ePrints

Calibrated Full-Waveform Airborne Laser Scanning for 3D Object Segmentation

Lookup NU author(s): Fanar ABED, Professor Jon MillsORCiD, Dr Pauline Miller

Downloads

• Published version [.pdf]

Licence

This work is licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0).

Abstract

Segmentation of urban features is considered a major research challenge in the fields of photogrammetry and remote sensing. However, the dense datasets now readily available through airborne laser scanning (ALS) offer increased potential for 3D object segmentation. Such potential is further augmented by the availability of full-waveform (FWF) ALS data. FWF ALS has demonstrated enhanced performance in segmentation and classification through the additional physical observables which can be provided alongside standard geometric information. However, use of FWF information is not recommended without prior radiometric calibration, taking into account all parameters affecting the backscatter energy. This paper reports the implementation of a radiometric calibration workflow for FWF ALS data, and demonstrates how the resultant FWF information can be used to improve segmentation of an urban area. The developed segmentation algorithm presents a novel approach which uses the calibrated backscatter cross-section as a weighting function to estimate the segmentation similarity measure. The normal vector and the local Euclidian distance are used as criteria to segment the point clouds through a region growing approach. The paper demonstrates the potential to enhance 3D object segmentation in urban areas by integrating the FWF physical backscattered energy alongside geometric information. The method is demonstrated through application to an interest area sampled from a relatively dense FWF ALS dataset. The results are assessed through comparison to those delivered from utilising only geometric information. Validation against a manual segmentation demonstrates a successful automatic implementation, achieving a segmentation accuracy of 82%, and out-performs a purely geometric approach.

Publication metadata

Author(s): Abed FM, Mills JP, Miller PE

Publication type: Article

Publication status: Published

Journal: Remote Sensing

Year: 2014

Volume: 6

Issue: 5

Pages: 4109-4132

Print publication date: 02/05/2014

Online publication date: 02/05/2014

Acceptance date: 15/04/2014

Date deposited: 23/07/2014

ISSN (electronic): 2072-4292

Publisher: MDPI

URL: http://dx.doi.org/10.3390/rs6054109

DOI: 10.3390/rs6054109

Altmetrics

Share