Vehicle-mounted laser point cloud and sequence panoramic image registration method

Abstract

The invention belongs to the technical field of surveying and mapping, and discloses a vehicle-mounted laser point cloud and sequence panoramic image registration method, which comprises the steps ofobtaining vehicle-mounted laser point cloud data and sequence panoramic image data, and obtaining a first skyline vector and a second skyline vector; respectively constructing a laser point cloud registration primitive graph and an image registration primitive graph; forming a conjugate registration primitive pair; resolving a pre-established 2D-3D coarse registration model by using the conjugateregistration primitive pair to obtain space coordinate initial conversion parameter information between a photogrammetric coordinate system and a LiDAR reference coordinate system, and completing 2D-3D coarse registration; generating an MVS image dense point cloud based on the sequence panoramic image data; and taking the space coordinate initial conversion parameter information as an initial value, and performing optimal registration between the MVS image dense point cloud and the vehicle-mounted laser point cloud data to complete 3D-3D accurate registration. According to the invention, the registration result precision is high, and the precision consistency is good.

Description

Technical field [0001] The present invention relates to the field of surveying and mapping, and in particular, to a registration method of a sequence panoramic image of an in-vehicle laser point cloud. Background technique [0002] With the rapid development of electronics, sensors and computer technology, measuring cameras, laser scanners, POS (POSITION AND ORIENTATION SYSTEM, Positioning System), etc., the sensor, etc., to slightly, and light weight, make low costs, Light small, integrated, multi-sensor remote sensor remote sensor remote sensing system has become a reality. At the same time, the sensor mounting platform has been diversely developed, especially the rapid development of low-altitude UAV (Unmanned Aerial Vehicle, UAV) and car mobile measuring system, which makes up for traditional satellites and human machine photographic measurements and remote sensing in medium and low sky and surface High resolution multi-source space data coverage. Under the common driving of ...

AI Technical Summary

Problems solved by technology

This type of method uses the output value of the POS system as the initial registration, and uses the non-rigid closest point iterative method to realize the registration of the two, but the iterative closest point (ICP) method has higher requirements for the approximate value of the initial conversion parameters. In the case of a large deviation of the initial value provided by POS, it is difficult to guarantee the convergence of the method, and the adaptability of the algorithm is limited.

Images