Journal of Nondestructive Evaluation最新文献_第6页

Research on Infrasound-based Pipeline Leakage Signal Identification Using Random Forest 基于随机森林的管道泄漏信号次声识别研究

IF 3 3区材料科学

Journal of Nondestructive Evaluation Pub Date : 2026-06-30 DOI: 10.1007/s10921-026-01391-1

Wenjing Du, Yongmei Hao, Mudan Yang, Jun Shen, Fei Li

{"title":"Research on Infrasound-based Pipeline Leakage Signal Identification Using Random Forest","authors":"Wenjing Du, Yongmei Hao, Mudan Yang, Jun Shen, Fei Li","doi":"10.1007/s10921-026-01391-1","DOIUrl":"10.1007/s10921-026-01391-1","url":null,"abstract":"<div>To address the challenges of identifying small leaks in buried pipelines and the low detection accuracy of infrasound leak signals, a multi-feature collaborative leakage discrimination framework is constructed signals. First, 12-dimensional leak features are extracted, and a Mutual Information optimized Random Forest (MI-RF) is introduced to perform preliminary feature screening. On this basis, a Full Subset Optimal Search—Mahalanobis Distance Adaptive Discrimination Mechanism (MDFSS) is proposed to establish a leak classification and discrimination framework. The experimental results show that for sudden leakage, the optimal feature combination is Peak-to-Peak Value and Mean Frequency, achieving an average identification Accuracy of 96.33% across different apertures; for persistent leakage, the optimal feature combination is Root Mean Square and Standard Deviation of Frequency, achieving an average identification accuracy of 94.17%. Ablation experiments verify the advantages of each stage of the MI-RF-MDFSS framework. Under the typical 2 mm persistent leakage condition, the proposed method, relying on the optimal feature combination, achieves a leak identification accuracy of 96.5%, outperforming traditional discrimination methods whose accuracy ranges only between 92 and 95%. Future work will further validate the generalization capability and engineering applicability of this method in complex industrial noise environments.</div>","PeriodicalId":655,"journal":{"name":"Journal of Nondestructive Evaluation","volume":"45 3","pages":""},"PeriodicalIF":3.0,"publicationDate":"2026-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148380812","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A Lightweight GNN-SR Framework for Artifact-Resilient CT Imaging in Industrial Nondestructive Evaluation 工业无损评估中伪影弹性CT成像的轻量级GNN-SR框架

IF 3 3区材料科学

Journal of Nondestructive Evaluation Pub Date : 2026-06-30 DOI: 10.1007/s10921-026-01392-0

Ying Zhou, Mei Choo Ang, Ummul Hanan Mohamad, Bo Ao, Ah-Lian Kor, Kok Weng Ng

{"title":"A Lightweight GNN-SR Framework for Artifact-Resilient CT Imaging in Industrial Nondestructive Evaluation","authors":"Ying Zhou, Mei Choo Ang, Ummul Hanan Mohamad, Bo Ao, Ah-Lian Kor, Kok Weng Ng","doi":"10.1007/s10921-026-01392-0","DOIUrl":"10.1007/s10921-026-01392-0","url":null,"abstract":"<div>Turbine blades in aero-engines contain intricate internal cooling structures embedded within dense superalloy materials, making reliable non-destructive evaluation (NDE) particularly challenging. Industrial computed tomography (CT) is widely used for internal inspection; however, metal-induced artifacts and resolution degradation often disrupt structural continuity, especially in sub-millimeter regions, limiting defect visibility and measurement accuracy. To address this challenge, this study proposes a lightweight, two-stage structure-aware CT image restoration framework tailored for industrial inspection. In the first stage, a graph neural network (GNN)-based artifact suppression module is developed to capture non-local structural dependencies via dilated convolutions and residual refinement, thereby enabling effective mitigation of streaks, rings, and beam-hardening artifacts. In the second stage, a pre-trained Enhanced Super-Resolution Generative Adversarial Network (ESRGAN) is applied as a fixed post-processing module to enhance high-frequency details and local contrast without domain-specific retraining. Experimental results on real turbine blade CT data demonstrate that the proposed framework outperforms conventional filtering methods and convolutional baselines, including CNNs and U-Net, in terms of both PSNR and SSIM. The results further indicate that structure-aware restoration is essential for maintaining image fidelity under severe artifact conditions. By improving structural clarity, the proposed method may support downstream tasks such as defect localization and dimensional verification in industrial inspection workflows.</div>","PeriodicalId":655,"journal":{"name":"Journal of Nondestructive Evaluation","volume":"45 3","pages":""},"PeriodicalIF":3.0,"publicationDate":"2026-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148380811","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Machine Learning-Assisted Fiber Bragg Grating Humidity Sensor for Nondestructive Condition Monitoring 用于无损状态监测的机器学习辅助光纤光栅湿度传感器

