Vision-based detection of unsafe actions of a construction worker: Case study of ladder climbing

SangUk Han, Sang Hyun Lee, Feniosky Peña-Mora

Research output: Contribution to journalArticle

53 Citations (Scopus)

Abstract

About 80-90% of accidents are caused by the unsafe actions and behaviors of employees in construction. Behavior management thus plays a key role in enhancing safety, and particularly, behavior observation is the most critical element for modifying workers' behavior in a safe manner. However, there is a lack of practical methods to measure workers' behavior in construction. To analyze workers' actions, this paper uses an advanced and economical depth sensor to collect motion data and then investigates consequent motion-analysis techniques to detect the unsafe actions of workers, which is the main focus of this paper. First, motion data are transformed onto a three-dimensional (3D) space as a preprocess, motion classification is performed to identify a typical prior, and the selected prior is used to detect the same action in a testing data set. As a case study, motion data for unsafe actions in ladder climbing (i.e., backward-facing climbing, climbing with an object, and reaching far to a side) are collected and used to detect the actions in a new testing data set in which the actions are randomly taken. The result shows that 90.91% of unsafe actions are correctly detected in the experiment.

Original languageEnglish
Pages (from-to)635-644
Number of pages10
JournalJournal of Computing in Civil Engineering
Volume27
Issue number6
DOIs
StatePublished - 2013 Nov 1

Fingerprint

Ladders
Testing
Accidents
Personnel
Sensors
Experiments
Motion analysis

Keywords

  • Behavior observation
  • Dimension reduction
  • Motion classification
  • Motion recognition
  • Motion sensor
  • Safety

Cite this

@article{a2e146bd88fb42e7ac980994b1c33848,
title = "Vision-based detection of unsafe actions of a construction worker: Case study of ladder climbing",
abstract = "About 80-90{\%} of accidents are caused by the unsafe actions and behaviors of employees in construction. Behavior management thus plays a key role in enhancing safety, and particularly, behavior observation is the most critical element for modifying workers' behavior in a safe manner. However, there is a lack of practical methods to measure workers' behavior in construction. To analyze workers' actions, this paper uses an advanced and economical depth sensor to collect motion data and then investigates consequent motion-analysis techniques to detect the unsafe actions of workers, which is the main focus of this paper. First, motion data are transformed onto a three-dimensional (3D) space as a preprocess, motion classification is performed to identify a typical prior, and the selected prior is used to detect the same action in a testing data set. As a case study, motion data for unsafe actions in ladder climbing (i.e., backward-facing climbing, climbing with an object, and reaching far to a side) are collected and used to detect the actions in a new testing data set in which the actions are randomly taken. The result shows that 90.91{\%} of unsafe actions are correctly detected in the experiment.",
keywords = "Behavior observation, Dimension reduction, Motion classification, Motion recognition, Motion sensor, Safety",
author = "SangUk Han and Lee, {Sang Hyun} and Feniosky Pe{\~n}a-Mora",
year = "2013",
month = "11",
day = "1",
doi = "10.1061/(ASCE)CP.1943-5487.0000279",
language = "English",
volume = "27",
pages = "635--644",
journal = "Journal of Computing in Civil Engineering",
issn = "0887-3801",
number = "6",

}

Vision-based detection of unsafe actions of a construction worker : Case study of ladder climbing. / Han, SangUk; Lee, Sang Hyun; Peña-Mora, Feniosky.

In: Journal of Computing in Civil Engineering, Vol. 27, No. 6, 01.11.2013, p. 635-644.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Vision-based detection of unsafe actions of a construction worker

T2 - Case study of ladder climbing

AU - Han, SangUk

AU - Lee, Sang Hyun

AU - Peña-Mora, Feniosky

PY - 2013/11/1

Y1 - 2013/11/1

N2 - About 80-90% of accidents are caused by the unsafe actions and behaviors of employees in construction. Behavior management thus plays a key role in enhancing safety, and particularly, behavior observation is the most critical element for modifying workers' behavior in a safe manner. However, there is a lack of practical methods to measure workers' behavior in construction. To analyze workers' actions, this paper uses an advanced and economical depth sensor to collect motion data and then investigates consequent motion-analysis techniques to detect the unsafe actions of workers, which is the main focus of this paper. First, motion data are transformed onto a three-dimensional (3D) space as a preprocess, motion classification is performed to identify a typical prior, and the selected prior is used to detect the same action in a testing data set. As a case study, motion data for unsafe actions in ladder climbing (i.e., backward-facing climbing, climbing with an object, and reaching far to a side) are collected and used to detect the actions in a new testing data set in which the actions are randomly taken. The result shows that 90.91% of unsafe actions are correctly detected in the experiment.

AB - About 80-90% of accidents are caused by the unsafe actions and behaviors of employees in construction. Behavior management thus plays a key role in enhancing safety, and particularly, behavior observation is the most critical element for modifying workers' behavior in a safe manner. However, there is a lack of practical methods to measure workers' behavior in construction. To analyze workers' actions, this paper uses an advanced and economical depth sensor to collect motion data and then investigates consequent motion-analysis techniques to detect the unsafe actions of workers, which is the main focus of this paper. First, motion data are transformed onto a three-dimensional (3D) space as a preprocess, motion classification is performed to identify a typical prior, and the selected prior is used to detect the same action in a testing data set. As a case study, motion data for unsafe actions in ladder climbing (i.e., backward-facing climbing, climbing with an object, and reaching far to a side) are collected and used to detect the actions in a new testing data set in which the actions are randomly taken. The result shows that 90.91% of unsafe actions are correctly detected in the experiment.

KW - Behavior observation

KW - Dimension reduction

KW - Motion classification

KW - Motion recognition

KW - Motion sensor

KW - Safety

UR - http://www.scopus.com/inward/record.url?scp=84886485512&partnerID=8YFLogxK

U2 - 10.1061/(ASCE)CP.1943-5487.0000279

DO - 10.1061/(ASCE)CP.1943-5487.0000279

M3 - Article

AN - SCOPUS:84886485512

VL - 27

SP - 635

EP - 644

JO - Journal of Computing in Civil Engineering

JF - Journal of Computing in Civil Engineering

SN - 0887-3801

IS - 6

ER -