Computers in Industry最新文献_第6页

Apple varieties and growth prediction with time series classification based on deep learning to impact the harvesting decisions 利用基于深度学习的时间序列分类进行苹果品种和生长预测，以影响收获决策

IF 8.2 1区计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers in Industry

Pub Date : 2024-09-28 DOI: 10.1016/j.compind.2024.104191

Mustafa Mhamed , Zhao Zhang , Wanjia Hua , Liling Yang , Mengning Huang , Xu Li , Tiecheng Bai , Han Li , Man Zhang

Apples are among the most popular fruits globally due to their health and nutritional benefits for humans. Artificial intelligence in agriculture has advanced, but vision, which improves machine efficiency, speed, and production, still needs to be improved. Managing apple development from planting to harvest affects productivity, quality, and economics. In this study, by establishing a vision system platform with a range of camera types that conforms with orchard standard specifications for data gathering, this work provides two new apple collections: Orchard Fuji Growth Stages (OFGS) and Orchard Apple Varieties (OAV), with preliminary benchmark assessments. Secondly, this research proposes the orchard apple vision transformer method (POA-VT), incorporating novel regularization techniques (CRT) that assist us in boosting efficiency and optimizing the loss functions. The highest accuracy scores are 91.56 % for OFGS and 94.20 % for OAV. Thirdly, an ablation study will be conducted to demonstrate the importance of CRT to the proposed method. Fourthly, the CRT outperforms the baselines by comparing it with the standard regularization functions. Finally, time series analyses predict the ‘Fuji’ growth stage, with the outstanding training and validation RMSE being 19.29 and 19.26, respectively. The proposed method offers high efficiency via multiple tasks and improves the automation of apple systems. It is highly flexible in handling various tasks related to apple fruits. Furthermore, it can integrate with real-time systems, such as UAVs and sorting systems. This research benefits the growth of apple’s robotic vision, development policies, time-sensitive harvesting schedules, and decision-making.

苹果因其对人类健康和营养的益处而成为全球最受欢迎的水果之一。农业领域的人工智能已经取得了长足进步，但提高机器效率、速度和产量的视觉技术仍有待改进。管理苹果从种植到收获的整个过程会影响其生产率、质量和经济效益。在这项研究中，通过建立一个视觉系统平台，配备一系列符合果园数据采集标准规范的相机类型，这项工作提供了两种新的苹果采集方法：果园富士生长阶段（OFGS）和果园苹果品种（OAV），并进行初步基准评估。其次，本研究提出了果园苹果视觉转换器方法（POA-VT），并结合了新颖的正则化技术（CRT），帮助我们提高效率并优化损失函数。结果表明，OFGS 和 OAV 的准确率分别达到 91.56% 和 94.20%。第三，将进行一项消融研究，以证明 CRT 对拟议方法的重要性。第四，通过与标准正则化函数比较，CRT 的性能优于基线。最后，时间序列分析预测了 "富士 "生长阶段，训练和验证均方根误差分别为 19.29 和 19.26，表现突出。所提出的方法通过多重任务实现了高效率，提高了苹果系统的自动化程度。它在处理与苹果果实相关的各种任务时具有很高的灵活性。此外，它还能与无人机和分拣系统等实时系统集成。这项研究有利于苹果机器人视觉、开发政策、时间敏感的收获计划和决策的发展。

{"title":"Apple varieties and growth prediction with time series classification based on deep learning to impact the harvesting decisions","authors":"Mustafa Mhamed , Zhao Zhang , Wanjia Hua , Liling Yang , Mengning Huang , Xu Li , Tiecheng Bai , Han Li , Man Zhang","doi":"10.1016/j.compind.2024.104191","DOIUrl":"10.1016/j.compind.2024.104191","url":null,"abstract":"<div><div>Apples are among the most popular fruits globally due to their health and nutritional benefits for humans. Artificial intelligence in agriculture has advanced, but vision, which improves machine efficiency, speed, and production, still needs to be improved. Managing apple development from planting to harvest affects productivity, quality, and economics. In this study, by establishing a vision system platform with a range of camera types that conforms with orchard standard specifications for data gathering, this work provides two new apple collections: Orchard Fuji Growth Stages (OFGS) and Orchard Apple Varieties (OAV), with preliminary benchmark assessments. Secondly, this research proposes the orchard apple vision transformer method (POA-VT), incorporating novel regularization techniques (CRT) that assist us in boosting efficiency and optimizing the loss functions. The highest accuracy scores are 91.56 % for OFGS and 94.20 % for OAV. Thirdly, an ablation study will be conducted to demonstrate the importance of CRT to the proposed method. Fourthly, the CRT outperforms the baselines by comparing it with the standard regularization functions. Finally, time series analyses predict the ‘Fuji’ growth stage, with the outstanding training and validation RMSE being 19.29 and 19.26, respectively. The proposed method offers high efficiency via multiple tasks and improves the automation of apple systems. It is highly flexible in handling various tasks related to apple fruits. Furthermore, it can integrate with real-time systems, such as UAVs and sorting systems. This research benefits the growth of apple’s robotic vision, development policies, time-sensitive harvesting schedules, and decision-making.</div></div>","PeriodicalId":55219,"journal":{"name":"Computers in Industry","volume":"164 ","pages":"Article 104191"},"PeriodicalIF":8.2,"publicationDate":"2024-09-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142358147","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Maximum subspace transferability discriminant analysis: A new cross-domain similarity measure for wind-turbine fault transfer diagnosis 最大子空间转移性判别分析：用于风力涡轮机故障转移诊断的新型跨域相似性测量方法

