Real-time monitoring of work-at-height safety hazards in construction sites using drones and deep learning

Introduction: The construction field is considered one of the most dangerous industries. Accidents and fatalities take place on a daily basis in construction projects. Globally, different levels of government have implemented strict rules and regulations to protect workers on job sites. However, des...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Journal of safety research 2022-12, Vol.83, p.364-370
Hauptverfasser: Shanti, Mohammad Z., Cho, Chung-Suk, de Soto, Borja Garcia, Byon, Young-Ji, Yeun, Chan Yeob, Kim, Tae Yeon
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page 370
container_issue
container_start_page 364
container_title Journal of safety research
container_volume 83
creator Shanti, Mohammad Z.
Cho, Chung-Suk
de Soto, Borja Garcia
Byon, Young-Ji
Yeun, Chan Yeob
Kim, Tae Yeon
description Introduction: The construction field is considered one of the most dangerous industries. Accidents and fatalities take place on a daily basis in construction projects. Globally, different levels of government have implemented strict rules and regulations to protect workers on job sites. However, despite the efforts to implement the rules and regulations, accidents occur frequently. Falling from heights is considered the most common cause of death in construction. This study developed a novel system integrating deep learning and drones to monitor workers in real-time when performing at-height activities. Method: Specifically, a pre-trained deep learning model was used to detect Personal Fall Arrest System components (e.g., safety harness, lifeline, and helmet). The drone was utilized to take images and videos from the construction site, and the data were relayed to the model to detect safety violations. The system was tested and validated in real construction sites and in a controlled lab environment to verify the model’s effectiveness under different light and weather conditions. Results: The overall accuracy of the system was 90%. The model’s precision and recall were 97.2 % and 90.2%, respectively. The average time taken to detect a violation was around 12 seconds. Conclusions: Moreover, the Area Under Curve - Receiver Operating Characteristics chart showed that the trained model was very good and precise in detecting and differentiating the desired objects. Practical Applications: This fast, reliable, and economical system can aid in saving many lives if implemented and utilized properly in real construction sites.
doi_str_mv 10.1016/j.jsr.2022.09.011
format Article
fullrecord <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_miscellaneous_2753291504</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><els_id>S0022437522001475</els_id><sourcerecordid>2753291504</sourcerecordid><originalsourceid>FETCH-LOGICAL-c283t-334925933d6582543c756513cabfb7a5bec124dcc3c18329680bd363e484bd173</originalsourceid><addsrcrecordid>eNp9kEtr3DAUhUVp6EwePyCbomU3dvWw_KCrEtIkMBAIyVrI0vWMHFuaSnJL-uujYaZddnW53HMO93wIXVNSUkLrr2M5xlAywlhJupJQ-gGtadt0BWVV8xGtSb4UFW_ECp3HOBJCakHpJ7TiddVSwro1Gp9ATUWyM-DZO5t8sG6L_YB_-_BaqFTswG53CUc1QHrDO_VHBROxdVh7F1NYdLLe4WgTRLzEg9kE7_KinMEGYI8nUMHlwyU6G9QU4eo0L9DLj9vnm_ti83j3cPN9U2jW8lRwXnVMdJybWrRMVFw3Ir_NteqHvlGiB53rGa25pi1nXd2S3vCaQ9VWvaENv0Bfjrn74H8uEJOcbdQwTcqBX6Jkjcg2KkiVpfQo1cHHGGCQ-2BnFd4kJfKAWI4yI5YHxJJ0MiPOns-n-KWfwfxz_GWaBd-OAsglf1kIMmoLToOxAXSSxtv_xL8DufuM3g</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2753291504</pqid></control><display><type>article</type><title>Real-time monitoring of work-at-height safety hazards in construction sites using drones and deep learning</title><source>MEDLINE</source><source>Access via ScienceDirect (Elsevier)</source><creator>Shanti, Mohammad Z. ; Cho, Chung-Suk ; de Soto, Borja Garcia ; Byon, Young-Ji ; Yeun, Chan Yeob ; Kim, Tae Yeon</creator><creatorcontrib>Shanti, Mohammad Z. ; Cho, Chung-Suk ; de