IF 5.2 2区工程技术

Measurement Pub Date : 2025-06-30 DOI: 10.1016/j.measurement.2025.118213

Yongsheng Yao , Chen Liu , Jue Li , Jinliang Wu

{"title":"Transformer-based data generation and lightweight robust detection network for complex pavement defects","authors":"Yongsheng Yao , Chen Liu , Jue Li , Jinliang Wu","doi":"10.1016/j.measurement.2025.118213","DOIUrl":"10.1016/j.measurement.2025.118213","url":null,"abstract":"<div><div>Ground-penetrating radar (GPR) accurate detection of road surface hidden defects is very important for road maintenance. However, due to the complexity of disease waveforms and the scarcity of sample data, existing models face problems such as insufficient accuracy and have weak generalization ability in engineering scenarios. To this end, this paper proposes a two-stage method for improving the accuracy of complex pavement diseases with small-scale data. Firstly, a B-scan Latent Generative Adversarial Network (BL-GAN) was proposed to effectively expand the small-scale dataset by combining the self-attention mechanism and hierarchical feature fusion to synthesize realistic defect patterns. The t-SNE visualization confirms the high feature agreement between the synthetic and real defect images. Secondly, by systematically integrating MobileViTv2 and the channel attention mechanism, a lightweight transformer-based detector ECA-MobileViTv2 YOLOv5s(EV2-YOLOv5s) is designed to achieve accurate localization of multi-scale defects while maintaining computational efficiency. Experimental results show that the proposed method achieves 96.5% <em>[email protected]</em> with 144.1 <em>FPS</em> (GPU) inference speed, 19.7 <em>FPS</em> (CPU) inference speed, and 7.4 MB model size, which is significantly better than traditional methods. Finally, the contributions of Multi-head Attention (MHA) and Linearly Separable Self-Attention (LSA) in global feature aggregation and local feature extraction were analyzed to support the subsequent optimization of Transformer-based defect detection algorithms. This work provides a practical solution for the detection of hidden road surface diseases in resource-constrained scenarios.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"256 ","pages":"Article 118213"},"PeriodicalIF":5.2,"publicationDate":"2025-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144516925","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

Cost-efficient PON fault localization utilizing Wilcoxon signed rank test 基于Wilcoxon符号秩检验的高效PON故障定位

IF 5.2 2区工程技术

Measurement Pub Date : 2025-06-30 DOI: 10.1016/j.measurement.2025.118314

Jianfeng Deng, Xuan Zhang, Xinhong Jia

{"title":"Cost-efficient PON fault localization utilizing Wilcoxon signed rank test","authors":"Jianfeng Deng, Xuan Zhang, Xinhong Jia","doi":"10.1016/j.measurement.2025.118314","DOIUrl":"10.1016/j.measurement.2025.118314","url":null,"abstract":"<div><div>We introduce a novel fault localization method for passive optical networks (PONs) based on the Wilcoxon signed rank test. This approach seamlessly integrates with existing PON monitoring techniques that rely on time-domain recognition, enabling the extraction of latent features from time-domain traces used in fault detection. It effectively addresses a key limitation of traditional methods, which are unable to utilize amplitude data for fault localization. Our approach accurately identifies features corresponding to fault point data regions and maps these positions onto acquired time-domain traces through a constructed model, thereby enabling secondary analysis of the trace data. This innovation overcomes the limitations inherent in most existing PON monitoring techniques, which can detect faults but not localize them, achieving fault localization without the need for additional components in the original PON monitoring system. Experimental validation conducted on a PON monitoring system based on time-domain recognition demonstrates the feasibility of the proposed approach. Furthermore, we compare the results obtained from our method with those from OTDR tests, confirming that our approach achieves high localization accuracy with only 64 averages. Since the data utilized for fault localization is derived from the fault detection phase itself rather than from external sources, this method significantly facilitates integrated design and enhances measurement efficiency. This cost-effective solution holds significant promise in the PON market, offering extensive applicability and considerable value in enhancing fault localization capabilities.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"256 ","pages":"Article 118314"},"PeriodicalIF":5.2,"publicationDate":"2025-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144524226","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

Evaluation on tensile and shear failure characteristics and mechanisms of cement-asphalt interface in semi-flexible paving materials (SFPM) using innovative interface measurement approaches 基于创新界面测量方法的半柔性铺装材料水泥-沥青界面拉剪破坏特征及机制评价