IF 8.2 1区计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers in Industry

Pub Date : 2024-09-27 DOI: 10.1016/j.compind.2024.104194

Quan Qian , Fei Wu , Yi Wang , Yi Qin

In the field of fault transfer diagnosis, many approaches only focus on the distribution alignment and knowledge transfer between the source domain and target domain. However, most of these approaches ignore the precondition of whether this transfer task is transferable. Current mainstream transferability discrimination methods heavily depend on expert knowledge and are extremely vulnerable to the noise interference and variations in feature scale. This limits their applicability due to the intelligent requirements and complex industrial environment. To address the challenges mentioned previously, this paper introduces a novel cross-domain similarity measure called maximum subspace transferability discriminant analysis (MSTDA) with zero-label prior knowledge. MSTDA is comprised of a maximum subspace representation and a similarity measurement criterion. During the phase of maximum subspace representation, a new kernel-induced Hilbert space is designed to map the low-dimensional original samples into the high-dimensional space to maximize the separability of different faults and then solve the separable intrinsic fault features. Following that, a novel similarity measurement criterion that is resistant to variations in feature scale is developed. This criterion is based on the orthogonal bases of intrinsic feature subspaces. The mini-batch sampling strategy is used to ensure the timeliness of MSTDA. Finally, the experimental results on three cases, particularly in the actual wind turbine dataset, confirm that the proposed MSTDA outperforms other well-known similarity measure methods in terms of transferability evaluation. The related code can be downloaded from https://qinyi-team.github.io/2024/09/Maximum-subspace-transferability-discriminant-analysis.

在故障转移诊断领域，许多方法只关注源域和目标域之间的分布对齐和知识转移。然而，这些方法大多忽略了这一转移任务是否具有可转移性这一前提条件。目前主流的可转移性判别方法严重依赖专家知识，极易受到噪声干扰和特征尺度变化的影响。由于智能化要求和复杂的工业环境，这限制了它们的适用性。为应对上述挑战，本文引入了一种新型跨域相似性测量方法，即零标签先验知识下的最大子空间可转移性判别分析（MSTDA）。MSTDA 由最大子空间表示和相似性测量标准组成。在最大子空间表示阶段，设计一个新的内核诱导希尔伯特空间，将低维原始样本映射到高维空间，以最大限度地分离不同故障，然后求解可分离的内在故障特征。随后，开发了一种新型的相似性测量准则，该准则可抵御特征尺度的变化。该准则基于内在特征子空间的正交基。微型批量采样策略用于确保 MSTDA 的及时性。最后，三个案例的实验结果，尤其是实际风力涡轮机数据集的实验结果，证实了所提出的 MSTDA 在可移植性评估方面优于其他著名的相似性度量方法。相关代码可从 https://qinyi-team.github.io/2024/09/Maximum-subspace-transferability-discriminant-analysis 下载。

{"title":"Maximum subspace transferability discriminant analysis: A new cross-domain similarity measure for wind-turbine fault transfer diagnosis","authors":"Quan Qian , Fei Wu , Yi Wang , Yi Qin","doi":"10.1016/j.compind.2024.104194","DOIUrl":"10.1016/j.compind.2024.104194","url":null,"abstract":"<div><div>In the field of fault transfer diagnosis, many approaches only focus on the distribution alignment and knowledge transfer between the source domain and target domain. However, most of these approaches ignore the precondition of whether this transfer task is transferable. Current mainstream transferability discrimination methods heavily depend on expert knowledge and are extremely vulnerable to the noise interference and variations in feature scale. This limits their applicability due to the intelligent requirements and complex industrial environment. To address the challenges mentioned previously, this paper introduces a novel cross-domain similarity measure called maximum subspace transferability discriminant analysis (MSTDA) with zero-label prior knowledge. MSTDA is comprised of a maximum subspace representation and a similarity measurement criterion. During the phase of maximum subspace representation, a new kernel-induced Hilbert space is designed to map the low-dimensional original samples into the high-dimensional space to maximize the separability of different faults and then solve the separable intrinsic fault features. Following that, a novel similarity measurement criterion that is resistant to variations in feature scale is developed. This criterion is based on the orthogonal bases of intrinsic feature subspaces. The mini-batch sampling strategy is used to ensure the timeliness of MSTDA. Finally, the experimental results on three cases, particularly in the actual wind turbine dataset, confirm that the proposed MSTDA outperforms other well-known similarity measure methods in terms of transferability evaluation. The related code can be downloaded from https://qinyi-team.github.io/2024/09/Maximum-subspace-transferability-discriminant-analysis.</div></div>","PeriodicalId":55219,"journal":{"name":"Computers in Industry","volume":"164 ","pages":"Article 104194"},"PeriodicalIF":8.2,"publicationDate":"2024-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142327883","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Dual channel visible graph convolutional neural network for microleakage monitoring of pipeline weld homalographic cracks 用于监测管道焊缝同色裂纹微渗漏的双通道可见图卷积神经网络

