Range Versus Surface Denoising of Terrestrial Laser Scanning Data for Rock Discontinuity Roughness Estimation

Maja Bitenc, Daniel Scott Kieffer, Kurosh Khoshelham

Publikation: Beitrag in einer FachzeitschriftArtikelForschungBegutachtung

Abstract

Surface roughness represents a major component of rock discontinuity shear strength. To achieve comprehensive, accurate, and efficient estimates of in situ discontinuity roughness, the traditional contact measuring methods are being replaced by advanced remote-sensing technologies. Terrestrial laser scanner (TLS) is well suited for measuring large inaccessible discontinuities; however, inherent TLS range noise strongly influences the surface details and roughness estimation. The aim of this research is to establish an optimal wavelet-denoising procedure for the TLS data acquired with different scanning configurations (range and incidence angle), and for rock discontinuities having different roughness characteristics and surface reflectivity. The conventional discrete wavelet transform and stationary wavelet transform in combination with four threshold selection methods are applied on TLS data in the direction of range measurements (range denoising) and in the direction perpendicular to the best-fit plane (surface denoising). The performance of the denoising procedures is assessed by comparing the range and surface-denoised TLS surfaces with reference surfaces acquired with the Advanced TOpometric Sensor. Comparative analyses of the roughness calculated according to the angular thresholding method (Grasselli, in Shear strength of rock joints based on quantified surface description, Ph.D. thesis. EPF Lausanne, Lausanne; Grasselli, Shear strength of rock joints based on quantified surface description, Ph.D. thesis, EPF Lausanne, Lausanne, 2001) indicate that all the denoising methods improve the roughness estimated from the TLS data appreciably; however, the level of improvement depends intrinsically on geometrical characteristics of the rock surface and scanning configuration. Range denoising has been found to provide more reliable noise estimations.
Originalspracheenglisch
Seitenumfang15
FachzeitschriftRock Mechanics and Rock Engineering
DOIs
PublikationsstatusElektronische Veröffentlichung vor Drucklegung. - 27 Feb 2019

Fingerprint

roughness
discontinuity
laser
Surface roughness
Rocks
scanner
Scanning
Lasers
rock
Shear strength
shear strength
wavelet
transform
Discrete wavelet transforms
surface roughness
reflectivity
Wavelet transforms
Remote sensing
sensor
remote sensing

Fields of Expertise

  • Sustainable Systems

Dies zitieren

Range Versus Surface Denoising of Terrestrial Laser Scanning Data for Rock Discontinuity Roughness Estimation. / Bitenc, Maja; Kieffer, Daniel Scott; Khoshelham, Kurosh.

in: Rock Mechanics and Rock Engineering, 27.02.2019.

Publikation: Beitrag in einer FachzeitschriftArtikelForschungBegutachtung

@article{5132a0d92cc64812849e073755730557,
title = "Range Versus Surface Denoising of Terrestrial Laser Scanning Data for Rock Discontinuity Roughness Estimation",
abstract = "Surface roughness represents a major component of rock discontinuity shear strength. To achieve comprehensive, accurate, and efficient estimates of in situ discontinuity roughness, the traditional contact measuring methods are being replaced by advanced remote-sensing technologies. Terrestrial laser scanner (TLS) is well suited for measuring large inaccessible discontinuities; however, inherent TLS range noise strongly influences the surface details and roughness estimation. The aim of this research is to establish an optimal wavelet-denoising procedure for the TLS data acquired with different scanning configurations (range and incidence angle), and for rock discontinuities having different roughness characteristics and surface reflectivity. The conventional discrete wavelet transform and stationary wavelet transform in combination with four threshold selection methods are applied on TLS data in the direction of range measurements (range denoising) and in the direction perpendicular to the best-fit plane (surface denoising). The performance of the denoising procedures is assessed by comparing the range and surface-denoised TLS surfaces with reference surfaces acquired with the Advanced TOpometric Sensor. Comparative analyses of the roughness calculated according to the angular thresholding method (Grasselli, in Shear strength of rock joints based on quantified surface description, Ph.D. thesis. EPF Lausanne, Lausanne; Grasselli, Shear strength of rock joints based on quantified surface description, Ph.D. thesis, EPF Lausanne, Lausanne, 2001) indicate that all the denoising methods improve the roughness estimated from the TLS data appreciably; however, the level of improvement depends intrinsically on geometrical characteristics of the rock surface and scanning configuration. Range denoising has been found to provide more reliable noise estimations.",
author = "Maja Bitenc and Kieffer, {Daniel Scott} and Kurosh Khoshelham",
year = "2019",
month = "2",
day = "27",
doi = "10.1007/s00603-019-01755-2",
language = "English",
journal = "Rock Mechanics and Rock Engineering",
issn = "0723-2632",
publisher = "Springer Wien",

