Yang Wang, Hong Xiao, Zhihai Zhang, Xiaoxuan Guo, Qiang Liu
The noise within train is a paradox; while harmful to passenger health, it is useful to operators as it provides insights into the working status of vehicles and tracks. Recently, methods for identifying defects based on interior noise signals are emerging, among which representation learning is the foundation for deep neural network models to understand the key information and structure of the data. To provide foundational data for track fault detection, a representation learning framework for interior noise, named the interior noise representation framework, is introduced. The method includes: (i) using wavelet transform to represent the original noise signal and designing a soft and hard denoising module for dataset denoising; (ii) deep residual convolutional denoising variational autoencoder (VAE) module performs representation learning with a VAE and deep residual convolutional neural networks, enabling richer data augmentation for sparsely labeled samples by manipulating the embedding space; (iii) deep embedding clustering submodule balances the representation of reconstruction and clustering features through the joint optimization of these aspects, categorizing metro noise into three distinct classes and effectively discriminating significantly different features. The experimental results show that, compared to traditional mechanism‐based models for characterizing interior noise, this approach offers a data‐driven general analysis framework, providing a foundational model for downstream tasks.
{"title":"Self‐supervised representation learning of metro interior noise based on variational autoencoder and deep embedding clustering","authors":"Yang Wang, Hong Xiao, Zhihai Zhang, Xiaoxuan Guo, Qiang Liu","doi":"10.1111/mice.13336","DOIUrl":"https://doi.org/10.1111/mice.13336","url":null,"abstract":"The noise within train is a paradox; while harmful to passenger health, it is useful to operators as it provides insights into the working status of vehicles and tracks. Recently, methods for identifying defects based on interior noise signals are emerging, among which representation learning is the foundation for deep neural network models to understand the key information and structure of the data. To provide foundational data for track fault detection, a representation learning framework for interior noise, named the interior noise representation framework, is introduced. The method includes: (i) using wavelet transform to represent the original noise signal and designing a soft and hard denoising module for dataset denoising; (ii) deep residual convolutional denoising variational autoencoder (VAE) module performs representation learning with a VAE and deep residual convolutional neural networks, enabling richer data augmentation for sparsely labeled samples by manipulating the embedding space; (iii) deep embedding clustering submodule balances the representation of reconstruction and clustering features through the joint optimization of these aspects, categorizing metro noise into three distinct classes and effectively discriminating significantly different features. The experimental results show that, compared to traditional mechanism‐based models for characterizing interior noise, this approach offers a data‐driven general analysis framework, providing a foundational model for downstream tasks.","PeriodicalId":156,"journal":{"name":"Computer-Aided Civil and Infrastructure Engineering","volume":"22 1","pages":""},"PeriodicalIF":11.775,"publicationDate":"2024-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142160481","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper presents the implementation of a novel monitoring system in which video images and conventional sensor network data are simultaneously analyzed to identify the structural flexibility from the ambient vibrations. The magnitude ratio between the flexibility estimated from known/unknown input force are theoretically derived and decomposed into two parts: and . The first scale factor related to basic modal parameters can be acquired using the general modal identification methods. Aiming to tackle the difficulty in identifying the second scale factor related to the force intensity, a video stream of traffic is processed to detect and classify vehicles to determine the vehicle's location while displacement measurements are simultaneously collected. By integrating the toll station data, the vehicle loads are assigned to the vehicle on the bridge deck through the uniqueness of the license plate number. Thus, a structural input–output relationship is established to solve the second scale factor . Finally, the flexibility estimated from the ambient vibration are scaled by and , respectively to obtain the exact flexibility , which are same as the analytical ones . Both numerical example and a laboratory test are performed to demonstrate the accuracy of the proposed methodology. The algorithms, approaches, and results given in the paper demonstrate its effectiveness and shows great potential for its application on a real‐life bridge's condition assessment.
