Building model generation system based on depth map analysis

Abstract

The invention relates to the technical field of building model analysis systems, in particular to a building model generation system based on depth map analysis, and the system comprises: a depth collection module which is used for obtaining an image of the distance from an image collector to each point in a scene; and the depth cutting module that is used for performing preliminary depth cutting on the whole depth image through pixel value analysis. An error between a point cloud position and an actual building contour salient point is reduced by analyzing and filtering the preliminarily acquired depth image, so that the precision of analyzing the building shape by using the depth image is improved; pixels which may cause errors in building shape analysis in the depth image are filtered and processed in sequence in a layered manner through the depth cutting module, the linear mapping module and the noise filtering module, so that building salient points in the depth image can be judged in the later stage. Therefore, the accuracy of building shape judgment is improved, and the efficiency of building model analysis is improved.

Description

technical field [0001] The invention relates to the technical field of building model analysis systems, in particular to a building model generation system based on depth map analysis. Background technique [0002] With the development of science and technology, the concept of 3D digital city is gradually applied to urban construction work. 3D digital city needs to establish a virtual urban spatial information environment, which can solve many complex problems in urban construction, and depth maps are gradually being used in In the work of real building measurement, the depth image is the three-dimensional representation of the object, which is generally obtained by a stereo camera or a TOF camera. If the internal calibration parameters of the camera are available, the depth image can be converted into a point cloud, which provides a basis for the construction of a three-dimensional model of urban buildings. reliable data. [0003] In the prior art, for example, the Chinese...

AI Technical Summary

Problems solved by technology

[0004] However, in the existing technology, when using the depth image to measure the actual building, the acquired depth image is not filtered and analyzed for irrelevant pixels and error pixels, but only relies on point cloud registration and solving the transformation matrix R and T to calculate the error. Update the position of the point cloud to analyze the shape of the building. Failure to use the initially acquired depth image as the starting point for error analysis will increase the actual measurement error and increase the workload of error analysis, which will greatly reduce the efficiency of building model generation using depth images.

Images