, camera orientation, or 3D measurements or serving as control points in UAV photogrammetry for scenarios with complex structures in the future.