{"title":"A computer vision–aided methodology for bridge flexibility identification from ambient vibrations","authors":"Yuyao Cheng, Siqi Jia, Jianliang Zhang, Jian Zhang","doi":"10.1111/mice.13329","DOIUrl":"https://doi.org/10.1111/mice.13329","url":null,"abstract":"This paper presents the implementation of a novel monitoring system in which video images and conventional sensor network data are simultaneously analyzed to identify the structural flexibility from the ambient vibrations. The magnitude ratio between the flexibility estimated from known/unknown input force are theoretically derived and decomposed into two parts: and . The first scale factor related to basic modal parameters can be acquired using the general modal identification methods. Aiming to tackle the difficulty in identifying the second scale factor related to the force intensity, a video stream of traffic is processed to detect and classify vehicles to determine the vehicle's location while displacement measurements are simultaneously collected. By integrating the toll station data, the vehicle loads are assigned to the vehicle on the bridge deck through the uniqueness of the license plate number. Thus, a structural input–output relationship is established to solve the second scale factor . Finally, the flexibility estimated from the ambient vibration are scaled by and , respectively to obtain the exact flexibility , which are same as the analytical ones . Both numerical example and a laboratory test are performed to demonstrate the accuracy of the proposed methodology. The algorithms, approaches, and results given in the paper demonstrate its effectiveness and shows great potential for its application on a real‐life bridge's condition assessment.","PeriodicalId":156,"journal":{"name":"Computer-Aided Civil and Infrastructure Engineering","volume":"17 1","pages":""},"PeriodicalIF":11.775,"publicationDate":"2024-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142131013","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Recent studies in dam displacement monitoring primarily focus on single‐response monitoring or model updating using advanced techniques. Few studies involve the combination analysis of displacement with other synchronized responses utilizing their monitoring characteristics. In situ strain data provide a strength‐safety perspective for dam displacement monitoring. The challenge lies in that estimating displacement directly using limited discrete strain data may be misleading. This paper analyzes the relationship between displacement and global, and multipoint local strains from the perspective of the differences in load effects of gravity dams, and indicates that introducing appropriate state factors improves the estimation. A displacement estimation model driven by strain data and state factors is developed using stacked convolutional neural network, and the variable relationships within the model are interpretated via accumulated local effects. Incorporating specific strength criteria, a novel displacement monitoring indicator based on the tensile safety of the dam heel is proposed. A case study of a gravity dam showcases the effectiveness of the proposed approach in comparison with the solely strain‐based model and the traditional hydrostatic‐seasonal‐time factors‐based model.
{"title":"Gravity dam displacement monitoring using in situ strain and deep learning","authors":"Xin Wu, Dongjian Zheng, Xingqiao Chen, Yongtao Liu, Jianchun Qiu, Haifeng Jiang","doi":"10.1111/mice.13333","DOIUrl":"https://doi.org/10.1111/mice.13333","url":null,"abstract":"Recent studies in dam displacement monitoring primarily focus on single‐response monitoring or model updating using advanced techniques. Few studies involve the combination analysis of displacement with other synchronized responses utilizing their monitoring characteristics. In situ strain data provide a strength‐safety perspective for dam displacement monitoring. The challenge lies in that estimating displacement directly using limited discrete strain data may be misleading. This paper analyzes the relationship between displacement and global, and multipoint local strains from the perspective of the differences in load effects of gravity dams, and indicates that introducing appropriate state factors improves the estimation. A displacement estimation model driven by strain data and state factors is developed using stacked convolutional neural network, and the variable relationships within the model are interpretated via accumulated local effects. Incorporating specific strength criteria, a novel displacement monitoring indicator based on the tensile safety of the dam heel is proposed. A case study of a gravity dam showcases the effectiveness of the proposed approach in comparison with the solely strain‐based model and the traditional hydrostatic‐seasonal‐time factors‐based model.","PeriodicalId":156,"journal":{"name":"Computer-Aided Civil and Infrastructure Engineering","volume":"100 1","pages":""},"PeriodicalIF":11.775,"publicationDate":"2024-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142117977","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The cover image is based on the Article A traffic state prediction method based on spatial–temporal data mining of floating car data by using autoformer architecture by Shuangzhi Yu et al., https://doi.org/10.1111/mice.13179.
