{"title":"基于改进AlphaPose模型的工程机械位姿估计方法","authors":"Jiayue Zhao, Yunzhong Cao, Yuanzhi Xiang","doi":"10.1108/ecam-05-2022-0476","DOIUrl":null,"url":null,"abstract":"PurposeThe safety management of construction machines is of primary importance. Considering that traditional construction machine safety monitoring and evaluation methods cannot adapt to the complex construction environment, and the monitoring methods based on sensor equipment cost too much. This paper aims to introduce computer vision and deep learning technologies to propose the YOLOv5-FastPose (YFP) model to realize the pose estimation of construction machines by improving the AlphaPose human pose model.Design/methodology/approachThis model introduced the object detection module YOLOv5m to improve the recognition accuracy for detecting construction machines. Meanwhile, to better capture the pose characteristics, the FastPose network optimized feature extraction was introduced into the Single-Machine Pose Estimation Module (SMPE) of AlphaPose. This study used Alberta Construction Image Dataset (ACID) and Construction Equipment Poses Dataset (CEPD) to establish the dataset of object detection and pose estimation of construction machines through data augmentation technology and Labelme image annotation software for training and testing the YFP model.FindingsThe experimental results show that the improved model YFP achieves an average normalization error (NE) of 12.94 × 10–3, an average Percentage of Correct Keypoints (PCK) of 98.48% and an average Area Under the PCK Curve (AUC) of 37.50 × 10–3. Compared with existing methods, this model has higher accuracy in the pose estimation of the construction machine.Originality/valueThis study extends and optimizes the human pose estimation model AlphaPose to make it suitable for construction machines, improving the performance of pose estimation for construction machines.","PeriodicalId":11888,"journal":{"name":"Engineering, Construction and Architectural Management","volume":null,"pages":null},"PeriodicalIF":3.6000,"publicationDate":"2022-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":"{\"title\":\"Pose estimation method for construction machine based on improved AlphaPose model\",\"authors\":\"Jiayue Zhao, Yunzhong Cao, Yuanzhi Xiang\",\"doi\":\"10.1108/ecam-05-2022-0476\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"PurposeThe safety management of construction machines is of primary importance. Considering that traditional construction machine safety monitoring and evaluation methods cannot adapt to the complex construction environment, and the monitoring methods based on sensor equipment cost too much. This paper aims to introduce computer vision and deep learning technologies to propose the YOLOv5-FastPose (YFP) model to realize the pose estimation of construction machines by improving the AlphaPose human pose model.Design/methodology/approachThis model introduced the object detection module YOLOv5m to improve the recognition accuracy for detecting construction machines. Meanwhile, to better capture the pose characteristics, the FastPose network optimized feature extraction was introduced into the Single-Machine Pose Estimation Module (SMPE) of AlphaPose. This study used Alberta Construction Image Dataset (ACID) and Construction Equipment Poses Dataset (CEPD) to establish the dataset of object detection and pose estimation of construction machines through data augmentation technology and Labelme image annotation software for training and testing the YFP model.FindingsThe experimental results show that the improved model YFP achieves an average normalization error (NE) of 12.94 × 10–3, an average Percentage of Correct Keypoints (PCK) of 98.48% and an average Area Under the PCK Curve (AUC) of 37.50 × 10–3. Compared with existing methods, this model has higher accuracy in the pose estimation of the construction machine.Originality/valueThis study extends and optimizes the human pose estimation model AlphaPose to make it suitable for construction machines, improving the performance of pose estimation for construction machines.\",\"PeriodicalId\":11888,\"journal\":{\"name\":\"Engineering, Construction and Architectural Management\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":3.6000,\"publicationDate\":\"2022-10-17\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"2\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Engineering, Construction and Architectural Management\",\"FirstCategoryId\":\"5\",\"ListUrlMain\":\"https://doi.org/10.1108/ecam-05-2022-0476\",\"RegionNum\":2,\"RegionCategory\":\"工程技术\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"ENGINEERING, CIVIL\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Engineering, Construction and Architectural Management","FirstCategoryId":"5","ListUrlMain":"https://doi.org/10.1108/ecam-05-2022-0476","RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENGINEERING, CIVIL","Score":null,"Total":0}
Pose estimation method for construction machine based on improved AlphaPose model
PurposeThe safety management of construction machines is of primary importance. Considering that traditional construction machine safety monitoring and evaluation methods cannot adapt to the complex construction environment, and the monitoring methods based on sensor equipment cost too much. This paper aims to introduce computer vision and deep learning technologies to propose the YOLOv5-FastPose (YFP) model to realize the pose estimation of construction machines by improving the AlphaPose human pose model.Design/methodology/approachThis model introduced the object detection module YOLOv5m to improve the recognition accuracy for detecting construction machines. Meanwhile, to better capture the pose characteristics, the FastPose network optimized feature extraction was introduced into the Single-Machine Pose Estimation Module (SMPE) of AlphaPose. This study used Alberta Construction Image Dataset (ACID) and Construction Equipment Poses Dataset (CEPD) to establish the dataset of object detection and pose estimation of construction machines through data augmentation technology and Labelme image annotation software for training and testing the YFP model.FindingsThe experimental results show that the improved model YFP achieves an average normalization error (NE) of 12.94 × 10–3, an average Percentage of Correct Keypoints (PCK) of 98.48% and an average Area Under the PCK Curve (AUC) of 37.50 × 10–3. Compared with existing methods, this model has higher accuracy in the pose estimation of the construction machine.Originality/valueThis study extends and optimizes the human pose estimation model AlphaPose to make it suitable for construction machines, improving the performance of pose estimation for construction machines.
期刊介绍:
ECAM publishes original peer-reviewed research papers, case studies, technical notes, book reviews, features, discussions and other contemporary articles that advance research and practice in engineering, construction and architectural management. In particular, ECAM seeks to advance integrated design and construction practices, project lifecycle management, and sustainable construction. The journal’s scope covers all aspects of architectural design, design management, construction/project management, engineering management of major infrastructure projects, and the operation and management of constructed facilities. ECAM also addresses the technological, process, economic/business, environmental/sustainability, political, and social/human developments that influence the construction project delivery process.
ECAM strives to establish strong theoretical and empirical debates in the above areas of engineering, architecture, and construction research. Papers should be heavily integrated with the existing and current body of knowledge within the field and develop explicit and novel contributions. Acknowledging the global character of the field, we welcome papers on regional studies but encourage authors to position the work within the broader international context by reviewing and comparing findings from their regional study with studies conducted in other regions or countries whenever possible.