Soto, Borja Garcia ; Byon, Young-Ji ; Yeun, Chan Yeob ; Kim, Tae Yeon</creatorcontrib><description>Introduction: The construction field is considered one of the most dangerous industries. Accidents and fatalities take place on a daily basis in construction projects. Globally, different levels of government have implemented strict rules and regulations to protect workers on job sites. However, despite the efforts to implement the rules and regulations, accidents occur frequently. Falling from heights is considered the most common cause of death in construction. This study developed a novel system integrating deep learning and drones to monitor workers in real-time when performing at-height activities. Method: Specifically, a pre-trained deep learning model was used to detect Personal Fall Arrest System components (e.g., safety harness, lifeline, and helmet). The drone was utilized to take images and videos from the construction site, and the data were relayed to the model to detect safety violations. The system was tested and validated in real construction sites and in a controlled lab environment to verify the model’s effectiveness under different light and weather conditions. Results: The overall accuracy of the system was 90%. The model’s precision and recall were 97.2 % and 90.2%, respectively. The average time taken to detect a violation was around 12 seconds. Conclusions: Moreover, the Area Under Curve - Receiver Operating Characteristics chart showed that the trained model was very good and precise in detecting and differentiating the desired objects. Practical Applications: This fast, reliable, and economical system can aid in saving many lives if implemented and utilized properly in real construction sites.</description><identifier>ISSN: 0022-4375</identifier><identifier>EISSN: 1879-1247</identifier><identifier>DOI: 10.1016/j.jsr.2022.09.011</identifier><identifier>PMID: 36481029</identifier><language>eng</language><publisher>United States: Elsevier Ltd</publisher><subject>Deep Learning ; Fall from heights ; Humans ; Law Enforcement ; Machine learning ; Personal Fall Arrest System (PFAS) ; Real-time detection ; Unmanned Aerial Vehicles (UAV) ; Workplace</subject><ispartof>Journal of safety research, 2022-12, Vol.83, p.364-370</ispartof><rights>2022 National Safety Council and Elsevier Ltd</rights><rights>Copyright © 2022 National Safety Council and Elsevier Ltd. All rights reserved.</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c283t-334925933d6582543c756513cabfb7a5bec124dcc3c18329680bd363e484bd173</citedby><cites>FETCH-LOGICAL-c283t-334925933d6582543c756513cabfb7a5bec124dcc3c18329680bd363e484bd173</cites><orcidid>0000-0002-6112-7967 ; 0000-0003-1209-1803</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://dx.doi.org/10.1016/j.jsr.2022.09.011$$EHTML$$P50$$Gelsevier$$H</linktohtml><link.rule.ids>314,780,784,3550,27924,27925,45995</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/36481029$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Shanti, Mohammad Z.</creatorcontrib><creatorcontrib>Cho, Chung-Suk</creatorcontrib><creatorcontrib>de Soto, Borja Garcia</creatorcontrib><creatorcontrib>Byon, Young-Ji</creatorcontrib><creatorcontrib>Yeun, Chan Yeob</creatorcontrib><creatorcontrib>Kim, Tae Yeon</creatorcontrib><title>Real-time monitoring of work-at-height safety hazards in construction sites using drones and deep learning</title><title>Journal of safety research</title><addtitle>J Safety Res</addtitle><description>Introduction: The construction field is considered one of the most dangerous industries. Accidents and fatalities take place on a daily basis in construction projects. Globally, different levels of government have implemented strict rules and regulations to protect workers on job sites. However, despite the efforts to implement the rules and regulations, accidents occur frequently. Falling from heights is considered the most common cause of death in