封面图像基于余双志等人撰写的文章《一种基于时空数据挖掘的浮动车数据交通状态预测方法》（A Traffic State prediction method based on spatial-temporal data mining of floating car data by using autoformer architecture），https://doi.org/10.1111/mice.13179。
{"title":"Cover Image, Volume 39, Issue 18","authors":"","doi":"10.1111/mice.13335","DOIUrl":"10.1111/mice.13335","url":null,"abstract":"<p><b>The cover image</b> is based on the Article <i>A traffic state prediction method based on spatial–temporal data mining of floating car data by using autoformer architecture</i> by Shuangzhi Yu et al., https://doi.org/10.1111/mice.13179.\u0000\u0000 <figure>\u0000 <div><picture>\u0000 <source></source></picture><p></p>\u0000 </div>\u0000 </figure>\u0000 </p>","PeriodicalId":156,"journal":{"name":"Computer-Aided Civil and Infrastructure Engineering","volume":"39 18","pages":""},"PeriodicalIF":8.5,"publicationDate":"2024-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1111/mice.13335","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142118224","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Spatial causality in urban traffic networks explores how events or conditions in one location affect those in another. Unveiling congestion spatial causality is crucial for identifying congestion‐inducing bottlenecks in traffic networks and offering valuable insights for traffic network management and control. This study introduces the traffic‐convergent‐cross‐mapping (T‐CCM) method, a state‐space‐reconstruction approach from the dynamic system perspective, to identify causality among roads within urban traffic networks using time series data. Simultaneously, it effectively addresses the intricate challenges of uncertainty and interdependency among sensors caused by traffic flow dynamics. Empirical findings from real‐world (PeMS‐Bay area) traffic speed data validate the effectiveness of the T‐CCM method in detecting causality. This study reveals bidirectional causal effects between downstream and upstream roads in short‐term congestion generation and dissipation periods, which can pinpoint congestion origins and inform quick traffic management response. Furthermore, it elucidates the long‐term causality impacts between distant roads, particularly with regard to traveler choices and road land use attributes, guiding infrastructure investment and public transit improvements.
{"title":"A convergent cross‐mapping approach for unveiling congestion spatial causality in urban traffic networks","authors":"Jiannan Mao, Hao Huang, Yu Gu, Weike Lu, Tianli Tang, Fan Ding","doi":"10.1111/mice.13334","DOIUrl":"https://doi.org/10.1111/mice.13334","url":null,"abstract":"Spatial causality in urban traffic networks explores how events or conditions in one location affect those in another. Unveiling congestion spatial causality is crucial for identifying congestion‐inducing bottlenecks in traffic networks and offering valuable insights for traffic network management and control. This study introduces the traffic‐convergent‐cross‐mapping (T‐CCM) method, a state‐space‐reconstruction approach from the dynamic system perspective, to identify causality among roads within urban traffic networks using time series data. Simultaneously, it effectively addresses the intricate challenges of uncertainty and interdependency among sensors caused by traffic flow dynamics. Empirical findings from real‐world (PeMS‐Bay area) traffic speed data validate the effectiveness of the T‐CCM method in detecting causality. This study reveals bidirectional causal effects between downstream and upstream roads in short‐term congestion generation and dissipation periods, which can pinpoint congestion origins and inform quick traffic management response. Furthermore, it elucidates the long‐term causality impacts between distant roads, particularly with regard to traveler choices and road land use attributes, guiding infrastructure investment and public transit improvements.","PeriodicalId":156,"journal":{"name":"Computer-Aided Civil and Infrastructure Engineering","volume":"99 1","pages":""},"PeriodicalIF":11.775,"publicationDate":"2024-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142101192","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The reliance of contractor selection for specific construction activities on subjective judgments remains a complex decision‐making process with high stakes due to its impact on project success. Existing methods of contractor selection lack a data‐driven decision‐support approach, leading to suboptimal contractor assignments. Here, an advanced node2vec‐based recommendation system is proposed that addresses the shortcomings of conventional contractor selection by incorporating a broad range of quantitative performance indicators. This study utilizes semi‐supervised machine learning to analyze contractor records, creating a network in which nodes represent activities and weighted edges correspond to contractors and their performances, particularly cost and schedule performance indicators. Node2vec is found to display a prediction accuracy of 88.16% and 84.08% when processing cost and schedule performance rating networks, respectively. The novelty of this research lies in its proposed network‐based, multi‐criteria decision‐making method for ranking construction contractors using embedding information obtained from quantitative contractor performance data and processed by the node2vec procedure, along with the measurement of cosine similarity between contractors and the ideal as related to a given activity.
