Method for clustering and identification of objects in laser scanning point clouds using dynamic logic

Abstract

Today there is a gap between a presence of various new equipment on the market which provides streams of various digital data about the environment, in particular in the form of laser scanning point clouds, and the lack of adequate e cient methods and software for information extraction from such data. A solution to the problem of bridging this gap is proposed on the basis of neural modeling eld theory and dynamic logic (DL). We present a DL-based method of extracting and analyzing information from hybrid point clouds, which include not only spatial coordinates and intensity, but also the color of each point, and can be from multiple sources including terrestrial, mobile and airborne laser scanning data. The proposed method is signi cant for creating a fundamental theoretical basis for new application algorithms and software for many new applications, including building information modeling, "smart city" environment, etc. The proposed method is fairly new to solving various problems related to extracting semantically rich information from a nontraditional type of digital data, especially hybrid point clouds created from laser scanning. This method will allow to signi cantly expand the existing boundaries of knowledge in the eld of extraction and analysis of information from various digital data, because neural modeling eld theory and DL can improve the performance of relevant calculations and close the existing gap in analysis of digital images.

Saint Petersburg 3D: Creating a Large-Scale Hybrid Mobile LiDAR Point Cloud Dataset for Geospatial Applications

“…The results of our work are the following: (1) we developed the dataset, which is described in this section; (2) we evaluated the performance of deep learning models for the semantic segmentation of dense large-scale point clouds in accordance with the proposed classification scheme in the context of geospatial applications; and (3) we assessed the possibility of both real and synthetic data joint use for point cloud representation learning. In this section, we describe the dataset created and the results of corresponding experiments, which we conducted to evaluate the performance of the modern deep learning models for the semantic segmentation of a dense large-scale point cloud trained on our SP3D dataset.…”

“…A point cloud is a digital record of objects and scenes [1] represented by a large number of 3D points, usually with different densities. Point clouds are a widespread type of data used in robotics, remote sensing, architecture, the construction industry [2], autonomous vehiclea, perception systems [3], geospatial applications [4], digital heritage [5], and many other areas. Most often, point clouds are obtained using active sensors based on LiDAR technologies mounted on a static or mobile platform.…”

Construction 4.0, Industry 4.0, and Building Information Modeling (BIM) for Sustainable Building Development within the Smart City

“…This derivative-free optimization-based method automatically generates a semantically rich as-built BIM from complex scenes using 3D point clouds [46]. The dynamic logic-based method for clustering and identifying objects in laser-scanned point clouds aids in creating a theoretical basis for many new application algorithms and software for BIM and smart urban environments, which addresses various problems related to the extraction of semantically rich information from non-traditional types of digital data [47]. In addition, a reliance on BIM, data mining, and semantic data modeling enable researchers to create a performance-oriented design decision-supporting system, which will inform future design decisions [48].…”