IF 8.2 1区计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers in Industry

Pub Date : 2024-09-26 DOI: 10.1016/j.compind.2024.104193

Jing Huang , Zhifen Zhang , Rui Qin , Yanlong Yu , Yongjie Li , Quanning Xu , Ji Xing , Guangrui Wen , Wei Cheng , Xuefeng Chen

When using a single sensor to monitor early microleakage of nuclear power pressure pipeline leakage, there are problems such as low monitoring accuracy and poor early warning reliability due to the limitations of the monitoring range and weak difference between the leakage signals. To address these challenges, this paper proposes a dual channel visible graph convolutional neural network (DCV-GCN). Firstly, the acoustic emission time-series data of each channel are truncated and divided, and the significant frequency bands are selected based on the envelope spectrum. On this basis, the sequence group is averaged to obtain the graph structure sequence. Then, the limited penetrable visibility (LPV) graph construction algorithm is used to calculate the adjacency matrix, and the important nodes is reserved according to the eigenvector centrality. Furthermore, the inverse ratio of the distance from the sensor in each single channel to the center of the crack is used as the fusion weight, and the adjacency matrices are merged after normalization to transform the construction of the graph structure dataset. Finally, the dataset is input into the graph convolutional neural network, and the effectiveness of the method is verified by carefully designing three homalographic cracks. The results show that the proposed method can effectively extract the distinguishing features with similar frequency components and similar leakage rates, and the recognition accuracy of different leakage states can reach 98.56 %. In addition, through ablation experiments and different parameter strategy settings, the operating mechanism is explained, which can provide a reference for monitoring and analysis by industrial technicians.

在使用单传感器监测核电压力管道泄漏早期微泄漏时，由于监测范围的限制和泄漏信号之间的微弱差异，存在监测精度低、预警可靠性差等问题。针对这些难题，本文提出了一种双通道可见图卷积神经网络（DCV-GCN）。首先，对每个通道的声发射时间序列数据进行截断和分割，并根据包络谱选择重要频段。在此基础上，对序列组进行平均处理，得到图结构序列。然后，利用有限可穿透性（LPV）图构建算法计算邻接矩阵，并根据特征向量中心性保留重要节点。此外，将每个单通道传感器到裂缝中心的距离的反比作为融合权重，并将邻接矩阵归一化后合并，以转换图结构数据集的构建。最后，将数据集输入图卷积神经网络，并通过精心设计的三条同源裂纹验证了该方法的有效性。结果表明，所提出的方法能有效提取频率成分相似、泄漏率相似的识别特征，对不同泄漏状态的识别准确率可达 98.56%。此外，通过烧蚀实验和不同的参数策略设置，解释了运行机理，可为工业技术人员的监测和分析提供参考。

{"title":"Dual channel visible graph convolutional neural network for microleakage monitoring of pipeline weld homalographic cracks","authors":"Jing Huang , Zhifen Zhang , Rui Qin , Yanlong Yu , Yongjie Li , Quanning Xu , Ji Xing , Guangrui Wen , Wei Cheng , Xuefeng Chen","doi":"10.1016/j.compind.2024.104193","DOIUrl":"10.1016/j.compind.2024.104193","url":null,"abstract":"<div><div>When using a single sensor to monitor early microleakage of nuclear power pressure pipeline leakage, there are problems such as low monitoring accuracy and poor early warning reliability due to the limitations of the monitoring range and weak difference between the leakage signals. To address these challenges, this paper proposes a dual channel visible graph convolutional neural network (DCV-GCN). Firstly, the acoustic emission time-series data of each channel are truncated and divided, and the significant frequency bands are selected based on the envelope spectrum. On this basis, the sequence group is averaged to obtain the graph structure sequence. Then, the limited penetrable visibility (LPV) graph construction algorithm is used to calculate the adjacency matrix, and the important nodes is reserved according to the eigenvector centrality. Furthermore, the inverse ratio of the distance from the sensor in each single channel to the center of the crack is used as the fusion weight, and the adjacency matrices are merged after normalization to transform the construction of the graph structure dataset. Finally, the dataset is input into the graph convolutional neural network, and the effectiveness of the method is verified by carefully designing three homalographic cracks. The results show that the proposed method can effectively extract the distinguishing features with similar frequency components and similar leakage rates, and the recognition accuracy of different leakage states can reach 98.56 %. In addition, through ablation experiments and different parameter strategy settings, the operating mechanism is explained, which can provide a reference for monitoring and analysis by industrial technicians.</div></div>","PeriodicalId":55219,"journal":{"name":"Computers in Industry","volume":"164 ","pages":"Article 104193"},"PeriodicalIF":8.2,"publicationDate":"2024-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142323436","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Video-based automatic people counting for public transport: On-bus versus off-bus deployment 基于视频的公共交通人员自动计数：公交车上与公交车外的部署

