Empirical Evaluation of Terrestrial Laser Scanner Calibration Strategies: Manufacturer-Based, Target-Based and Keypoint-Based

Tomislav Medić, Heiner Kuhlmann, Christoph Holst

Research output: Chapter in Book/Report/Conference proceedingChapterpeer-review

3 Scopus citations

Abstract

To assure accurate terrestrial laser scanner (TLS) point clouds, the instruments should be repeatedly calibrated in the manufacturer’s facilities, either in regular time intervals or if they fail the examination on a test-field. This workflow has drawbacks for the end-users: there is no deeper understanding of the calibration procedure and it is time- and money-wise burdening. As an alternative, scientists developed several user-oriented calibration approaches, which are not well accepted in practice. We investigated the differences between the factory calibration and two user-oriented calibration approaches, as well as their interaction. The investigation focuses on the raw point clouds without factory calibration. Our results show that the user-oriented calibration can replace the manufacturer calibration to some degree. However, the user-oriented calibration on top of the manufacturers’ calibration improves the measurement quality beyond the factory calibration. Hence, user-oriented calibration should be considered as an alternative for the re-calibration to deliver up-to-date calibration parameters.

Original languageEnglish
Title of host publicationSpringer Proceedings in Earth and Environmental Sciences
PublisherSpringer Nature
Pages41-56
Number of pages16
DOIs
StatePublished - 2021
Externally publishedYes

Publication series

NameSpringer Proceedings in Earth and Environmental Sciences
ISSN (Print)2524-342X
ISSN (Electronic)2524-3438

Keywords

  • Accuracy
  • Features
  • Point cloud
  • Quality assurance
  • Systematic errors
  • TLS

Fingerprint

Dive into the research topics of 'Empirical Evaluation of Terrestrial Laser Scanner Calibration Strategies: Manufacturer-Based, Target-Based and Keypoint-Based'. Together they form a unique fingerprint.

Cite this