IF 5.2 2区工程技术

Measurement Pub Date : 2025-06-30 DOI: 10.1016/j.measurement.2025.118299

Jiashu Li , Di Li , Yinshan Xu , Kangqing Yang , Jiaolong Ren

{"title":"Evaluation on tensile and shear failure characteristics and mechanisms of cement-asphalt interface in semi-flexible paving materials (SFPM) using innovative interface measurement approaches","authors":"Jiashu Li , Di Li , Yinshan Xu , Kangqing Yang , Jiaolong Ren","doi":"10.1016/j.measurement.2025.118299","DOIUrl":"10.1016/j.measurement.2025.118299","url":null,"abstract":"<div><div>The interaction between the inorganic components of cement phase and the organic components of asphalt phase tends to form a weak interface in the SFPM. However, characteristics of cement-asphalt interface in the SFPM remain unclear owing to a lack of direct and effective measurement methods for the interface. Hence, an interfacial tensile measurement approach and a 45° interfacial shear measurement approach are developed to investigate tensile and shear properties and failure mechanisms of this interface in the SFPM, considering various asphalt binders and interface modifiers. Results show that: the modification effects of interface modifiers are particularly pronounced at temperatures below 0 °C. The application of various interface modifiers results in enhancements of 20–30 % in tensile resistance and 6–8 % in shear resistance at the interface. Among these modifiers, cationic emulsified asphalt demonstrates the most substantial effect. As the temperature increases, the modification effects on the asphalt binder become more pronounced. Notably, the interface modified with high-viscosity modified asphalt binder exhibits the most significant improvements, with tensile resistance increasing by over 50 % and average shear strength rising by 38 %. Additionally, an examination of interface failure mechanisms reveals that, under low-temperature conditions, failure predominantly occurs at the cement phase, whereas at higher temperatures, cohesive failure of the asphalt phase becomes the primary failure mode.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"256 ","pages":"Article 118299"},"PeriodicalIF":5.2,"publicationDate":"2025-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144524210","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

Enhancing leather defect detection with cross-stage fused attention partial convolution network 跨阶段融合注意部分卷积网络增强皮革缺陷检测

IF 5.2 2区工程技术

Measurement Pub Date : 2025-06-30 DOI: 10.1016/j.measurement.2025.118177

Zhiqiang Chen , Feng He , Daxing Xu , Hailun Wang , Jiehang Deng , Yi Chen , Chuan Li

{"title":"Enhancing leather defect detection with cross-stage fused attention partial convolution network","authors":"Zhiqiang Chen , Feng He , Daxing Xu , Hailun Wang , Jiehang Deng , Yi Chen , Chuan Li","doi":"10.1016/j.measurement.2025.118177","DOIUrl":"10.1016/j.measurement.2025.118177","url":null,"abstract":"<div><div>Leather surface defects are in various colours, shapes, and sizes. Leather products enterprises usually require low-cost edge computing and embedded devices. These factors pose significant challenges to machine vision-based leather surface defect detection. Under limited computing resources, it is hoped that the learning ability of visual models is enhanced to maintain sufficient accuracy while being lightweight. To this end, the Cross Stage Fused Attention Partial Convolution Network (CSFAPCNet) is proposed which integrates multiple lightweight attention mechanisms to improve feature learning while combining cross-stage partial connections and partial convolutions to reduce computational complexity. Research was conducted on leather defect detection from three levels: leather anomaly detection, multi-type defect detection, and simultaneous detection of positioning and classification. Ablation experiments and generalization performance analysis were also performed. Systematic and in-depth experiments have shown that CSFAPCNet outperforms state-of-the-art methods in various evaluation metrics, and maintains sufficient accuracy while significantly reducing computational complexity.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"256 ","pages":"Article 118177"},"PeriodicalIF":5.2,"publicationDate":"2025-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144516926","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

Design of a double-gradient scanner apparatus to assess the uncertainty component associated with inhomogeneity in flexible wire thermocouples 双梯度扫描装置的设计，以评估柔性线热电偶中与不均匀性相关的不确定度成分

IF 5.2 2区工程技术

Measurement Pub Date : 2025-06-30 DOI: 10.1016/j.measurement.2025.118184

Gilvan César de Oliveira , Matheus Pereira Porto , Rafael Augusto Magalhães Ferreira

