Geoscience Reference
In-Depth Information
Improvement in Accuracy Through
Self-Calibration for Panoramic Scanner
Mohd Azwan Abbas, Halim Setan, Zulkepli Majid, Albert K. Chong,
Derek D. Lichti and Khairulnizam M. Idris
Abstract Currently, three-dimensional (3D) information has become a necessity
for many purposes especially for documentation, management and analysis. With
the rapid and dense 3D data (point clouds) and considerably at high accuracy has
made terrestrial laser scanner (TLS) widely used for these purposes. However,
similar to other 3D instruments, TLS measurements also cannot escape from the
occurrence of various systematic errors. Through self-calibration, the significant
systematic errors consisted in TLS can be modelled and subsequently removed to
improve the accuracy of the data. To prove that statement, this study has performed
self-calibration for panoramic scanner (Faro Photon 120) at a laboratory with
dimensions of 15.5 m 9 9m9 3 m. By employing optimal network configura-
tion, all 138 well-distributed planar targets were measured from seven scanner
positions to derive four calibration parameters. Statistical analysis (e.g. t test) has
shown that only two parameters, the constant rangefinder offset error (9.3 mm) and
the vertical circle index error (9.4
00
) were significant for the calibrated scanner. To
ensure that self-calibration can improve the accuracy of TLS data, photogrammetric
technique was utilised to establish 15 3D test points at the calibration field. These
test points were used to graphically and statistically demonstrate the improvement
in accuracy between raw data and calibrated data of Faro Photon 120 scanner.
Keywords Terrestrial
laser
scanner
Accuracy
Systematic
errors
Self-calibration
