Data Fusion of Lidar Into a Region Growing Stereo Algorithm

Abstract

ABSTRACT:Stereo vision and LIDAR continue to dominate standoff 3D measurement techniques in photogrammetry although the two techniques are normally used in competition. Stereo matching algorithms generate dense 3D data, but perform poorly on low-texture image features. LIDAR measurements are accurate, but imaging requires scanning and produces sparse point clouds. Clearly the two techniques are complementary, but recent attempts to improve stereo matching performance on low-texture surfaces using data fusion have focused on the use of time-of-flight cameras, with comparatively little work involving LIDAR.A low-level data fusion method is shown, involving a scanning LIDAR system and a stereo camera pair. By directly imaging the LIDAR laser spot during a scan, unique stereo correspondences are obtained. These correspondences are used to seed a regiongrowing stereo matcher until the whole image is matched. The iterative nature of the acquisition process minimises the number of LIDAR points needed. This method also enables simple calibration of stereo cameras without the need for targets and trivial coregistration between the stereo and LIDAR point clouds. Examples of this data fusion technique are provided for a variety of scenes.

LiDAR-guided stereo matching with a spatial consistency constraint

“…Dense image matching constrained by LiDAR data is advantageous for automated 3D reconstruction due to the complementary characteristics of LiDAR and images (Nickels et al, 2003). Several studies used LiDAR data projected on an image as a seed point and then generated a depth/disparity map of the entire image through a region growing algorithm (Gandhi et al, 2012;Veitch-Michaelis et al, 2015). Other studies adjusted or added the penalty parameters in the cost aggregation process that depends on the distance between the current disparity and the disparity of the LiDAR projection point (Huang et al, 2018;Huber and Kanade, 2011).…”

Remote sensing of chemical agents within nuclear facilities using Raman spectroscopy

“…The Raman instrument could also be deployed via a robotic arm if required. The Raman probe section of the instrument was mounted on an ROV and combined with a 3‐D vision stereo camera system [ 4 ] to view potential target samples. The 3‐D vision system also provided ranging information to optimise the working distance of the probe and hence quality of the Raman signal captured.…”

Real Time Dense Depth Estimation by Fusing Stereo with Sparse Depth Measurements

“…They use lidar measurements as support points for the stereo method mentioned in [9]. Veitch-Michaelis et al also propose a similar approach with a region growing based stereo algorithm [10].…”