{"title":"Design of a double-gradient scanner apparatus to assess the uncertainty component associated with inhomogeneity in flexible wire thermocouples","authors":"Gilvan César de Oliveira , Matheus Pereira Porto , Rafael Augusto Magalhães Ferreira","doi":"10.1016/j.measurement.2025.118184","DOIUrl":"10.1016/j.measurement.2025.118184","url":null,"abstract":"<div><div>Current thermocouple calibration standards identify primary sources of uncertainty, but often lack systematic methodologies for their evaluation. As a result, practitioners frequently rely on generic normative references, which can compromise metrological quality and increase measurement uncertainty. This study develops and validates an experimental setup for assessing calibration uncertainty components related to the thermoelectric inhomogeneity of flexible thermocouple wires. The system features adjustable temperature gradients and a precision movement mechanism, comprising inclined metallic clamps mounted on linear rails, driven by stepper motors, and controlled via an Arduino board. A uniform wire displacement is ensured by an aluminum insert within a metrology well-characterized by known axial thermal homogeneity. The inhomogeneity scanner was validated using tests that simulate artificially aged segments ranging from 15 mm to 400 mm in type T flexible wires, with gauges from 24 AWG to 30 AWG. The results confirmed the system’s ability to detect and quantify thermoelectric inhomogeneity, even under varying artificial aging conditions. Furthermore, the experimentally determined uncertainty values were compared with estimates from the EURAMET Calibration Guide No. 8, demonstrating the robustness and reliability of the proposed approach. This solution provides a cost-effective and easily implementable alternative for calibrating new and used thermocouple wires, ensuring compliance with normative requirements while enhancing metrological quality in temperature measurements.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"256 ","pages":"Article 118184"},"PeriodicalIF":5.2,"publicationDate":"2025-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144516927","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

Calculation method of wilting index for the soybean canopy based on the three-dimensional Fourier transform 基于三维傅里叶变换的大豆冠层萎蔫指数计算方法

IF 5.2 2区工程技术

Measurement Pub Date : 2025-06-30 DOI: 10.1016/j.measurement.2025.118301

Xiaodan Ma, Yingying Cheng, Haiou Guan, Hongyu Lin

{"title":"Calculation method of wilting index for the soybean canopy based on the three-dimensional Fourier transform","authors":"Xiaodan Ma, Yingying Cheng, Haiou Guan, Hongyu Lin","doi":"10.1016/j.measurement.2025.118301","DOIUrl":"10.1016/j.measurement.2025.118301","url":null,"abstract":"<div><div>Drought stress significantly affecting soybean yield, thus, rapid detection of phenotypic traits for the soybean canopy is crucial for the breeding and cultivation. Traditional methods in the determination of soybean wilting index have the problems of cumbersome and time consuming, thus, a calculation method of wilting index was proposed for the soybean canopy based on three-dimensional (3D) Fourier transform. First, the Kinect 2.0 sensor was used to acquire the 3D point cloud data of the soybean canopy, and the filter integration algorithm was applied to remove the invalid point cloud. Second, a height-based slicing algorithm was used to convert the 3D spatial structure of the canopy into a 2D image sequences. Third, a constraint function based on spectral features was constructed to select the optimal sequence position as 1/2 of the canopy, which realized a 3D Fourier transform algorithm of the soybean canopy. Finally, a 3D Fourier transform-based soybean wilting index was proposed based on the proportion difference of the direct component (DC) and the spectral radius energy to the total energy in the 3D canopy based on the normal water treatment (NWT) and the drought stress treatment (DST). The result showed that the test coefficient <span><math><mrow><mi>p</mi></mrow></math></span> of the significant difference reached 1.12 × 10<sup>−7</sup> under NWT and DST, and the determination coefficients <span><math><mrow><msup><mi>R</mi><mn>2</mn></msup></mrow></math></span> between the calculated wilting index and the corresponding mean leaf inclination were all above 0.83. The research result can accurately characterize the wilting degree, and provide technical support for irrigation, seed examination of the drought soybean plant.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"256 ","pages":"Article 118301"},"PeriodicalIF":5.2,"publicationDate":"2025-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144524223","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

High-precision calibration method for the multiview measurement of the mirror–assisted digital image correlation 反光镜辅助数字图像相关多视点测量的高精度标定方法

