Rancang Bangun Aplikasi Deteksi Alat Pelindung Diri (APD) untuk Pekerja Proyek dengan Menggunakan Algoritma Yolov5

Jurnal JTIK (Jurnal Teknologi Informasi dan Komunikasi) Pub Date : 2024-04-01 DOI:10.35870/jtik.v8i2.1960

Muhamad Alfin Taufiqurrochman, Herny Februariyanti

{"title":"Rancang Bangun Aplikasi Deteksi Alat Pelindung Diri (APD) untuk Pekerja Proyek dengan Menggunakan Algoritma Yolov5","authors":"Muhamad Alfin Taufiqurrochman, Herny Februariyanti","doi":"10.35870/jtik.v8i2.1960","DOIUrl":null,"url":null,"abstract":"The risk of work accidents that can be experienced by project workers is very high in the world of construction. This can be caused by behavioral factors, one of which is project workers' indiscipline in wearing Personal Protective Equipment (PPE), which can endanger the personal safety of project workers. By utilizing Artificial Intelligence technology, with the computer vision domain, researchers created a PPE detection application using the YoloV5 algorithm. The stage of creating this detection application starts from the process of problem scoping, data acquisition, data exploration, modeling, evaluation and deployment. The dataset used in making this application was taken from Saravana Alagar via Google Drive, covering 4 PPE objects, namely helmets, masks, vests and shoes. By conducting a training dataset of 100 epochs, the percentage results given were very good, namely helmets 96%, vests 96%, masks 95%, and shoes 92%. It is hoped that making this application can minimize cases of work accidents that occur in the project worker area and can increase discipline in using project PPE","PeriodicalId":474679,"journal":{"name":"Jurnal JTIK (Jurnal Teknologi Informasi dan Komunikasi)","volume":"83 ","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Jurnal JTIK (Jurnal Teknologi Informasi dan Komunikasi)","FirstCategoryId":"0","ListUrlMain":"https://doi.org/10.35870/jtik.v8i2.1960","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 0

Abstract

The risk of work accidents that can be experienced by project workers is very high in the world of construction. This can be caused by behavioral factors, one of which is project workers' indiscipline in wearing Personal Protective Equipment (PPE), which can endanger the personal safety of project workers. By utilizing Artificial Intelligence technology, with the computer vision domain, researchers created a PPE detection application using the YoloV5 algorithm. The stage of creating this detection application starts from the process of problem scoping, data acquisition, data exploration, modeling, evaluation and deployment. The dataset used in making this application was taken from Saravana Alagar via Google Drive, covering 4 PPE objects, namely helmets, masks, vests and shoes. By conducting a training dataset of 100 epochs, the percentage results given were very good, namely helmets 96%, vests 96%, masks 95%, and shoes 92%. It is hoped that making this application can minimize cases of work accidents that occur in the project worker area and can increase discipline in using project PPE

查看原文本刊更多论文

利用 Yolov5 算法为项目工人设计个人防护设备 (PPE) 检测应用程序

在建筑领域，项目工人发生工伤事故的风险非常高。这可能是由行为因素造成的，其中之一就是项目工人不按规定穿戴个人防护设备（PPE），这会危及项目工人的人身安全。通过利用人工智能技术和计算机视觉领域，研究人员使用 YoloV5 算法创建了一个个人防护设备检测应用程序。创建该检测应用程序的阶段从问题范围界定、数据采集、数据探索、建模、评估和部署过程开始。用于制作该应用程序的数据集来自 Saravana Alagar 通过 Google Drive 提供的数据集，涵盖 4 个个人防护设备对象，即头盔、面罩、背心和鞋子。通过进行 100 个历元的训练数据集，得出的百分比结果非常好，即头盔 96%、防弹背心 96%、面具 95%、鞋子 92%。希望这一应用能最大限度地减少项目工人区域发生的工伤事故，并能提高使用项目个人防护设备的纪律性。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Jurnal JTIK (Jurnal Teknologi Informasi dan Komunikasi)

自引率

0.00%

发文量