{"title":"Performance‐driven contractor recommendation system using a weighted activity–contractor network","authors":"Fatemeh Mostofi, Onur Behzat Tokdemir, Ümit Bahadır, Vedat Toğan","doi":"10.1111/mice.13332","DOIUrl":"https://doi.org/10.1111/mice.13332","url":null,"abstract":"The reliance of contractor selection for specific construction activities on subjective judgments remains a complex decision‐making process with high stakes due to its impact on project success. Existing methods of contractor selection lack a data‐driven decision‐support approach, leading to suboptimal contractor assignments. Here, an advanced node2vec‐based recommendation system is proposed that addresses the shortcomings of conventional contractor selection by incorporating a broad range of quantitative performance indicators. This study utilizes semi‐supervised machine learning to analyze contractor records, creating a network in which nodes represent activities and weighted edges correspond to contractors and their performances, particularly cost and schedule performance indicators. Node2vec is found to display a prediction accuracy of 88.16% and 84.08% when processing cost and schedule performance rating networks, respectively. The novelty of this research lies in its proposed network‐based, multi‐criteria decision‐making method for ranking construction contractors using embedding information obtained from quantitative contractor performance data and processed by the node2vec procedure, along with the measurement of cosine similarity between contractors and the ideal as related to a given activity.","PeriodicalId":156,"journal":{"name":"Computer-Aided Civil and Infrastructure Engineering","volume":"8 1","pages":""},"PeriodicalIF":11.775,"publicationDate":"2024-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142101193","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Jiandong Qiu, Sicheng Fu, Jushang Ou, Kai Tang, Xinming Qu, Shixiao Liang, Xin Wang, Bin Ran
In modern transportation systems, network‐wide traffic flow estimation is crucial for informed decision making, strategic infrastructure planning, and effective traffic management. While the limited availability of observed road‐segment traffic flow data presents a significant challenge, the emerging collection of Global Navigation Satellite System (GNSS) speed data across the entire network provides an alternative method for estimating the missing traffic flow information. To this end, this paper introduces a novel approach to estimating network‐wide road‐segment traffic flow. This approach takes advantage of the abundantly available GNSS speed data, coupled with only sparsely observed traffic flow samples. By integrating the principles of dynamic traffic assignment models with sparse recovery techniques, we formulate the problem of traffic flow estimation as a Least Absolute Shrinkage and Selection Operator (LASSO) optimization task. The efficacy and practical applicability of our proposed method are validated through evaluations using both hypothetical and real‐world case studies. The experimental findings exhibit a close alignment between the estimated and ground‐truth link flows across different time periods. Additionally, the method consistently produces low mean estimation errors for the majority of road segments, underlining the potential for our approach in effectively managing traffic flow estimation for large‐scale road networks, particularly in situations characterized by data scarcity.
{"title":"Estimating link flow through link speed with sparse flow data sampling","authors":"Jiandong Qiu, Sicheng Fu, Jushang Ou, Kai Tang, Xinming Qu, Shixiao Liang, Xin Wang, Bin Ran","doi":"10.1111/mice.13323","DOIUrl":"https://doi.org/10.1111/mice.13323","url":null,"abstract":"In modern transportation systems, network‐wide traffic flow estimation is crucial for informed decision making, strategic infrastructure planning, and effective traffic management. While the limited availability of observed road‐segment traffic flow data presents a significant challenge, the emerging collection of Global Navigation Satellite System (GNSS) speed data across the entire network provides an alternative method for estimating the missing traffic flow information. To this end, this paper introduces a novel approach to estimating network‐wide road‐segment traffic flow. This approach takes advantage of the abundantly available GNSS speed data, coupled with only sparsely observed traffic flow samples. By integrating the principles of dynamic traffic assignment models with sparse recovery techniques, we formulate the problem of traffic flow estimation as a Least Absolute Shrinkage and Selection Operator (LASSO) optimization task. The efficacy and practical applicability of our proposed method are validated through evaluations using both hypothetical and real‐world case studies. The experimental findings exhibit a close alignment between the estimated and ground‐truth link flows across different time periods. Additionally, the method consistently produces low mean estimation errors for the majority of road segments, underlining the potential for our approach in effectively managing traffic flow estimation for large‐scale road networks, particularly in situations characterized by data scarcity.","PeriodicalId":156,"journal":{"name":"Computer-Aided