IF 3 3区材料科学

Journal of Nondestructive Evaluation Pub Date : 2026-06-26 DOI: 10.1007/s10921-026-01394-y

Ahmad Adel Abu-Shareha, K. Narsimha Reddy, Manoranjan Parhi, Mahendihasan S. Heera, B. Jayaprakash, Sudeep Varshney, B. Bharathi, Sumit Sharma, Devendra Singh

{"title":"Machine Learning-Assisted Fiber Bragg Grating Humidity Sensor for Nondestructive Condition Monitoring","authors":"Ahmad Adel Abu-Shareha, K. Narsimha Reddy, Manoranjan Parhi, Mahendihasan S. Heera, B. Jayaprakash, Sudeep Varshney, B. Bharathi, Sumit Sharma, Devendra Singh","doi":"10.1007/s10921-026-01394-y","DOIUrl":"10.1007/s10921-026-01394-y","url":null,"abstract":"<div>Fiber Bragg Grating (FBG)-based sensors have emerged as reliable optical devices for monitoring humidity owing to their wavelength-encoded response, compactness, and immunity to electromagnetic interference. Accurate humidity measurement is a critical requirement in nondestructive evaluation (NDE) and condition monitoring, as moisture strongly influences the degradation and long-term integrity of polymer-based and composite materials. However, the intrinsic humidity sensitivity of bare FBGs is minimal and can be enhanced using a polymer coating that swells upon moisture absorption, inducing Bragg wavelength shifts. This study presents an optical modeling and simulation framework for a polymer-coated FBG humidity sensor, supported by machine learning (ML) algorithms to address nonlinear polymer swelling, hysteresis, and temperature cross-sensitivity. Theoretical formulations governing Bragg wavelength variation were implemented for polyimide-coated FBGs under varying relative humidity levels (0–100% RH). Reflectivity spectra were simulated using the coupling coefficient model and analyzed using Random Forest (RF) and Support Vector Regression (SVR) algorithms. The results demonstrate that ML-assisted modeling effectively captures the nonlinear optical response, yielding significant improvements in sensitivity and accuracy. The proposed approach contributes to nondestructive environmental condition monitoring by improving the reliability of optical fiber sensing systems, thereby supporting material health assessment and structural monitoring applications.</div>","PeriodicalId":655,"journal":{"name":"Journal of Nondestructive Evaluation","volume":"45 3","pages":""},"PeriodicalIF":3.0,"publicationDate":"2026-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148323828","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Metal Magnetic Memory Testing Signals Response to Alternating Stress during On-line Measurement of Ferromagnetic Steel 铁磁性钢在线测量中金属磁记忆检测信号对交变应力的响应

IF 3 3区材料科学

Journal of Nondestructive Evaluation Pub Date : 2026-06-18 DOI: 10.1007/s10921-025-01323-5

Huipeng Wang, Jialong Shi, Qiaogen Wang, Haihang Li, Lihong Dong, Haidou Wang, Huizhong Liu

{"title":"Metal Magnetic Memory Testing Signals Response to Alternating Stress during On-line Measurement of Ferromagnetic Steel","authors":"Huipeng Wang, Jialong Shi, Qiaogen Wang, Haihang Li, Lihong Dong, Haidou Wang, Huizhong Liu","doi":"10.1007/s10921-025-01323-5","DOIUrl":"10.1007/s10921-025-01323-5","url":null,"abstract":"<div>Metal magnetic memory testing (MMMT) has attracted considerable interest for its capability in stress concentration evaluation. An experimental investigation of the response of MMMT signals to the alternating stress in a single tension–compression stress cycle was carried out in Q235 steel. The normal component of the MMMT signals, Bz was an inclined line throughout the test and the tangential component of the MMMT signals, Bx was horizontal. The slope of Bz, KBz, and the mean value of Bx, Bxave, were extracted and the variation behaviour was presented and analyzed. During the tensile loading stage, KBz and Bxave initially increased with the applied stress, and then decreased. During the tensile unloading stage, KBz and Bxave did not change in the original manner, but continued to increase, and then decreased when the stress was completely released. KBz and Bxave kept decreasing during the compressive loading stage. While KBz and Bxave continued to increase throughout the compressive unloading stage, and they were smaller than those when the applied stress was the same during the compressive loading stage. When the compressive stress was completely unloaded, KBz and Bxave were slightly larger than when the tensile stress was completely unloaded. The results indicate that in on-line MMMT for stress evaluation, the stress state for each measurement (i.e., the applied stress that the workpiece was subjected to when performing MMMT) should be consistent to ensure the accuracy of stress evaluation.</div>","PeriodicalId":655,"journal":{"name":"Journal of Nondestructive Evaluation","volume":"45 3","pages":""},"PeriodicalIF":3.0,"publicationDate":"2026-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148281038","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Intelligent Micro-Cracks Identification in Acoustic Emission Based on Moment Tensor Theory and Convolutional Neural Networks 基于矩张量理论和卷积神经网络的声发射微裂纹智能识别

