Advanced Engineering Informatics最新文献

Modeling and risk assessment of workers’ situation awareness in human-machine collaborative construction operations: A computational cognitive modeling and simulation approach

IF 8 1区工程技术

Advanced Engineering Informatics Pub Date : 2024-11-30 DOI: 10.1016/j.aei.2024.102951

Jue Li , Sihan He , Hui Lu , Gangyan Xu , Hongwei Wang

{"title":"Modeling and risk assessment of workers’ situation awareness in human-machine collaborative construction operations: A computational cognitive modeling and simulation approach","authors":"Jue Li , Sihan He , Hui Lu , Gangyan Xu , Hongwei Wang","doi":"10.1016/j.aei.2024.102951","DOIUrl":"10.1016/j.aei.2024.102951","url":null,"abstract":"<div><div>Insufficient situation awareness (SA) among workers remains a prominent factor contributing to construction accidents in complex and high-risk human-machine collaborative construction operations. However, previous studies have not fully explored the impact of various internal and external factors on the formation of workers’ SA, making it difficult to understand the potential changes in SA and respond to its error risks in specific scenarios. To address this issue, this paper proposes a proactive analysis approach of worker’s SA and the corresponding error risk based on computational cognitive modeling and simulation. This approach establishes a perception model by quantitatively depicting the mechanism underlying workers’ attention formation. Bayesian network is employed to represent the belief propagation process involved in worker’s comprehension and projection of the situation. The Monte Carlo method is applied to dynamically analyze the uncertainty inherent in the formation of workers’ SA. To demonstrate the feasibility and validity of the proposed approach, a shield tunneling construction project was adopted as an example. The results indicate that crucial factors such as stress, mental fatigue, and risk preference significantly impact shield machine operation workers’ SA, revealing dynamic changes and interactions of cognitive components within the SA formation process. The findings suggest that the proposed approach can serve as a proactive analysis tool to offer new insights for predicting and controlling risks associated with workers’ SA errors.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"63 ","pages":"Article 102951"},"PeriodicalIF":8.0,"publicationDate":"2024-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142744341","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}

引用次数: 0

A state of the art in digital twin for intelligent fault diagnosis

IF 8 1区工程技术

Advanced Engineering Informatics Pub Date : 2024-11-29 DOI: 10.1016/j.aei.2024.102963

Changhua Hu , Zeming Zhang , Chuanyang Li, Mingzhe Leng, Zhaoqiang Wang, Xinyi Wan, Chen Chen

{"title":"A state of the art in digital twin for intelligent fault diagnosis","authors":"Changhua Hu , Zeming Zhang , Chuanyang Li, Mingzhe Leng, Zhaoqiang Wang, Xinyi Wan, Chen Chen","doi":"10.1016/j.aei.2024.102963","DOIUrl":"10.1016/j.aei.2024.102963","url":null,"abstract":"<div><div>The intelligent manufacturing and digital technologies have rapidly advanced with the advent of the industry 4.0 era, placing higher demands on the stability, reliability, and safety of industrial equipment. Fault diagnosis (FD), a crucial step ensuring the regular operations, its accuracy and efficiency directly influence the stable operation of the equipment and economic benefits. With the progress of the artificial intelligence (AI) technology, data-driven FD methods have been developing in the area of intelligence, i.e., the intelligent fault diagnosis (IFD). Recently, a new solution is provided for IFD. That is the digital twin (DT), a technology serving as a bridge connecting the physical and virtual worlds. Numerous researchers have published studies on the use of DT technology for IFD of equipment. This paper analyzes 260 articles from 2017 to 2024, offering a systematic discussion of DT, IFD, and the application of DT in IFD. Firstly, the concepts, key technologies, and application scenarios of DT and IFD are described in detail; then, the application of DT technology in the field of IFD is emphasized; finally, this paper summarizes the existing problems and challenges, puts forward suggestions to solve the issues, and looks forward to the future development. This work is expected to provide valuable references and utilization for researchers in related fields, as well as, promoting the further development and application of DT technology in the IFD domain.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"63 ","pages":"Article 102963"},"PeriodicalIF":8.0,"publicationDate":"2024-11-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142744342","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}

引用次数: 0

Spatio-temporal attention-based hidden physics-informed neural network for remaining useful life prediction