IF 5.2 2区工程技术

Measurement Pub Date : 2025-06-29 DOI: 10.1016/j.measurement.2025.118283

Pengxiang Ge , Xizuo Dan , Haoyu Wang , Haoran Gao , Yonghong Wang , Guihua Li

{"title":"High-precision calibration method for the multiview measurement of the mirror–assisted digital image correlation","authors":"Pengxiang Ge , Xizuo Dan , Haoyu Wang , Haoran Gao , Yonghong Wang , Guihua Li","doi":"10.1016/j.measurement.2025.118283","DOIUrl":"10.1016/j.measurement.2025.118283","url":null,"abstract":"<div><div>The plane mirror-assisted digital image correlation (DIC) technique facilitates multiview measurement, but the technology is hindered by low utilization of the plane mirror, contamination of mirror surfaces, and low calibration accuracy. To address these issues, a high-precision calibration method that employs a gauge block and fluorescent speckles has been proposed. The method establishes a symmetric relationship about the plane mirror through the fluorescent speckles on the front and rear working surfaces of the gauge block. Additionally, the gauge block is also used to calibrate the accuracy of the posture of the plane mirror during the coordinate transformation process. To verify the performance of the method, morphology reconstruction experiments of complex surfaces, dynamic experiments, and comparative analysis are performed. The results show that the reconstruction accuracy of the plane mirror (size:<span><math><mrow><mn>150</mn><mo>×</mo><mn>150</mn><mo>×</mo><mn>1.5</mn></mrow></math></span> mm) after calibration was improved by 0.08 mm over the direct reconstruction accuracy of the DIC, with significant improvements in both dynamic and static measurement accuracy. With high calibration accuracy and convenient operation, this method shows great potential for application in plane mirror-assisted optical measurements.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"256 ","pages":"Article 118283"},"PeriodicalIF":5.2,"publicationDate":"2025-06-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144518112","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

3D ergonomics parameter measurement using video-based deep learning 基于视频的深度学习的三维人体工程学参数测量

IF 5.2 2区工程技术

Measurement Pub Date : 2025-06-28 DOI: 10.1016/j.measurement.2025.118034

Xiaojing Zhou, Wei Wang, Yanzhong Hu

{"title":"3D ergonomics parameter measurement using video-based deep learning","authors":"Xiaojing Zhou, Wei Wang, Yanzhong Hu","doi":"10.1016/j.measurement.2025.118034","DOIUrl":"10.1016/j.measurement.2025.118034","url":null,"abstract":"<div><div>In contrast to strapping sensors or scanning with a 3D scanner, obtaining ergonomic parameters by taking a video of a single person is undoubtedly one of the simplest, most efficient, and most comfortable measurement methods. In this paper, we propose an ergonomic parameter measuring method based on extracting the keypoints of the human body in the video using deep learning techniques,aiming to achieve high accuracy under simple hardware conditions. A YOLOv8-FSC-Pose network model is developed for human target detection and joint point localization in binocular video. Kalman filtering compensates for joint point outliers caused by occlusion or noise in each frame. Then the optimized DeepLabv3+ semantic segmentation network is used to extract a complete and fine human motion contour from the target region. We extract the head and foot points from the contour required to measure human parameters. Based on ergonomic principles, we designed specific calculation methods to determine walking height, upper and lower limb range of motion (ROM), and gait parameters by analyzing the three-dimensional coordinates extracted from human keypoints. The error analysis results show that this method can fully apply to the real-time measurement of human parameters in complex outdoor scenes, with the advantages of good real-time, ease of operation, and high accuracy.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"256 ","pages":"Article 118034"},"PeriodicalIF":5.2,"publicationDate":"2025-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144516920","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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 quantification method for model parameter uncertainty based on an improved sequential Monte Carlo filter 基于改进序贯蒙特卡罗滤波的模型参数不确定性量化方法