IF 8.2 1区计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers in Industry

Pub Date : 2024-09-26 DOI: 10.1016/j.compind.2024.104195

Chris McCarthy, Hadi Ghaderi, Felip Martí, Prem Jayaraman, Hussein Dia

Interest in Automatic People Counting (APC) for crowd detection and management is rapidly growing. While a range of Internet of Things (IoT) sensors and systems exist, video analytics is emerging as a particularly attractive option — especially for applications where more traditional methods of people counting are not available, unreliable or expensive. In this paper we focus on automatic people counting in the public transport context – specifically, rail replacement bus services – in which bus companies are typically contracted to provide bus services to replace train services during periods of planned and unplanned line disruption. This presents a particularly compelling use-case for video-based people counting, while also presenting unique challenges. Field trials are thus vital to the proper assessment of video-based APC solutions, however remain relatively scarce in the literature. While datasets to support research and benchmarking exist, these do not capture the intrinsic complexities of real-world deployment and the implications of selected configurations — in particular, on-vehicle versus off-vehicle use cases. In this paper, we evaluate our own video-based APC solution, representative of state-of-the-art approaches in the literature, in two separate extensive (i.e, multi-day) metropolitan field trials covering both on and off-bus use-cases. Through real-world deployment of the system in both settings, we highlight key differences with respect to APC accuracy, as well as other practical considerations, and the validity of underlying assumptions in both on and off-bus scenarios.

人们对用于人群检测和管理的自动人员计数（APC）的兴趣正在迅速增长。虽然已有一系列物联网（IoT）传感器和系统，但视频分析技术正成为一种特别有吸引力的选择--尤其是在没有更传统的人员计数方法、不可靠或昂贵的应用领域。在本文中，我们将重点讨论公共交通环境下的自动人员计数，特别是铁路替代公交服务，公交公司通常与铁路公司签订合同，在计划内或计划外线路中断期间提供公交服务，以替代火车服务。在这种情况下，公交公司通常会在计划内或计划外的线路中断期间承包公交服务，以取代火车服务。这为基于视频的人员计数提供了一个特别有吸引力的使用案例，同时也带来了独特的挑战。因此，实地试验对于正确评估基于视频的旅客自助计数解决方案至关重要，但在文献中却相对较少。虽然存在支持研究和基准测试的数据集，但这些数据集无法捕捉真实世界部署的内在复杂性和所选配置的影响，特别是车载与非车载用例。在本文中，我们对自己的基于视频的 APC 解决方案进行了评估，该方案代表了文献中最先进的方法，分别在两个大城市进行了广泛（即多天）的实地试验，涵盖了公交车上和公交车下的使用案例。通过在这两个环境中对系统的实际部署，我们强调了 APC 精度方面的主要差异、其他实际考虑因素以及在公交车内和公交车外场景中基本假设的有效性。

{"title":"Video-based automatic people counting for public transport: On-bus versus off-bus deployment","authors":"Chris McCarthy, Hadi Ghaderi, Felip Martí, Prem Jayaraman, Hussein Dia","doi":"10.1016/j.compind.2024.104195","DOIUrl":"10.1016/j.compind.2024.104195","url":null,"abstract":"<div><div>Interest in Automatic People Counting (APC) for crowd detection and management is rapidly growing. While a range of Internet of Things (IoT) sensors and systems exist, video analytics is emerging as a particularly attractive option — especially for applications where more traditional methods of people counting are not available, unreliable or expensive. In this paper we focus on automatic people counting in the public transport context – specifically, rail replacement bus services – in which bus companies are typically contracted to provide bus services to replace train services during periods of planned and unplanned line disruption. This presents a particularly compelling use-case for video-based people counting, while also presenting unique challenges. Field trials are thus vital to the proper assessment of video-based APC solutions, however remain relatively scarce in the literature. While datasets to support research and benchmarking exist, these do not capture the intrinsic complexities of real-world deployment and the implications of selected configurations — in particular, on-vehicle versus off-vehicle use cases. In this paper, we evaluate our own video-based APC solution, representative of state-of-the-art approaches in the literature, in two separate extensive (i.e, multi-day) metropolitan field trials covering both on and off-bus use-cases. Through real-world deployment of the system in both settings, we highlight key differences with respect to APC accuracy, as well as other practical considerations, and the validity of underlying assumptions in both on and off-bus scenarios.</div></div>","PeriodicalId":55219,"journal":{"name":"Computers in Industry","volume":"164 ","pages":"Article 104195"},"PeriodicalIF":8.2,"publicationDate":"2024-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142323437","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

TDAD: Self-supervised industrial anomaly detection with a two-stage diffusion model TDAD：利用两阶段扩散模型进行自我监督式工业异常检测