IF 8 1区工程技术

Advanced Engineering Informatics Pub Date : 2024-11-29 DOI: 10.1016/j.aei.2024.102958

Feilong Jiang , Xiaonan Hou , Min Xia

{"title":"Spatio-temporal attention-based hidden physics-informed neural network for remaining useful life prediction","authors":"Feilong Jiang , Xiaonan Hou , Min Xia","doi":"10.1016/j.aei.2024.102958","DOIUrl":"10.1016/j.aei.2024.102958","url":null,"abstract":"<div><div>Predicting the Remaining Useful Life (RUL) is essential in Prognostic Health Management (PHM) for industrial systems. Although deep learning approaches have achieved considerable success in predicting RUL, challenges such as low prediction accuracy and interpretability pose significant challenges, hindering their practical implementation. In this work, we introduce a Spatio-temporal Attention-based Hidden Physics-informed Neural Network (STA-HPINN) for RUL prediction, which can utilize the associated physics of the system degradation. The spatio-temporal attention mechanism can extract important features from the input data. With the self-attention mechanism on both the sensor dimension and time step dimension, the proposed model can effectively extract degradation information. The hidden physics-informed neural network is utilized to capture the physics mechanisms that govern the evolution of RUL. With the constraint of physics, the model can achieve higher accuracy and reasonable predictions. The approach is validated on a benchmark dataset, demonstrating exceptional performance when compared to cutting-edge methods, especially in the case of complex conditions.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"63 ","pages":"Article 102958"},"PeriodicalIF":8.0,"publicationDate":"2024-11-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142744344","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}

引用次数: 0

A comprehensive multi-stage decision-making model for supplier selection and order allocation approach in the digital economy

IF 8 1区工程技术

Advanced Engineering Informatics Pub Date : 2024-11-28 DOI: 10.1016/j.aei.2024.102961

Xuhui Chen , Yong He , Golnaz Hooshmand Pakdel , Xiaofan Liu , Sai Wang

{"title":"A comprehensive multi-stage decision-making model for supplier selection and order allocation approach in the digital economy","authors":"Xuhui Chen , Yong He , Golnaz Hooshmand Pakdel , Xiaofan Liu , Sai Wang","doi":"10.1016/j.aei.2024.102961","DOIUrl":"10.1016/j.aei.2024.102961","url":null,"abstract":"<div><div>The increasingly serious environmental issues and fierce competition caused by globalization have brought pressure on supply chain managers who seek to allocate multiple purchase demands comprehensively, highlighting the significance of supplier assessment considering sustainability and technique. Moreover, many multi-criteria decision-making (MCDM) methods fail to quantify the risk preference of decision-makers (DMs) when conducting the supplier assessment process. Indeed, a hybrid supplier selection and order allocation model that integrates such requirements is yet to be proposed. Thus, this work develops a comprehensive decision-making model that constructs a deep learning model to forecast the potential demand and addresses the sustainable supplier selection based on cumulative prospect theory (CPT) and multi-material order allocation problem simultaneously. The proposed order allocation model is solved by the second generation of adaptive geometry estimation based many-objective evolutionary algorithm, with the technique of order preference similarity to the ideal solution used to filter out the best Pareto solution for DMs as the reference. Through implementing an illustrative case study of a leading Chinese engineering machinery manufacturer followed by a sensitivity analysis, the relatively strong applicability and scalability of the proposed model and methods are demonstrated. The results show that introducing Weibull distribution to estimate the theoretical obsolescence rate of historically sold accessories can result in higher demand prediction accuracy for consumable mechanical accessories. Integrating CPT into the MCDM framework allows us to evaluate suppliers more comprehensively by capturing the effect of DMs’ risk preferences and gain or loss sensitivity.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"63 ","pages":"Article 102961"},"PeriodicalIF":8.0,"publicationDate":"2024-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142744343","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}

引用次数: 0

Artificial rabbits optimization–based motion balance system for the impact recovery of a bipedal robot

IF 8 1区工程技术

Advanced Engineering Informatics Pub Date : 2024-11-28 DOI: 10.1016/j.aei.2024.102965

