A technique to increase the efficiency of artefacts identification in lidar-based canopy height models

ABSTRACT

Canopy height models (CHM) acquired from aerial laser scanning contain local minima (LM) that include artefacts; however, not all LM are artefacts, hence LM were designated as potential artefacts. This article is a part of a new method of removing artefacts or pits from CHM where we focused on improving the efficiency of LM determination as potential artefacts. The minimum filter (MF) was used as the basic tool of LM identification while interpolation was used to increase the number of identified LM. Obtained results have shown that the best interpolation method is bilinear interpolation with a combination of factors: the optimal ratio of the number of successfully defined LM, commission errors and minimum processing time. It has been established that if bilinear interpolation is applied (depth of LM 3–12.9 m), the suggested method can better identify LM compared to the MF without interpolation and might reach up to 22.5%, but not less than 4.3%.

Acknowledgements

The authors would like to thank Professor Arthur P. Cracknell, Editor, Dr Craig Cassells, and two anonymous reviewers for their constructive and helpful comments. We would also like to thank Dr V.G. Pavelyev for scientific discussion.

Additional information

Funding

This work was supported by the Russian Foundation for Basic Research (RFBR), research project No. 14-05-31206.