IF 8.2 1区计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers in Industry

Pub Date : 2024-09-26 DOI: 10.1016/j.compind.2024.104192

Changyun Wei , Hui Han , Yu Xia , Ze Ji

Visual anomaly detection has emerged as a highly applicable solution in practical industrial manufacturing, owing to its notable effectiveness and efficiency. However, it also presents several challenges and uncertainties. To address the complexity of anomaly types and the high cost associated with data annotation, this paper introduces a self-supervised learning framework called TDAD, based on a two-stage diffusion model. TDAD consists of three key components: anomaly synthesis, image reconstruction, and defect segmentation. It is trained end-to-end, with the goal of improving pixel-level segmentation accuracy of anomalies and reducing false detection rates. By synthesizing anomalies from normal samples, designing a diffusion model-based reconstruction network, and incorporating a multiscale semantic feature fusion module for defect segmentation, TDAD achieves state-of-the-art performance in image-level detection and anomaly localization on challenging and widely used datasets such as MVTec and VisA benchmarks.

视觉异常检测因其显著的有效性和效率，已成为实际工业生产中非常适用的解决方案。然而，它也带来了一些挑战和不确定性。针对异常类型的复杂性和数据标注的高成本，本文基于两阶段扩散模型，介绍了一种名为 TDAD 的自监督学习框架。TDAD 由三个关键部分组成：异常合成、图像重建和缺陷分割。它采用端到端训练，目的是提高异常点像素级分割的准确性，降低误检率。通过从正常样本中合成异常点，设计基于扩散模型的重建网络，并结合多尺度语义特征融合模块进行缺陷分割，TDAD 在具有挑战性且广泛使用的数据集（如 MVTec 和 VisA 基准）上实现了一流的图像级检测和异常点定位性能。

引用次数: 0

A novel anomaly detection method for magnetic flux leakage signals via a feature-based unsupervised detection network 通过基于特征的无监督检测网络对漏磁通信号进行异常检测的新方法

IF 8.2 1区计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers in Industry

Pub Date : 2024-09-25 DOI: 10.1016/j.compind.2024.104190

He Zhao, Jinhai Liu, Qiannan Wang, Xiangkai Shen, Lin Jiang

High-precision anomaly detection, as the key technology of magnetic flux leakage (MFL) signal detection, is a challenging task. It is difficult to detect anomalies in MFL signals due to the variety of anomalies and the characteristics of the anomalies are easily submerged in the variation of the natural signals. To address the above issues, a feature-based unsupervised detection network (FUDet) is designed, which accomplishes the unsupervised anomaly detection task through feature discrimination and feature reconstruction. Firstly, a bidirectional discrimination module is proposed, which can input normal and anomaly feature distributions to mine the characteristics of samples, so as to enhance the ability of the model to recognize anomaly signals. Secondly, a dynamic noise generation module is designed to generate different feature distributions for each input that are consistent with the characteristics of MFL signals. This module creates an adversarial effect with the discriminator, allowing it to identify more subtle feature differences through training. Finally, a reconstruction classification module is designed to naturally reconstruct the non-normal features and normal features into normal signals, which can be used to detect anomalies by comparing the difference between the input signals and the reconstructed signals. Experimentally, the method is proved to outperform the P-AUROC of the state-of-the-art method by 3.1% under MFL signals and achieves outstanding results in MFL signal anomaly detection.

高精度异常检测作为磁通量泄漏（MFL）信号检测的关键技术，是一项具有挑战性的任务。由于异常点种类繁多，且异常点的特征容易被淹没在自然信号的变化中，因此很难检测出 MFL 信号中的异常点。针对上述问题，设计了一种基于特征的无监督检测网络（FUDet），通过特征判别和特征重构完成无监督异常检测任务。首先，提出了一个双向判别模块，它可以输入正常和异常特征分布来挖掘样本的特征，从而增强模型识别异常信号的能力。其次，设计了动态噪声生成模块，为每次输入生成符合 MFL 信号特征的不同特征分布。该模块会对鉴别器产生对抗效应，使其能够通过训练识别更细微的特征差异。最后，设计了一个重构分类模块，将非正常特征和正常特征自然重构为正常信号，通过比较输入信号和重构信号之间的差异来检测异常。实验证明，该方法在 MFL 信号下的 P-AUROC 优于最先进方法的 3.1%，在 MFL 信号异常检测中取得了优异的成绩。