IF 3 3区材料科学

Journal of Nondestructive Evaluation Pub Date : 2026-06-18 DOI: 10.1007/s10921-026-01396-w

Xuewei Liu, Chuang Jia, Bin Liu, Yuanguang Zhu, Quansheng Liu

{"title":"Intelligent Micro-Cracks Identification in Acoustic Emission Based on Moment Tensor Theory and Convolutional Neural Networks","authors":"Xuewei Liu, Chuang Jia, Bin Liu, Yuanguang Zhu, Quansheng Liu","doi":"10.1007/s10921-026-01396-w","DOIUrl":"10.1007/s10921-026-01396-w","url":null,"abstract":"<div>The identification of crack propagation processes and microcrack types in surrounding rocks is of great importance for ensuring the safety and stability of underground engineering during subterranean resource extraction and underground space utilization. In this study, a deep learning framework integrating moment tensor (MT) theory and convolutional neural networks (CNN), termed MT-CNN, is proposed for automatic microcrack classification. Acoustic emission (AE) waveform-based representations are used as model inputs, while the corresponding crack types determined by MT inversion serve as classification labels. A series of uniaxial compression experiments was conducted to establish the AE database. Subsequently, three types of AE signal representations and four CNN architectures, including MobileNetV2, InceptionV3, ResNet50, and VGG16, were employed to construct and evaluate the MT-CNN models in order to identify the optimal framework. By analyzing the learning curves and classification performance, the results indicate that the VGG16 model combined with continuous wavelet transform (CWT) time–frequency representations achieves the best performance, with a test accuracy of 84.02%. Furthermore, a multi-metric evaluation framework incorporating accuracy, ROC curve, F1-score, Cohen’s kappa coefficient, and Hamming loss was introduced for comprehensive model assessment. The results demonstrate that the CWT-VGG16 model achieves the highest overall score of 0.82, confirming its superiority among all tested configurations.</div>","PeriodicalId":655,"journal":{"name":"Journal of Nondestructive Evaluation","volume":"45 3","pages":""},"PeriodicalIF":3.0,"publicationDate":"2026-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10921-026-01396-w.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148281049","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

SAHI-Enhanced YOLOv11 with Dual-Stage Label Fusion for Intelligent Weld Defect Detection in X-Ray Radiographs 采用sahi增强的YOLOv11双级标签融合技术，用于x射线图像中焊缝缺陷的智能检测

IF 3 3区材料科学

Journal of Nondestructive Evaluation Pub Date : 2026-06-16 DOI: 10.1007/s10921-026-01384-0

Jiaying Yan, Gefei Kuang, Chenggang Wang, Rui Feng, Linhao Wang, Jiahao Zhang, Mingjun Zhang, Tao Liu