construction. This study developed a novel system integrating deep learning and drones to monitor workers in real-time when performing at-height activities. Method: Specifically, a pre-trained deep learning model was used to detect Personal Fall Arrest System components (e.g., safety harness, lifeline, and helmet). The drone was utilized to take images and videos from the construction site, and the data were relayed to the model to detect safety violations. The system was tested and validated in real construction sites and in a controlled lab environment to verify the model’s effectiveness under different light and weather conditions. Results: The overall accuracy of the system was 90%. The model’s precision and recall were 97.2 % and 90.2%, respectively. The average time taken to detect a violation was around 12 seconds. Conclusions: Moreover, the Area Under Curve - Receiver Operating Characteristics chart showed that the trained model was very good and precise in detecting and differentiating the desired objects. Practical Applications: This fast, reliable, and economical system can aid in saving many lives if implemented and utilized properly in real construction sites.</description><subject>Deep Learning</subject><subject>Fall from heights</subject><subject>Humans</subject><subject>Law Enforcement</subject><subject>Machine learning</subject><subject>Personal Fall Arrest System (PFAS)</subject><subject>Real-time detection</subject><subject>Unmanned Aerial Vehicles (UAV)</subject><subject>Workplace</subject><issn>0022-4375</issn><issn>1879-1247</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2022</creationdate><recordtype>article</recordtype><sourceid>EIF</sourceid><recordid>eNp9kEtr3DAUhUVp6EwePyCbomU3dvWw_KCrEtIkMBAIyVrI0vWMHFuaSnJL-uujYaZddnW53HMO93wIXVNSUkLrr2M5xlAywlhJupJQ-gGtadt0BWVV8xGtSb4UFW_ECp3HOBJCakHpJ7TiddVSwro1Gp9ATUWyM-DZO5t8sG6L_YB_-_BaqFTswG53CUc1QHrDO_VHBROxdVh7F1NYdLLe4WgTRLzEg9kE7_KinMEGYI8nUMHlwyU6G9QU4eo0L9DLj9vnm_ti83j3cPN9U2jW8lRwXnVMdJybWrRMVFw3Ir_NteqHvlGiB53rGa25pi1nXd2S3vCaQ9VWvaENv0Bfjrn74H8uEJOcbdQwTcqBX6Jkjcg2KkiVpfQo1cHHGGCQ-2BnFd4kJfKAWI4yI5YHxJJ0MiPOns-n-KWfwfxz_GWaBd-OAsglf1kIMmoLToOxAXSSxtv_xL8DufuM3g</recordid><startdate>202212</startdate><enddate>202212</enddate><creator>Shanti, Mohammad Z.</creator><creator>Cho, Chung-Suk</creator><creator>de Soto, Borja Garcia</creator><creator>Byon, Young-Ji</creator><creator>Yeun, Chan Yeob</creator><creator>Kim, Tae Yeon</creator><general>Elsevier Ltd</general><scope>CGR</scope><scope>CUY</scope><scope>CVF</scope><scope>ECM</scope><scope>EIF</scope><scope>NPM</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7X8</scope><orcidid>https://orcid.org/0000-0002-6112-7967</orcidid><orcidid>https://orcid.org/0000-0003-1209-1803</orcidid></search><sort><creationdate>202212</creationdate><title>Real-time monitoring of work-at-height safety hazards in construction sites using drones and deep learning</title><author>Shanti, Mohammad Z. ; Cho, Chung-Suk ; de Soto, Borja Garcia ; Byon, Young-Ji ; Yeun, Chan Yeob ; Kim, Tae Yeon</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c283t-334925933d6582543c756513cabfb7a5bec124dcc3c18329680bd363e484bd173</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2022</creationdate><topic>Deep Learning</topic><topic>Fall from heights</topic><topic>Humans</topic><topic>Law Enforcement</topic><topic>Machine learning</topic><topic>Personal Fall Arrest System (PFAS)</topic><topic>Real-time detection</topic><topic>Unmanned Aerial Vehicles (UAV)</topic><topic>Workplace</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Shanti, Mohammad Z.</creatorcontrib><creatorcontrib>Cho, Chung-Suk</creatorcontrib><creatorcontrib>de Soto, Borja Garcia</creatorcontrib><creatorcontrib>Byon, Young-Ji</creatorcontrib><creatorcontrib>Yeun, Chan Yeob</creatorcontrib><creatorcontrib>Kim, Tae Yeon</creatorcontrib><collection>Medline</collection><collection>MEDLINE</collection><collection>MEDLINE (Ovid)</collection><collection>MEDLINE</collection><collection>MEDLINE</collection><collection>PubMed</collection><collection>CrossRef</collection><collection>MEDLINE - Academic</collection><jtitle>Journal of safety research</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Shanti, Mohammad Z.