{"title":"A novel anomaly detection method for magnetic flux leakage signals via a feature-based unsupervised detection network","authors":"He Zhao, Jinhai Liu, Qiannan Wang, Xiangkai Shen, Lin Jiang","doi":"10.1016/j.compind.2024.104190","DOIUrl":"10.1016/j.compind.2024.104190","url":null,"abstract":"<div><div>High-precision anomaly detection, as the key technology of magnetic flux leakage (MFL) signal detection, is a challenging task. It is difficult to detect anomalies in MFL signals due to the variety of anomalies and the characteristics of the anomalies are easily submerged in the variation of the natural signals. To address the above issues, a feature-based unsupervised detection network (FUDet) is designed, which accomplishes the unsupervised anomaly detection task through feature discrimination and feature reconstruction. Firstly, a bidirectional discrimination module is proposed, which can input normal and anomaly feature distributions to mine the characteristics of samples, so as to enhance the ability of the model to recognize anomaly signals. Secondly, a dynamic noise generation module is designed to generate different feature distributions for each input that are consistent with the characteristics of MFL signals. This module creates an adversarial effect with the discriminator, allowing it to identify more subtle feature differences through training. Finally, a reconstruction classification module is designed to naturally reconstruct the non-normal features and normal features into normal signals, which can be used to detect anomalies by comparing the difference between the input signals and the reconstructed signals. Experimentally, the method is proved to outperform the P-AUROC of the state-of-the-art method by 3.1% under MFL signals and achieves outstanding results in MFL signal anomaly detection.</div></div>","PeriodicalId":55219,"journal":{"name":"Computers in Industry","volume":"164 ","pages":"Article 104190"},"PeriodicalIF":8.2,"publicationDate":"2024-09-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142319097","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Extended realities and discrete events simulations: A systematic review to define design trade-offs and directions 扩展现实与离散事件模拟：界定设计权衡和方向的系统审查

IF 8.2 1区计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers in Industry

Pub Date : 2024-09-24 DOI: 10.1016/j.compind.2024.104188

Giulia Wally Scurati , Francesco Ferrise , Marco Bertoni

Extended Reality (XR) technologies are increasingly popular to support the engagement of different audiences and stakeholders with Discrete Event Simulations (DES) due to their capability to deliver more accessible visual and immersive experiences. XR applications can be developed either using modules integrated into DES software or game engines, providing different sets of opportunities in the environment design. However, there is a lack of development guidelines for such environments, considering visualization, information presentation, interaction, and navigation aspects. The paper presents a systematic review of the use of XR for DES, relating the results to XR design heuristics to identify and discuss major design tradeoffs. Finally, a case study from the mining sector is exemplified to illustrate possible. solutions to balance the trade-offs.

由于扩展现实（XR）技术能够提供更容易获得的视觉和身临其境的体验，因此在支持不同受众和利益相关者参与离散事件模拟（DES）方面越来越受欢迎。XR 应用可以通过集成到 DES 软件中的模块或游戏引擎来开发，从而为环境设计提供不同的机会。然而，考虑到可视化、信息展示、交互和导航等方面，目前还缺乏针对此类环境的开发指南。本文系统回顾了 XR 在 DES 中的应用，并将结果与 XR 设计启发式方法联系起来，以确定和讨论主要的设计权衡。最后，还举例说明了矿业部门的一个案例研究，以说明平衡权衡的可能解决方案。

引用次数: 0

Development of immersive bridge digital twin platform to facilitate bridge damage assessment and asset model updates 开发沉浸式桥梁数字孪生平台，促进桥梁损坏评估和资产模型更新

IF 8.2 1区计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers in Industry

Pub Date : 2024-09-23 DOI: 10.1016/j.compind.2024.104189

Muhammad Fawad , Marek Salamak , Qian Chen , Mateusz Uscilowski , Kalman Koris , Marcin Jasinski , Piotr Lazinski , Dawid Piotrowski

Conventional infrastructure asset management practices have heavily relied on static data collection and suffered from decision lags. Though advanced Structural Health Monitoring (SHM) systems were extensively explored based on multi-functional sensor deployment, asset model updating has not been achieved to facilitate timely and effective decision-making of infrastructure managers due to a lack of system integration. To address this challenge, this study develops the Immersive Bridge Digital Twin Platform (IBDTP) to allow infrastructure managers to automate the SHM processes of bridges and engage them in immersive decision-making processes based on Scan-to-BIM and Augmented Reality (AR) technologies. A novel 3D game engine is proposed as part of IBDTP and was tested using a single-span concrete arch bridge located in Poland. Results show that the measurement data collected and presented in IBDTP improves the infrastructure managers' accessibility to major damage data of the bridge to plan for future interventions. The functions of the IBDTP can be potentially scaled for different types of bridges and critical infrastructure, substantially improving the traditional SHM in terms of data management and 3D structural visualization.

传统的基础设施资产管理实践严重依赖静态数据收集，存在决策滞后问题。虽然基于多功能传感器部署的先进结构健康监测（SHM）系统得到了广泛探索，但由于缺乏系统集成，资产模型更新尚未实现，无法促进基础设施管理者做出及时有效的决策。为了应对这一挑战，本研究开发了沉浸式桥梁数字孪生平台（IBDTP），使基础设施管理者能够自动化桥梁的 SHM 流程，并让他们参与基于扫描到 BIM 和增强现实（AR）技术的沉浸式决策流程。作为 IBDTP 的一部分，提出了一个新颖的 3D 游戏引擎，并使用位于波兰的一座单跨混凝土拱桥进行了测试。结果表明，在 IBDTP 中收集和展示的测量数据提高了基础设施管理者对桥梁主要损坏数据的可访问性，从而为未来的干预措施制定计划。IBDTP 的功能可根据不同类型的桥梁和关键基础设施进行扩展，在数据管理和三维结构可视化方面大大改进了传统的 SHM。