Ping-Huan Kuo , Wei-Cyuan Yang , Yu-Sian Lin , Chao-Chung Peng

{"title":"Artificial rabbits optimization–based motion balance system for the impact recovery of a bipedal robot","authors":"Ping-Huan Kuo , Wei-Cyuan Yang , Yu-Sian Lin , Chao-Chung Peng","doi":"10.1016/j.aei.2024.102965","DOIUrl":"10.1016/j.aei.2024.102965","url":null,"abstract":"<div><div>Research on the control of bipedal robots has predominantly focused on ensuring stability and balance during locomotion, often neglecting the robot’s ability to respond to unexpected external disturbances. In the present study, an algorithm is proposed to enable humanoid robots to maintain balance when they experience external impacts. In evaluation experiments, a robot was placed on flat surfaces and sloped terrain, where it experienced impacts from five angles. To evaluate the robot’s stability, data were collected before, during, and after each impact. The study utilized the artificial rabbits optimization (ARO) algorithm to optimize parameters and trained the robot’s control model by using a five-layer multilayer perceptron (MLP) neural network. Notably, the joint use of ARO and MLP yielded computational savings relative to conventional reinforcement learning methods. The proposed hybrid approach allowed the robot to adapt quickly to external forces and maintain balance effectively. The findings of this research hold considerable promise for enhancing the practical applications of bipedal robots in real-world scenarios, where unpredictable forces or impacts are common. By improving a robot’s ability to react dynamically and maintain balance, the proposed method enables humanoid robots to operate in highly challenging and dynamic environments, such as those associated with disaster response, industrial tasks, or everyday human interaction, without falling because of unexpected disturbances. Thus, the present study contributes to the field of humanoid robotics by addressing real-world challenges and providing a robust solution for impact resistance.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"63 ","pages":"Article 102965"},"PeriodicalIF":8.0,"publicationDate":"2024-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142744340","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}

引用次数: 0

A novel AI-driven EEG generalized classification model for cross-subject and cross-scene analysis 用于跨主体和跨场景分析的新型人工智能驱动脑电图广义分类模型

IF 8 1区工程技术

Advanced Engineering Informatics Pub Date : 2024-11-26 DOI: 10.1016/j.aei.2024.102971

Jingjing Li , Ching-Hung Lee , Yanhong Zhou , Tiange Liu , Tzyy-Ping Jung , Xianglong Wan , Dingna Duan , Dong Wen

{"title":"A novel AI-driven EEG generalized classification model for cross-subject and cross-scene analysis","authors":"Jingjing Li , Ching-Hung Lee , Yanhong Zhou , Tiange Liu , Tzyy-Ping Jung , Xianglong Wan , Dingna Duan , Dong Wen","doi":"10.1016/j.aei.2024.102971","DOIUrl":"10.1016/j.aei.2024.102971","url":null,"abstract":"<div><div>Artificial intelligence algorithms combined with electroencephalography (EEG) can effectively identify and interpret patterns of brain activity. However, the considerable variability in EEG signals among individuals and the challenges in transferring data and features among different scenarios result in a lack of universality in EEG signal analysis methods. To address these challenges, we introduce a novel AI-driven EEG general classification model called the Deformation Residual Compact Shrinkage Attention Mechanism (D-RCSAM) network. This low-parameter model improves spatial sampling positions using deformable convolution blocks and reduces computational costs while improving generalization performance through depthwise separable residual blocks. We further optimized the soft thresholding function to enhance the model’s nonlinearity and sparse representation, while also improving the loss function. We validated the proposed model on one public dataset and two private datasets, with results demonstrating that the D-RCSAM model effectively integrates both public and private EEG signal features. Visualization and interpretability results show that the D-RCSAM model can handle cross-subject and cross-scene classification tasks, outperforming state-of-the-art models in cognitive task classification. This research offers a new perspective on intelligent, comprehensive analysis across individuals and scenarios.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"63 ","pages":"Article 102971"},"PeriodicalIF":8.0,"publicationDate":"2024-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142723535","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}

引用次数: 0

CMPSO: A novel co-evolutionary multigroup particle swarm optimization for multi-mission UAVs path planning CMPSO：用于多任务无人机路径规划的新型协同进化多群粒子群优化技术