Civil and Infrastructure Engineering","volume":"19 1","pages":""},"PeriodicalIF":11.775,"publicationDate":"2024-08-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142101217","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pore structure characteristics of cementitious materials play a critical role in the transport properties of concrete structures. This paper develops a novel framework for modeling chloride penetration in concrete, considering the pore structure‐dependent model parameters. In the framework, a multi‐scale transport model was derived by linking the chloride diffusivities with pore size distributions (PSDs). Based on the three‐dimensional (3D) microstructure generated by “porous growth” and “hard core‐soft shell” methods, two sub‐models were computationally developed for determining the multi‐modal PSDs and pore size‐related chloride diffusivities. The predicted results by these series of models were compared with corresponding experimental data. The results indicated that by adopting pore size‐related diffusivities, even if the total porosities were the same, the proposed multi‐scale chloride transport model could better capture the effect of different PSDs on chloride penetration profiles, while the model without pore structure‐depended parameters would ignore the differences. Compared with the reference transport models, which adopt averaged chloride diffusivities, the chloride penetration depths predicted by the proposed multi‐scale model are in better agreement with experimental data, with 10%–25% reduced prediction error. This multi‐scale transport model is hoped to provide a novel computational approach on 3D microstructure generation and better reveal the underlying mechanism of the chloride penetration process in concrete from a microscopic perspective.
{"title":"Modeling the chloride transport in concrete from microstructure generation to chloride diffusivity prediction","authors":"Liang‐yu Tong, Qing‐feng Liu, Qingxiang Xiong, Zhaozheng Meng, Ouali Amiri, Mingzhong Zhang","doi":"10.1111/mice.13331","DOIUrl":"https://doi.org/10.1111/mice.13331","url":null,"abstract":"Pore structure characteristics of cementitious materials play a critical role in the transport properties of concrete structures. This paper develops a novel framework for modeling chloride penetration in concrete, considering the pore structure‐dependent model parameters. In the framework, a multi‐scale transport model was derived by linking the chloride diffusivities with pore size distributions (PSDs). Based on the three‐dimensional (3D) microstructure generated by “porous growth” and “hard core‐soft shell” methods, two sub‐models were computationally developed for determining the multi‐modal PSDs and pore size‐related chloride diffusivities. The predicted results by these series of models were compared with corresponding experimental data. The results indicated that by adopting pore size‐related diffusivities, even if the total porosities were the same, the proposed multi‐scale chloride transport model could better capture the effect of different PSDs on chloride penetration profiles, while the model without pore structure‐depended parameters would ignore the differences. Compared with the reference transport models, which adopt averaged chloride diffusivities, the chloride penetration depths predicted by the proposed multi‐scale model are in better agreement with experimental data, with 10%–25% reduced prediction error. This multi‐scale transport model is hoped to provide a novel computational approach on 3D microstructure generation and better reveal the underlying mechanism of the chloride penetration process in concrete from a microscopic perspective.","PeriodicalId":156,"journal":{"name":"Computer-Aided Civil and Infrastructure Engineering","volume":"29 1","pages":""},"PeriodicalIF":11.775,"publicationDate":"2024-08-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142101213","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Sina Tavasoli, Sina Poorghasem, Xiao Pan, T. Y. Yang, Y. Bao
This paper introduces an innovative autonomous survivor detection pipeline tailored for low‐cost micro aerial vehicles (MAVs) operating in post‐disaster indoor environments. This consists of three main components: (1) a novel pipeline for survivor geotagging, which includes autonomous navigation, mapping, and detection of survivors using thermal images; (2) a navigation strategy to ensure complete thermal scanning coverage for survivor detection using low‐cost commercial grade thermal camera; and (3) robust and accurate survivor detection using YOLOv8x and thermal imaging. To demonstrate the effectiveness of the proposed framework, first, the autonomous navigation algorithm is simulated in Robotic Operating System (ROS) and experimentally validated under different layouts. Second, the YOLOv8x algorithm is pretrained and achieves high accuracy. Finally, a real‐world implementation was conducted with partially covered survivors in a simulated post‐disaster environment. The results demonstrated the proposed pipeline can accurately map the layout of the environment and identify all survivors. This study demonstrates that affordable MAVs with basic thermal cameras can be effectively used to geotag survivors to support rescue missions during post‐disaster events.