{"title":"Development of immersive bridge digital twin platform to facilitate bridge damage assessment and asset model updates","authors":"Muhammad Fawad , Marek Salamak , Qian Chen , Mateusz Uscilowski , Kalman Koris , Marcin Jasinski , Piotr Lazinski , Dawid Piotrowski","doi":"10.1016/j.compind.2024.104189","DOIUrl":"10.1016/j.compind.2024.104189","url":null,"abstract":"<div><div>Conventional infrastructure asset management practices have heavily relied on static data collection and suffered from decision lags. Though advanced Structural Health Monitoring (SHM) systems were extensively explored based on multi-functional sensor deployment, asset model updating has not been achieved to facilitate timely and effective decision-making of infrastructure managers due to a lack of system integration. To address this challenge, this study develops the Immersive Bridge Digital Twin Platform (IBDTP) to allow infrastructure managers to automate the SHM processes of bridges and engage them in immersive decision-making processes based on Scan-to-BIM and Augmented Reality (AR) technologies. A novel 3D game engine is proposed as part of IBDTP and was tested using a single-span concrete arch bridge located in Poland. Results show that the measurement data collected and presented in IBDTP improves the infrastructure managers' accessibility to major damage data of the bridge to plan for future interventions. The functions of the IBDTP can be potentially scaled for different types of bridges and critical infrastructure, substantially improving the traditional SHM in terms of data management and 3D structural visualization.</div></div>","PeriodicalId":55219,"journal":{"name":"Computers in Industry","volume":"164 ","pages":"Article 104189"},"PeriodicalIF":8.2,"publicationDate":"2024-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S0166361524001179/pdfft?md5=a6ebf35ab4db5c14eeed0be759e6256a&pid=1-s2.0-S0166361524001179-main.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142313036","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Product digital twins: An umbrella review and research agenda for understanding their value 产品数字孪生：了解其价值的总体回顾和研究议程

IF 8.2 1区计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers in Industry

Pub Date : 2024-09-21 DOI: 10.1016/j.compind.2024.104181

Francisco Gomez Medina , Veronica Martinez Hernandez

Product Digital Twins (DTs) are digital representations of a physical asset that update synchronously throughout its lifecycle. Over the past decade, a rich and varied literature on the development of new technologies and approaches to implementing product DTs has emerged. This literature has been reviewed multiple times, but the variety in focus and scope of DT reviews has become so extensive that it is challenging to assess our collective understanding and knowledge of DT theory. We address this issue by conducting a systematic umbrella review of product DT reviews, classifying and analysing review themes to understand strengths and shortcomings of product DT literature. Our analysis reveals a key shortcoming in the product DT literature: There is currently little evidence and understanding of DT value. Understanding how DTs provide value to an organisation is of paramount importance, as it will determine the elements of the DT that truly have an effect on value, as well as the mechanisms by which that value is created. We conclude this work by presenting a five-item research agenda to address these shortcomings and develop our understanding of DT value. Since DTs can be complex and expensive to implement, research and practice should focus on those elements of the DT that provide value to the organisation.

产品数字孪生（DT）是物理资产的数字表示，在其整个生命周期中同步更新。在过去的十年中，出现了丰富多样的文献，介绍了新技术的发展和实施产品数字孪生的方法。这些文献已被多次评述，但 DT 评述的重点和范围已变得如此广泛，以至于评估我们对 DT 理论的集体理解和知识具有挑战性。为了解决这个问题，我们对产品 DT 评论进行了系统性的总体回顾，对评论主题进行了分类和分析，以了解产品 DT 文献的优势和不足。我们的分析揭示了产品研发文献中的一个关键缺陷：目前几乎没有关于 DT 价值的证据和理解。了解 DT 如何为组织提供价值至关重要，因为这将确定 DT 中真正对价值产生影响的要素，以及创造价值的机制。最后，我们提出了五项研究议程，以解决这些不足并加深我们对 DT 价值的理解。由于 DT 可能很复杂，实施起来也很昂贵，因此研究和实践应侧重于 DT 中那些能为组织带来价值的要素。

{"title":"Product digital twins: An umbrella review and research agenda for understanding their value","authors":"Francisco Gomez Medina , Veronica Martinez Hernandez","doi":"10.1016/j.compind.2024.104181","DOIUrl":"10.1016/j.compind.2024.104181","url":null,"abstract":"<div><p>Product Digital Twins (DTs) are digital representations of a physical asset that update synchronously throughout its lifecycle. Over the past decade, a rich and varied literature on the development of new technologies and approaches to implementing product DTs has emerged. This literature has been reviewed multiple times, but the variety in focus and scope of DT reviews has become so extensive that it is challenging to assess our collective understanding and knowledge of DT theory. We address this issue by conducting a systematic umbrella review of product DT reviews, classifying and analysing review themes to understand strengths and shortcomings of product DT literature. Our analysis reveals a key shortcoming in the product DT literature: There is currently little evidence and understanding of DT value. Understanding how DTs provide value to an organisation is of paramount importance, as it will determine the elements of the DT that truly have an effect on value, as well as the mechanisms by which that value is created. We conclude this work by presenting a five-item research agenda to address these shortcomings and develop our understanding of DT value. Since DTs can be complex and expensive to implement, research and practice should focus on those elements of the DT that provide value to the organisation.</p></div>","PeriodicalId":55219,"journal":{"name":"Computers in Industry","volume":"164 ","pages":"Article 104181"},"PeriodicalIF":8.2,"publicationDate":"2024-09-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S016636152400109X/pdfft?md5=9a355badb7559ca56d607a28d5927053&pid=1-s2.0-S016636152400109X-main.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142274794","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Automated corner grading of trading cards: Defect identification and confidence calibration through deep learning 交易卡的自动边角分级：通过深度学习进行缺陷识别和信心校准