IF 8 1区工程技术

Advanced Engineering Informatics Pub Date : 2024-11-26 DOI: 10.1016/j.aei.2024.102923

Gang Hu , Mao Cheng , Essam H. Houssein , Heming Jia

{"title":"CMPSO: A novel co-evolutionary multigroup particle swarm optimization for multi-mission UAVs path planning","authors":"Gang Hu , Mao Cheng , Essam H. Houssein , Heming Jia","doi":"10.1016/j.aei.2024.102923","DOIUrl":"10.1016/j.aei.2024.102923","url":null,"abstract":"<div><div>To cope with the situation where an unmanned aerial vehicle (UVA) needs to perform missions to multiple locations, this paper presents a new multi-mission UAVs path planning model and proposes a novel co-evolutionary multigroup particle swarm optimization (CMPSO) for solving this complex model. In this model, a new ball curve, the ball <em>λ</em>-Bezier curve (B<em>λ</em>B), is used to represent the path of UAVs. In particular, UAV needs to satisfy <em>G</em><sup>0</sup> and <em>G</em><sup>1</sup> continuity at the must-pass points. Using this as a basis, a new model is built to generate a feasible path that is safe, smooth and constrained by the angle of climb and flight altitude. To solve this model efficiently, CMPSO framed by two novel different grouping learning mechanisms is proposed. Two different group learning mechanisms, grouping based on fitness values and activity level, replace the original speed and position update methods in PSO. The grouping mechanism based on the activity level uses the median of the velocity vector modes as a criterion to divide the whole population into two. They effectively facilitate the transfer of information between particles. In addition, a mutation mechanism based on the activity level is introduced to address the defect of PSO’s proneness to converge to local optima. By comparing CMPSO with 15 excellent metaheuristics at CEC 2017, CMPSO is ranked first with an average ranking of 3.72. Also, CMPSO has the best and most stable performance on 18 of the 21 engineering application problems. Finally, CMPSO is applied to three different environments of the path planning model. CMPSO outperforms the other compared algorithms in all three environments with a success rate of 100. This shows the efficiency and practicality of CMPSO in facing complex path planning problems.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"63 ","pages":"Article 102923"},"PeriodicalIF":8.0,"publicationDate":"2024-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142723534","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}

引用次数: 0

Multi-graph attention temporal convolutional network-based radius prediction in three-roller bending of thin-walled parts 薄壁部件三辊弯曲中基于多图注意时序卷积网络的半径预测

IF 8 1区工程技术

Advanced Engineering Informatics Pub Date : 2024-11-26 DOI: 10.1016/j.aei.2024.102940

Liling Zuo , Jie Zhang , Youlong Lyu , Yiqing Chen , Lei Diao , Zhijun Zhang

{"title":"Multi-graph attention temporal convolutional network-based radius prediction in three-roller bending of thin-walled parts","authors":"Liling Zuo , Jie Zhang , Youlong Lyu , Yiqing Chen , Lei Diao , Zhijun Zhang","doi":"10.1016/j.aei.2024.102940","DOIUrl":"10.1016/j.aei.2024.102940","url":null,"abstract":"<div><div>Three-roller bending is the key production process for aerospace products, which forms thin-walled parts into a curved shape with a specific radius through multiple passes of these parts between loaded rollers. Radius prediction is of great importance for reasonable bending process control for desired curves. However, due to the various influence factors and the complicated interaction between successive passes, it prevents high accuracy of radius prediction. The prediction model based on multi-graph attention temporal convolutional network is therefore proposed to deal with these challenges. First, multiple graphs are constructed from multi-pass observations from three-roller bending process, with each graph representing the influencing factors of a specific pass. Second, graph attention mechanism explores the coupling effects of influence factors on the radius and realizes the extraction of key factors for each graph. Third, temporal convolutional network reveals the interaction between successive passes by establishing the connection between different graphs, and provides the radius prediction at each pass. In comparative experiments based on simulated data and experimental data collected from real cases, the results demonstrate the higher prediction accuracy of the proposed method over traditional methods.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"63 ","pages":"Article 102940"},"PeriodicalIF":8.0,"publicationDate":"2024-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142707315","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}