{"title":"Autonomous post‐disaster indoor navigation and survivor detection using low‐cost micro aerial vehicles","authors":"Sina Tavasoli, Sina Poorghasem, Xiao Pan, T. Y. Yang, Y. Bao","doi":"10.1111/mice.13319","DOIUrl":"https://doi.org/10.1111/mice.13319","url":null,"abstract":"This paper introduces an innovative autonomous survivor detection pipeline tailored for low‐cost micro aerial vehicles (MAVs) operating in post‐disaster indoor environments. This consists of three main components: (1) a novel pipeline for survivor geotagging, which includes autonomous navigation, mapping, and detection of survivors using thermal images; (2) a navigation strategy to ensure complete thermal scanning coverage for survivor detection using low‐cost commercial grade thermal camera; and (3) robust and accurate survivor detection using YOLOv8x and thermal imaging. To demonstrate the effectiveness of the proposed framework, first, the autonomous navigation algorithm is simulated in Robotic Operating System (ROS) and experimentally validated under different layouts. Second, the YOLOv8x algorithm is pretrained and achieves high accuracy. Finally, a real‐world implementation was conducted with partially covered survivors in a simulated post‐disaster environment. The results demonstrated the proposed pipeline can accurately map the layout of the environment and identify all survivors. This study demonstrates that affordable MAVs with basic thermal cameras can be effectively used to geotag survivors to support rescue missions during post‐disaster events.","PeriodicalId":156,"journal":{"name":"Computer-Aided Civil and Infrastructure Engineering","volume":"51 1","pages":""},"PeriodicalIF":11.775,"publicationDate":"2024-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142085434","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Building change detection (BCD) plays a crucial role in urban planning and development. However, several pressing issues remain unresolved in this field, including false detections of buildings in complex backgrounds, the occurrence of jagged edges in segmentation results, and detection blind spots in densely built‐up areas. To address these challenges, this study innovatively proposes a Hierarchical Adaptive Gradual Recognition Network (HAGR‐Net) to improve the accuracy and robustness of BCD. Additionally, this research is the first to employ the Reinforcement Learning Optimization Algorithm Based on Particle Swarm (ROPS) to optimize the training process of HAGR‐Net, thereby accelerating the training process and reducing memory overhead. Experimental results indicate that the optimized HAGR‐Net outperforms state‐of‐the‐art methods on the WHU_CD, Google_CD, and LEVIR_CD data sets, achieving F1 scores of 93.13%, 85.31%, and 91.72%, and mean intersection over union (mIoU) scores of 91.20%, 85.99%, and 90.01%, respectively.
{"title":"A hierarchical progressive recognition network for building change detection in high‐resolution remote sensing images","authors":"Zhihuan Liu, Zaichun Yang, Tingting Ren, Zhenzhen Wang, JinSheng Deng, Chenxi Deng, Hongmin Zhao, Guoxiong Zhou, Aibin Chen, Liujun Li","doi":"10.1111/mice.13330","DOIUrl":"https://doi.org/10.1111/mice.13330","url":null,"abstract":"Building change detection (BCD) plays a crucial role in urban planning and development. However, several pressing issues remain unresolved in this field, including false detections of buildings in complex backgrounds, the occurrence of jagged edges in segmentation results, and detection blind spots in densely built‐up areas. To address these challenges, this study innovatively proposes a Hierarchical Adaptive Gradual Recognition Network (HAGR‐Net) to improve the accuracy and robustness of BCD. Additionally, this research is the first to employ the Reinforcement Learning Optimization Algorithm Based on Particle Swarm (ROPS) to optimize the training process of HAGR‐Net, thereby accelerating the training process and reducing memory overhead. Experimental results indicate that the optimized HAGR‐Net outperforms state‐of‐the‐art methods on the WHU_CD, Google_CD, and LEVIR_CD data sets, achieving F1 scores of 93.13%, 85.31%, and 91.72%, and mean intersection over union (mIoU) scores of 91.20%, 85.99%, and 90.01%, respectively.","PeriodicalId":156,"journal":{"name":"Computer-Aided Civil and Infrastructure Engineering","volume":"13 1","pages":""},"PeriodicalIF":11.775,"publicationDate":"2024-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142085134","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：