University of Twente Student Theses


Hybrid adjustment of UAS-based LiDAR and image data

Yogender, . (2022) Hybrid adjustment of UAS-based LiDAR and image data.

[img] PDF
Abstract:Several advancements are going with Unmanned Aerial Systems (UAS) with the addition of multiple sensors and simultaneous data acquisition to obtain detailed geo-data for various applications. However, simultaneous data acquisition with multiple sensors, namely camera, and LiDAR, will also result in possible discrepancies associated with them. These discrepancies must be solved to use a reliable and accurate final product. This research aimed to minimize the discrepancies/errors between the LiDAR and the image data acquired simultaneously with an Unmanned Aerial Systems (UAS) by implementing a hybrid adjustment approach. There can be several discrepancies associated with both the datasets due to the different characteristics of the sensors and the terrain conditions. The initial trajectory of the UAS, raw LiDAR measurements, and image observations were the inputs used for the hybrid adjustment. The UAS trajectory, LiDAR strips, intrinsic calibration of Lidar and camera sensors, and exterior orientations of the images were adjusted and estimated correctly in this hybrid adjustment approach. After hybrid adjustment, both LiDAR and camera-based point clouds are expected to be in the same reference system, with minimal discrepancies between them. In this hybrid adjustment workflow, the discrepancies were minimized with a least-squares-based simultaneous adjustment for both LiDAR and image datasets. For the hybrid adjustment process, three types of correspondences were established, namely: between image pairs (IMG-to-IMG), between LiDAR strips (STR-to-STR), and between image and LiDAR strips (IMG-to-STR). The hybrid adjustment process was experimented with coupled images (coupled to a common LiDAR/image trajectory by the time stamp of images), loose images (not tied to a common LiDAR/image trajectory), and raw LiDAR measurements. We have also experimented with the UAS trajectory correction with bias and linear trajectory correction models in the hybrid adjustment process. After each iteration of hybrid adjustment, a convergence criterion is tested (relative change of the weighted sum of squared errors), and a new iteration cycle starts until a given number of iterations are completed. After hybrid adjustment, a Dense Image Matching (DIM) point cloud was generated with Pix4DMapper using the undistorted images and estimated image orientations from the hybrid adjustment without further optimization of the orientations. For quality control, the relative height difference between the LiDAR and DIM point clouds and Cloud-to-Cloud distances were compared between both the point clouds before and after hybrid adjustment. We also carried out the surface-level analysis of the results to better interpret the errors before and after hybrid adjustment. From the results, it was observed that the most accurate orientation between LiDAR and image data could be obtained by implementing the hybrid adjustment with coupled images and a bias trajectory correction model. It was observed that the alignment between the point clouds has significantly improved from the range of meters to a centimeter-level after implementing the hybrid adjustment process.
Item Type:Essay (Master)
Faculty:ITC: Faculty of Geo-information Science and Earth Observation
Programme:Geoinformation Science and Earth Observation MSc (75014)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page