Light detection and ranging (LiDAR) data of 3D point clouds acquired from laser sensors is a crucial form of geospatial data for recognition of complex objects since LiDAR data provides geometric information in terms of 3D coordinates with additional attributes such as intensity and multiple returns. In this paper, we focused on utilizing multiple returns in the training data for semantic segmentation, in particular building extraction using PointNet++. PointNet++ is known as one of the efficient and robust deep learning (DL) models for processing 3D point clouds. On most building boundaries, two returns of the laser pulse occur. The experimental results demonstrated that the proposed approach could improve building extraction by adding two returns to the training datasets. Specifically, the recall value of the predicted building boundaries for the test data was improved from 0.7417 to 0.7948 for the best case. However, no significant improvement was achieved for the new data because the new data had relatively lower point density compared to the training and test data.

Most object detection, recognition, and classification are performed using optical imagery. Images are unable to fully represent the real-world due to the limited range of the visible light spectrum reflected light from the surfaces of the objects. In this regard, physical and geometrical information from other data sources would compensate for the limitation of the optical imagery and bring a synergistic effect for training deep learning (DL) models. In this paper, we propose to classify terrain features using convolutional neural network (CNN) based SegNet model by utilizing 3D geospatial data including infrared (IR) orthoimages, digital surface model (DSM), and derived information. The slope, aspect, and shaded relief images (SRIs) were derived from the DSM and were used as training data for the DL model. The experiments were carried out using the Vaihingen and Potsdam dataset provided by the German Society for Photogrammetry, Remote Sensing and Geoinformation (DGPF) through the International Society for Photogrammetry and Remote Sensing (ISPRS). The dataset includes IR orthoimages, DSM, airborne LiDAR data, and label data. The motivation of utilizing 3D data and derived information for training the DL model is that real-world objects are 3D features. The experimental results demonstrate that the proposed approach of utilizing and integrating various informative feature data could improve the performance of the DL for semantic segmentation. In particular, the accuracy of building classification is higher compared with other natural objects because derived information could provide geometric characteristics. Intersection-of-union (IoU) of the buildings for the test data and the new unseen data with combining all derived data were 84.90% and 52.45%, respectively.

The present study investigates indirect effects of communication about suicide in interpersonal relationships and patient-provider relationships on suicidal intention via attitude, descriptive norm, injunctive norm, and self-efficacy in committing suicide among Korean adults, using longitudinal survey data. Moderation effects of perceived social support are also tested. As a result of path analysis (N = 984), communication about suicide with family, friends, and coworkers is significantly related to pro-suicide descriptive norm and pro-suicide injunctive norm. Positive attitude toward suicide, pro-suicide injunctive norm, and high self-efficacy in committing suicide are significantly related to suicidal intention. Interpersonal communication is positively related to pro-suicide injunctive norm, which in turn links to suicidal intention. Communication about suicide with healthcare professionals, however, did not show significant direct and indirect effects. Moderation effects of perceived social support are detected between interpersonal communication and pro-suicide attitude and pro-suicide injunctive norm, as well as between patient-provider communication and pro-suicide attitude.

Light detection and ranging (LiDAR) data collected from airborne laser scanning systems are one of the major sources of spatial data. Airborne laser scanning systems have the capacity for rapid and direct acquisition of accurate 3D coordinates. Use of LiDAR data is increasing in various applications, such as topographic mapping, building and city modeling, biomass measurement, and disaster management. Segmentation is a crucial process in the extraction of meaningful information for applications such as 3D object modeling and surface reconstruction. Most LiDAR processing schemes are based on digital image processing and computer vision algorithms. This paper introduces a shape descriptor method for segmenting LiDAR point clouds using a “multilevel cube code” that is an extension of the 2D chain code to 3D space. The cube operator segments point clouds into roof surface patches, including superstructures, removes unnecessary objects, detects the boundaries of buildings, and determines model key points for building modeling. Both real and simulated LiDAR data were used to verify the proposed approach. The experiments demonstrated the feasibility of the method for segmenting LiDAR data from buildings with a wide range of roof types. The method was found to segment point cloud data effectively.

Demand for orthoimages is increasing as a crucial component of geographic information systems (GISs). Orthoimages are geometrically equivalent to planimetric maps, which show true geographic locations of terrain features. To produce orthoimages, geometric distortions from camera tilt and relief displacement from perspective images must be corrected. Traditionally, removing such distortions has been accomplished by differential rectification in a pixel-by-pixel fashion. However, this method cannot produce true orthoimages because of the double-mapping problem. We propose a method of generating patch-based true orthoimages for surface patches in buildings. The proposed method utilizes three-dimensional (3D) building model data. Patches from the data were projected onto aerial images to extract image patches and analysis of the superstructures was performed. Because orthoimages are generated for each building, the orthoimage quality is enhanced when using building data with a high level of detail. Instead of performing the complex visibility analysis of existing approaches, this article identifies occlusion areas based on unit surfaces of buildings and presents mutual recovery of occlusions using multiple images. To evaluate the feasibility of the method, experiments were performed with real datasets: (1) a building with a dome superstructure, (2) high-rise buildings close to each other, and (3) buildings with various shapes.

This paper introduces an automated method for building height recovery through the integration of high‐resolution satellite images and digital vector maps. A cross‐correlation matching method along the vertical line locus on the Ikonos images was deployed to recover building heights. The rational function models composed of rational polynomial coefficients were utilized to create a stereopair of the epipolar resampled Ikonos images. Building footprints from the digital maps were used for locating the vertical guideline along the building edges. The digital terrain model (DTM) was generated from the contour layer in the digital maps. The terrain height derived from the DTM at each foot of the buildings was used as the starting location for image matching. At a preset incremental value of height along the vertical guidelines derived from vertical line loci, an evaluation process that is based on the cross‐correlation matching of the images was carried out to test if the top of the building has reached where maximum correlation occurs. The accuracy of the reconstructed buildings was evaluated by the comparison with manually digitized 3D building data derived from aerial photographs.