}

TY - JOUR

T1 - Range Versus Surface Denoising of Terrestrial Laser Scanning Data for Rock Discontinuity Roughness Estimation

AU - Bitenc, Maja

AU - Kieffer, Daniel Scott

AU - Khoshelham, Kurosh

PY - 2019/2/27

Y1 - 2019/2/27

N2 - Surface roughness represents a major component of rock discontinuity shear strength. To achieve comprehensive, accurate, and efficient estimates of in situ discontinuity roughness, the traditional contact measuring methods are being replaced by advanced remote-sensing technologies. Terrestrial laser scanner (TLS) is well suited for measuring large inaccessible discontinuities; however, inherent TLS range noise strongly influences the surface details and roughness estimation. The aim of this research is to establish an optimal wavelet-denoising procedure for the TLS data acquired with different scanning configurations (range and incidence angle), and for rock discontinuities having different roughness characteristics and surface reflectivity. The conventional discrete wavelet transform and stationary wavelet transform in combination with four threshold selection methods are applied on TLS data in the direction of range measurements (range denoising) and in the direction perpendicular to the best-fit plane (surface denoising). The performance of the denoising procedures is assessed by comparing the range and surface-denoised TLS surfaces with reference surfaces acquired with the Advanced TOpometric Sensor. Comparative analyses of the roughness calculated according to the angular thresholding method (Grasselli, in Shear strength of rock joints based on quantified surface description, Ph.D. thesis. EPF Lausanne, Lausanne; Grasselli, Shear strength of rock joints based on quantified surface description, Ph.D. thesis, EPF Lausanne, Lausanne, 2001) indicate that all the denoising methods improve the roughness estimated from the TLS data appreciably; however, the level of improvement depends intrinsically on geometrical characteristics of the rock surface and scanning configuration. Range denoising has been found to provide more reliable noise estimations.

AB - Surface roughness represents a major component of rock discontinuity shear strength. To achieve comprehensive, accurate, and efficient estimates of in situ discontinuity roughness, the traditional contact measuring methods are being replaced by advanced remote-sensing technologies. Terrestrial laser scanner (TLS) is well suited for measuring large inaccessible discontinuities; however, inherent TLS range noise strongly influences the surface details and roughness estimation. The aim of this research is to establish an optimal wavelet-denoising procedure for the TLS data acquired with different scanning configurations (range and incidence angle), and for rock discontinuities having different roughness characteristics and surface reflectivity. The conventional discrete wavelet transform and stationary wavelet transform in combination with four threshold selection methods are applied on TLS data in the direction of range measurements (range denoising) and in the direction perpendicular to the best-fit plane (surface denoising). The performance of the denoising procedures is assessed by comparing the range and surface-denoised TLS surfaces with reference surfaces acquired with the Advanced TOpometric Sensor. Comparative analyses of the roughness calculated according to the angular thresholding method (Grasselli, in Shear strength of rock joints based on quantified surface description, Ph.D. thesis. EPF Lausanne, Lausanne; Grasselli, Shear strength of rock joints based on quantified surface description, Ph.D. thesis, EPF Lausanne, Lausanne, 2001) indicate that all the denoising methods improve the roughness estimated from the TLS data appreciably; however, the level of improvement depends intrinsically on geometrical characteristics of the rock surface and scanning configuration. Range denoising has been found to provide more reliable noise estimations.

UR - http://www.scopus.com/inward/record.url?scp=85062624821&partnerID=8YFLogxK

U2 - 10.1007/s00603-019-01755-2

DO - 10.1007/s00603-019-01755-2

M3 - Article

JO - Rock Mechanics and Rock Engineering

JF - Rock Mechanics and Rock Engineering

SN - 0723-2632

ER -