Point cloud building reconstruction method based on segmentation

Abstract

The invention discloses a point cloud building reconstruction method based on segmentation. Particularly the method comprises the following steps of 1, segmenting a point cloud scene by means of area increase and a distance-based clustering algorithm, and extracting planes which forms the building in the point cloud scene; 2, performing longitudinal segmentation on the planes which are extracted in the step 1, calculating the length of each slice, and successively comparing a height difference between each slice and a subsequent slide for detecting key points, and successively connecting the key points for finishing plane reconstruction; 3, optimizing the initial reconstructed model of the building according to a topological relation between the planes; and 4, according to the topological relation between walls and windows, finishing detail reconstruction of the building. The point cloud building reconstruction method settles a problem of insufficient building reconstruction details caused by high point cloud data volume and incomplete point cloud data of a three-dimensional object.

Description

technical field [0001] The invention belongs to the technical field of building reconstruction methods, and relates to a segmentation-based point cloud building reconstruction method. Background technique [0002] 3D digital city has been widely used in many fields such as virtual tourism, city planning, city 3D map and digital entertainment. The three-dimensional construction of urban scenes can give users a more realistic visual experience and help them understand urban geographic information more accurately. Buildings are important components in urban scenes, and it is of great significance to quickly obtain complete and accurate 3D building models. Building reconstruction is also an important research field of computer vision and computer graphics. At present, existing building reconstruction methods are roughly divided into: building reconstruction methods based on shape fitting, and building reconstruction methods based on prior knowledge. [0003] 1. Building recon...

AI Technical Summary

Problems solved by technology

[0011] In the real 3D point cloud scene, due to occlusion, noise interference or single-sided scanning during the scanning process, it is difficult to obtain the complete point cloud data of the building, which makes the existing methods not suitable for the building in the point cloud scene. reconstruction

Images