IF 8.2 1区计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers in Industry

Pub Date : 2024-09-19 DOI: 10.1016/j.compind.2024.104187

Lutfun Nahar , Md. Saiful Islam , Mohammad Awrangjeb , Rob Verhoeve

This research focuses on trading card quality inspection, where defects have a significant effect on both the quality inspection and grading. The present inspection procedure is subjective which means the grading is sensitive to mistakes made by individuals. To address this, a deep neural network based on transfer learning for automated defect detection is proposed with a particular emphasis on corner grading which is a crucial factor in overall card grading. This paper presents an extension of our prior study, in which we achieved an accuracy of 78% by employing the VGG-net and InceptionV3 models. In this study, our emphasis is on the DenseNet model where convolutional layers are used to extract features and regularisation methods including batch normalisation and spatial dropout are incorporated for better defect classification. Our approach outperformed prior findings, as evidenced by experimental results based on a real dataset provided by our industry partner, achieving an 83% mean accuracy in defect classification. Additionally, this study investigates various calibration approaches to fine-tune the model confidence. To make the model more reliable, a rule-based approach is incorporated to classify defects based on confidence scores. Finally, a human-in-the-loop system is integrated to inspect the misclassified samples. Our results demonstrate that the model’s performance and confidence are expected to improve further when a large number of misclassified samples, along with human feedback, are used to retrain the network.

这项研究的重点是交易卡质量检验，缺陷对质量检验和评级都有重大影响。目前的检测程序是主观的，这意味着分级对个人所犯的错误很敏感。为了解决这个问题，我们提出了一种基于迁移学习的深度神经网络，用于自动缺陷检测，并特别强调了角分级，因为角分级是整个卡片分级的关键因素。我们在之前的研究中采用了 VGG 网络和 InceptionV3 模型，准确率达到了 78%。在本研究中，我们将重点放在 DenseNet 模型上，该模型使用卷积层提取特征，并采用包括批量归一化和空间剔除在内的正则化方法，以获得更好的缺陷分类效果。基于行业合作伙伴提供的真实数据集的实验结果表明，我们的方法优于之前的研究成果，缺陷分类的平均准确率达到 83%。此外，本研究还探讨了各种校准方法，以微调模型的置信度。为了使模型更加可靠，我们采用了一种基于规则的方法，根据置信度分数对缺陷进行分类。最后，还集成了一个人工在环系统，以检查分类错误的样本。我们的研究结果表明，如果使用大量的错误分类样本和人工反馈来重新训练网络，模型的性能和置信度有望进一步提高。

{"title":"Automated corner grading of trading cards: Defect identification and confidence calibration through deep learning","authors":"Lutfun Nahar , Md. Saiful Islam , Mohammad Awrangjeb , Rob Verhoeve","doi":"10.1016/j.compind.2024.104187","DOIUrl":"10.1016/j.compind.2024.104187","url":null,"abstract":"<div><p>This research focuses on trading card quality inspection, where defects have a significant effect on both the quality inspection and grading. The present inspection procedure is subjective which means the grading is sensitive to mistakes made by individuals. To address this, a deep neural network based on transfer learning for automated defect detection is proposed with a particular emphasis on corner grading which is a crucial factor in overall card grading. This paper presents an extension of our prior study, in which we achieved an accuracy of 78% by employing the VGG-net and InceptionV3 models. In this study, our emphasis is on the DenseNet model where convolutional layers are used to extract features and regularisation methods including batch normalisation and spatial dropout are incorporated for better defect classification. Our approach outperformed prior findings, as evidenced by experimental results based on a real dataset provided by our industry partner, achieving an 83% mean accuracy in defect classification. Additionally, this study investigates various calibration approaches to fine-tune the model confidence. To make the model more reliable, a rule-based approach is incorporated to classify defects based on confidence scores. Finally, a human-in-the-loop system is integrated to inspect the misclassified samples. Our results demonstrate that the model’s performance and confidence are expected to improve further when a large number of misclassified samples, along with human feedback, are used to retrain the network.</p></div>","PeriodicalId":55219,"journal":{"name":"Computers in Industry","volume":"164 ","pages":"Article 104187"},"PeriodicalIF":8.2,"publicationDate":"2024-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S0166361524001155/pdfft?md5=94620ae9f7ac6add13e46e3c2ecef436&pid=1-s2.0-S0166361524001155-main.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142274793","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0