Automatic extraction of blocks from 3D point clouds of fractured rock

Na Chen, John Kemeny, Qinghui Jiang, Zhiwen Pan

Research output: Research - peer-reviewArticle

Abstract

This paper presents a new method for extracting blocks and calculating block size automatically from rock surface 3D point clouds. Block size is an important rock mass characteristic and forms the basis for several rock mass classification schemes. The proposed method consists of four steps: 1) the automatic extraction of discontinuities using an improved Ransac Shape Detection method, 2) the calculation of discontinuity intersections based on plane geometry, 3) the extraction of block candidates based on three discontinuities intersecting one another to form corners, and 4) the identification of “true” blocks using an improved Floodfill algorithm. The calculated block sizes were compared with manual measurements in two case studies, one with fabricated cardboard blocks and the other from an actual rock mass outcrop. The results demonstrate that the proposed method is accurate and overcomes the inaccuracies, safety hazards, and biases of traditional techniques.

LanguageEnglish (US)
Pages149-161
Number of pages13
JournalComputers and Geosciences
Volume109
DOIs
StatePublished - Dec 1 2017

Fingerprint

discontinuity
rock
method
Rocks
rock mass classification
detection method
outcrop
hazard
safety
geometry
calculation
Hazards
Geometry

Keywords

  • 3D point clouds
  • Automatic extraction
  • Block size
  • Discontinuity orientation
  • Rock mass

ASJC Scopus subject areas

  • Information Systems
  • Computers in Earth Sciences

Cite this

Automatic extraction of blocks from 3D point clouds of fractured rock. / Chen, Na; Kemeny, John; Jiang, Qinghui; Pan, Zhiwen.

In: Computers and Geosciences, Vol. 109, 01.12.2017, p. 149-161.

Research output: Research - peer-reviewArticle

Chen, Na ; Kemeny, John ; Jiang, Qinghui ; Pan, Zhiwen. / Automatic extraction of blocks from 3D point clouds of fractured rock. In: Computers and Geosciences. 2017 ; Vol. 109. pp. 149-161
@article{924cd55811f044b49133c25a9e1c8764,
title = "Automatic extraction of blocks from 3D point clouds of fractured rock",
abstract = "This paper presents a new method for extracting blocks and calculating block size automatically from rock surface 3D point clouds. Block size is an important rock mass characteristic and forms the basis for several rock mass classification schemes. The proposed method consists of four steps: 1) the automatic extraction of discontinuities using an improved Ransac Shape Detection method, 2) the calculation of discontinuity intersections based on plane geometry, 3) the extraction of block candidates based on three discontinuities intersecting one another to form corners, and 4) the identification of “true” blocks using an improved Floodfill algorithm. The calculated block sizes were compared with manual measurements in two case studies, one with fabricated cardboard blocks and the other from an actual rock mass outcrop. The results demonstrate that the proposed method is accurate and overcomes the inaccuracies, safety hazards, and biases of traditional techniques.",
keywords = "3D point clouds, Automatic extraction, Block size, Discontinuity orientation, Rock mass",
author = "Na Chen and John Kemeny and Qinghui Jiang and Zhiwen Pan",
year = "2017",
month = "12",
doi = "10.1016/j.cageo.2017.08.013",
volume = "109",
pages = "149--161",
journal = "Computers and Geosciences",
issn = "0098-3004",
publisher = "Elsevier Limited",

}

TY - JOUR

T1 - Automatic extraction of blocks from 3D point clouds of fractured rock

AU - Chen,Na

AU - Kemeny,John

AU - Jiang,Qinghui

AU - Pan,Zhiwen

PY - 2017/12/1

Y1 - 2017/12/1

N2 - This paper presents a new method for extracting blocks and calculating block size automatically from rock surface 3D point clouds. Block size is an important rock mass characteristic and forms the basis for several rock mass classification schemes. The proposed method consists of four steps: 1) the automatic extraction of discontinuities using an improved Ransac Shape Detection method, 2) the calculation of discontinuity intersections based on plane geometry, 3) the extraction of block candidates based on three discontinuities intersecting one another to form corners, and 4) the identification of “true” blocks using an improved Floodfill algorithm. The calculated block sizes were compared with manual measurements in two case studies, one with fabricated cardboard blocks and the other from an actual rock mass outcrop. The results demonstrate that the proposed method is accurate and overcomes the inaccuracies, safety hazards, and biases of traditional techniques.

AB - This paper presents a new method for extracting blocks and calculating block size automatically from rock surface 3D point clouds. Block size is an important rock mass characteristic and forms the basis for several rock mass classification schemes. The proposed method consists of four steps: 1) the automatic extraction of discontinuities using an improved Ransac Shape Detection method, 2) the calculation of discontinuity intersections based on plane geometry, 3) the extraction of block candidates based on three discontinuities intersecting one another to form corners, and 4) the identification of “true” blocks using an improved Floodfill algorithm. The calculated block sizes were compared with manual measurements in two case studies, one with fabricated cardboard blocks and the other from an actual rock mass outcrop. The results demonstrate that the proposed method is accurate and overcomes the inaccuracies, safety hazards, and biases of traditional techniques.

KW - 3D point clouds

KW - Automatic extraction

KW - Block size

KW - Discontinuity orientation

KW - Rock mass

UR - http://www.scopus.com/inward/record.url?scp=85028655660&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85028655660&partnerID=8YFLogxK

U2 - 10.1016/j.cageo.2017.08.013

DO - 10.1016/j.cageo.2017.08.013

M3 - Article

VL - 109

SP - 149

EP - 161

JO - Computers and Geosciences

T2 - Computers and Geosciences

JF - Computers and Geosciences

SN - 0098-3004

ER -