IF 5.2 2区工程技术

Measurement Pub Date : 2025-06-28 DOI: 10.1016/j.measurement.2025.118287

Guizhong Xie , Yanhang Liu , Hao Li , Hongrui Geng , Shenshen Chen , Jie Zhou

{"title":"A quantification method for model parameter uncertainty based on an improved sequential Monte Carlo filter","authors":"Guizhong Xie , Yanhang Liu , Hao Li , Hongrui Geng , Shenshen Chen , Jie Zhou","doi":"10.1016/j.measurement.2025.118287","DOIUrl":"10.1016/j.measurement.2025.118287","url":null,"abstract":"<div><div>The quantification of model parameter uncertainty is of great importance for the safety of engineering structures. However, traditional particle filter methods face challenges in avoiding particle degradation and impoverishment when estimating and predicting model parameters. Thus, this paper uses three improved particle filters combined with a Gaussian mixture model to evaluate the uncertainty of the model parameters. Fatigue crack growth model is established based on the Paris equation, and the performance of the filter is validated through two numerical examples with engineering backgrounds. Numerical example 1 analyzes a pressure vessel with a central crack, comparing the performance of different filters in estimating crack length, stress intensity factor, and posterior parameter distribution. By comparing the performance of PFGM, IBIS, and SMC under 10<sup>7</sup> loading cycles, it was found that SMC, compared to the other two methods, effectively reduces the estimation error of the posterior parameter distribution, with the error not exceeding 3%. Additionally, the time consumption of the resampling process was reduced by 12.7%. Numerical example 2 studied the central crack model of a Q235 steel plate, combining the K-L expansion method to quantify the material parameter random field. The results show that SMC outperforms the other filtering methods in terms of dynamic adaptability, prediction accuracy (RMSE reduced by 23.5%), and computational efficiency. In summary, SMC, through its adaptive resampling method, significantly reduces computation time and improves prediction accuracy in parameter estimation. It effectively quantifies the uncertainty of structural model parameters, enhancing the precision and stability of uncertainty assessments, thus providing a reliable tool for uncertainty quantification in large-scale structural health monitoring.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"256 ","pages":"Article 118287"},"PeriodicalIF":5.2,"publicationDate":"2025-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144524209","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

GeoContrast: Geometric knowledge-based contrast learning for industrial point cloud segmentation GeoContrast：用于工业点云分割的基于几何知识的对比学习

IF 5.2 2区工程技术

Measurement Pub Date : 2025-06-28 DOI: 10.1016/j.measurement.2025.118204

Xueqi Wang , Yinhua Liu , Yinan Wang , Yanzheng Li , Daqiang Zhang

{"title":"GeoContrast: Geometric knowledge-based contrast learning for industrial point cloud segmentation","authors":"Xueqi Wang , Yinhua Liu , Yinan Wang , Yanzheng Li , Daqiang Zhang","doi":"10.1016/j.measurement.2025.118204","DOIUrl":"10.1016/j.measurement.2025.118204","url":null,"abstract":"<div><div>Process simulation of industrial robots is essential for improving operational accuracy and efficiency. Due to their three-dimensional structure, point clouds show great promise for building high-fidelity simulation environments. However, the absence of semantic information in unstructured point clouds exacerbates the consistency issue between physical entities and the corresponding virtual models. Several factors hinder improvements in the accuracy of point cloud semantic segmentation. These include the variability of process equipment with a high degree of freedom, the inherent complexity of industrial environments, and the scarcity of high-quality annotated data. To bridge this gap, this paper proposes a novel contrast pre-training framework, GeoContrast, tailored for point cloud segmentation in industrial environments. Firstly, a geometric knowledge module generates geometrically consistent features through clustering and multidimensional feature encoding, improving the segmentation accuracy and generalization ability of high degree of freedom entities. Secondly, an Upsampling Attention Module is implemented to efficiently fuse multi-layer features, significantly enhancing the robustness of entity boundary segmentation. Finally, an uncertainty-weighted contrast loss function is proposed that incorporates information entropy to dynamically adjust sample weights, improving the model’s ability to discern entity boundaries in complex industrial environments. The effectiveness of the proposed method was validated using the Actual Welding Scene Scanning (AWSS) dataset and the Stanford 3D Indoor Scene (S3DIS) dataset. Results demonstrated that the proposed method outperformed existing methods, achieving a maximum improvement of 45.2% in a mean intersection over Union (mIoU) on the AWSS dataset using only 1% of the labels. In addition, it also achieved excellent performance on S3DIS, with mIoU of 75.3%.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"256 ","pages":"Article 118204"},"PeriodicalIF":5.2,"publicationDate":"2025-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144516923","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

Measurement最新文献