引用次数: 0

A stacked graph neural network with self-exciting process for robotic cognitive strategy reasoning in proactive human-robot collaborative assembly 用于主动式人机协作装配中机器人认知策略推理的具有自激过程的堆叠图神经网络

IF 8 1区工程技术

Advanced Engineering Informatics Pub Date : 2024-11-26 DOI: 10.1016/j.aei.2024.102957

Pengfei Ding , Jie Zhang , Peng Zhang , Youlong Lv , Dexian Wang

{"title":"A stacked graph neural network with self-exciting process for robotic cognitive strategy reasoning in proactive human-robot collaborative assembly","authors":"Pengfei Ding , Jie Zhang , Peng Zhang , Youlong Lv , Dexian Wang","doi":"10.1016/j.aei.2024.102957","DOIUrl":"10.1016/j.aei.2024.102957","url":null,"abstract":"<div><div>Proactive human-robot collaborative assembly, a cognitively driven human-robot collaboration, requires research into robot cognitive strategy reasoning to ensure that the robot actively collaborates with the operator in task completion. However, current methods primarily focus on the pairwise relationships of assembly components in discrete snapshots. They could fail to represent the interconnected status of dynamic assembly, leading to inaccurate task allocation, thereby affecting robotic cognitive strategy. To address this problem, we propose a stacked graph neural network (GNN) with self-exciting process to capture the correlation and triggering mechanisms between time-varying tasks. Firstly, a temporal hypergraph with assembly knowledge is constructed to represent the non-pairwise relationships among assembly components in time-varying tasks, aiming to reduce the redundant information brought by pairwise relationships. Then, considering the characteristic of mutual influence between assembly events, a Hawkes process is introduced into the stacked GNN architecture to learn the event correlation representation in the temporal hypergraph. This point process models the self-exciting process of assembly events for simultaneously capturing the individual and collective features of events, thereby revealing the triggering mechanisms of the dynamic events. Finally, the effectiveness of proposed method is demonstrated by comparative experiments and the results of robotic cognitive strategy reasoning on dynamic assembly.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"63 ","pages":"Article 102957"},"PeriodicalIF":8.0,"publicationDate":"2024-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142707316","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}

引用次数: 0

UAV applications in intelligent traffic: RGBT image feature registration and complementary perception 无人机在智能交通中的应用：RGBT 图像特征注册与互补感知

IF 8 1区工程技术

Advanced Engineering Informatics Pub Date : 2024-11-26 DOI: 10.1016/j.aei.2024.102953

Yingying Ji , Kechen Song , Hongwei Wen , Xiaotong Xue , Yunhui Yan , Qinggang Meng

{"title":"UAV applications in intelligent traffic: RGBT image feature registration and complementary perception","authors":"Yingying Ji , Kechen Song , Hongwei Wen , Xiaotong Xue , Yunhui Yan , Qinggang Meng","doi":"10.1016/j.aei.2024.102953","DOIUrl":"10.1016/j.aei.2024.102953","url":null,"abstract":"<div><div>The flexibility of unmanned aerial vehicles (UAVs) has led to a wide range of applications in the field of intelligent traffic detection. In order to cope with the detection needs of various scenarios in all-weather, more and more attention has been focused on methods that incorporate RGB and thermal(T) images. The existing research focuses on manually registered RGBT image detection methods. However, for mobile devices such as UAVs, strict registration is almost impossible, and the detection performance of existing methods without registration is greatly reduced. In order to improve the efficiency, we consider the direct detection of unregistered raw images acquired by UAVs. Therefore, this paper introduces RGBT salient object detection and proposes a feature registration and complementary perception network (FRCPNet). To achieve accurate RGB-T SOD in the presence of misregistration issues, we progressively perform pixel-level alignment of multi-level features for each modality, while enhancing the semantic correlation between the two modalities. This is followed by complementary perception of global information, leading to improved detection performance. Experiments demonstrate that our proposed method is competitive with the current state-of-the-art methods on RGBT image pairs acquired in real scenes with large parallax. In addition, our method has application value in scenes such as automatic driving and intelligent monitoring in intelligent traffic. The source code will be published at <span><span>https://github.com/VDT-2048/FRCPNet</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"63 ","pages":"Article 102953"},"PeriodicalIF":8.0,"publicationDate":"2024-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142723533","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}

引用次数: 0