</au><au>Cho, Chung-Suk</au><au>de Soto, Borja Garcia</au><au>Byon, Young-Ji</au><au>Yeun, Chan Yeob</au><au>Kim, Tae Yeon</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Real-time monitoring of work-at-height safety hazards in construction sites using drones and deep learning</atitle><jtitle>Journal of safety research</jtitle><addtitle>J Safety Res</addtitle><date>2022-12</date><risdate>2022</risdate><volume>83</volume><spage>364</spage><epage>370</epage><pages>364-370</pages><issn>0022-4375</issn><eissn>1879-1247</eissn><abstract>Introduction: The construction field is considered one of the most dangerous industries. Accidents and fatalities take place on a daily basis in construction projects. Globally, different levels of government have implemented strict rules and regulations to protect workers on job sites. However, despite the efforts to implement the rules and regulations, accidents occur frequently. Falling from heights is considered the most common cause of death in construction. This study developed a novel system integrating deep learning and drones to monitor workers in real-time when performing at-height activities. Method: Specifically, a pre-trained deep learning model was used to detect Personal Fall Arrest System components (e.g., safety harness, lifeline, and helmet). The drone was utilized to take images and videos from the construction site, and the data were relayed to the model to detect safety violations. The system was tested and validated in real construction sites and in a controlled lab environment to verify the model’s effectiveness under different light and weather conditions. Results: The overall accuracy of the system was 90%. The model’s precision and recall were 97.2 % and 90.2%, respectively. The average time taken to detect a violation was around 12 seconds. Conclusions: Moreover, the Area Under Curve - Receiver Operating Characteristics chart showed that the trained model was very good and precise in detecting and differentiating the desired objects. Practical Applications: This fast, reliable, and economical system can aid in saving many lives if implemented and utilized properly in real construction sites.</abstract><cop>United States</cop><pub>Elsevier Ltd</pub><pmid>36481029</pmid><doi>10.1016/j.jsr.2022.09.011</doi><tpages>7</tpages><orcidid>https://orcid.org/0000-0002-6112-7967</orcidid><orcidid>https://orcid.org/0000-0003-1209-1803</orcidid></addata></record>
fulltext fulltext
identifier ISSN: 0022-4375
ispartof Journal of safety research, 2022-12, Vol.83, p.364-370
issn 0022-4375
1879-1247
language eng
recordid cdi_proquest_miscellaneous_2753291504
source MEDLINE; Access via ScienceDirect (Elsevier)
subjects Deep Learning
Fall from heights
Humans
Law Enforcement
Machine learning
Personal Fall Arrest System (PFAS)
Real-time detection
Unmanned Aerial Vehicles (UAV)
Workplace
title Real-time monitoring of work-at-height safety hazards in construction sites using drones and deep learning
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-23T10%3A33%3A42IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Real-time%20monitoring%20of%20work-at-height%20safety%20hazards%20in%20construction%20sites%20using%20drones%20and%20deep%20learning&rft.jtitle=Journal%20of%20safety%20research&rft.au=Shanti,%20Mohammad%20Z.&rft.date=2022-12&rft.volume=83&rft.spage=364&rft.epage=370&rft.pages=364-370&rft.issn=0022-4375&rft.eissn=1879-1247&rft_id=info:doi/10.1016/j.jsr.2022.09.011&rft_dat=%3Cproquest_cross%3E2753291504%3C/proquest_cross%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2753291504&rft_id=info:pmid/36481029&rft_els_id=S0022437522001475&rfr_iscdi=true