{"title":"SAHI-Enhanced YOLOv11 with Dual-Stage Label Fusion for Intelligent Weld Defect Detection in X-Ray Radiographs","authors":"Jiaying Yan, Gefei Kuang, Chenggang Wang, Rui Feng, Linhao Wang, Jiahao Zhang, Mingjun Zhang, Tao Liu","doi":"10.1007/s10921-026-01384-0","DOIUrl":"10.1007/s10921-026-01384-0","url":null,"abstract":"<div>In response to the challenges associated with radiographic weld inspection, such as multi-scale defects, image quality heterogeneity, and background interference, this study introduces an advanced weld defect detection framework. This framework integrates the YOLOv11 detector with SAHI slicing. A weld region extraction strategy employing watershed segmentation and frequency-domain filtering is utilized to suppress irrelevant background elements. Concurrently, an adaptive enhancement pipeline, which combines median filtering, non-local means denoising, and CLAHE, is implemented to improve image quality. The SAHI framework, featuring dynamic slicing, facilitates robust multi-scale defect detection, while a two-stage label fusion process reconstructs fragmented defects and resolves cross-category ambiguities. The method was evaluated on an original dataset encompassing five defect categories, achieving a precision of 94.9%, recall of 86.4%, mAP@0.5 of 95.1%, and mAP@[0.5:0.95] of 71.4%, thereby demonstrating high detection capability. When tested on an independent external dataset, the method maintained robust performance, with a precision of 78.8%, recall of 72.8%, mAP@0.5 of 77.8%, and mAP@[0.5:0.95] of 46.9%, confirming its effectiveness and generalizability in practical applications.</div>","PeriodicalId":655,"journal":{"name":"Journal of Nondestructive Evaluation","volume":"45 3","pages":""},"PeriodicalIF":3.0,"publicationDate":"2026-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148281273","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Modelling Investigation on Bandgap Properties of Fluid-Saturated Porous Lithium-Ion Battery Phononic Crystals 饱和流体多孔锂离子电池声子晶体带隙特性的建模研究

IF 3 3区材料科学

Journal of Nondestructive Evaluation Pub Date : 2026-06-16 DOI: 10.1007/s10921-026-01390-2

Chen Xiong, Huanyu Zhao, Jie Gao, Yan Lyu

{"title":"Modelling Investigation on Bandgap Properties of Fluid-Saturated Porous Lithium-Ion Battery Phononic Crystals","authors":"Chen Xiong, Huanyu Zhao, Jie Gao, Yan Lyu","doi":"10.1007/s10921-026-01390-2","DOIUrl":"10.1007/s10921-026-01390-2","url":null,"abstract":"<div>Lithium-ion batteries typically consist of stacked, wound micron-sized cell structures (cathode/separator/anode), exhibiting periodic structural characteristics akin to phononic crystals. In this study, we present a theoretical model based on Biot’s theory and Floquent-Bloch theorem to investigate wave propagation in a one-dimensional, multi-layered porous lithium-ion battery structure. We employ a numerical approach to compute the bandgap characteristics of this multi-layered porous anode. Our results demonstrate excellent agreement with those obtained from finite element simulations, incorporating a solid mechanics module, thereby validating the feasibility and accuracy of the proposed method. Considering the dynamic variations in electrode porosity and mechanical properties during battery cycling, our findings reveal a fundamental connection between the State of Charge (SOC) and the bandgap characteristics of lithium-ion batteries. A comparative analysis of the effects of structural properties (with and without Biot theoretical assumptions) on the frequency range of transverse and longitudinal bandgaps is also presented. Notably, we observe that the frequency difference of the bandgap shifts systematically during the charging cycle both with and without Biot theoretical assumptions. This mapping relationship provides valuable theoretical insight for the acoustic non-destructive characterization of battery state parameters.</div>","PeriodicalId":655,"journal":{"name":"Journal of Nondestructive Evaluation","volume":"45 3","pages":""},"PeriodicalIF":3.0,"publicationDate":"2026-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148281269","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Advancements in Citrus Quality Assessment: Utilizing X-ray Imaging with CitrusnetQuality Model for Non-Destructive Evaluation 柑橘品质评价的研究进展：利用x射线成像和CitrusnetQuality模型进行无损评价

IF 3 3区材料科学

Journal of Nondestructive Evaluation Pub Date : 2026-06-14 DOI: 10.1007/s10921-025-01316-4

Syed Mudassir Raza, Zia-Ul Haq, Mohamed Ibrahim Abdullah Babeker, Muhammad Waqar Akram, Tahir Iqbal, Muhammad Adnan Islam, Shanjun Li

{"title":"Advancements in Citrus Quality Assessment: Utilizing X-ray Imaging with CitrusnetQuality Model for Non-Destructive Evaluation","authors":"Syed Mudassir Raza, Zia-Ul Haq, Mohamed Ibrahim Abdullah Babeker, Muhammad Waqar Akram, Tahir Iqbal, Muhammad Adnan Islam, Shanjun Li","doi":"10.1007/s10921-025-01316-4","DOIUrl":"10.1007/s10921-025-01316-4","url":null,"abstract":"<div>Citrus is a globally consumed commodity. The quality of its fruit is crucial for marketability and consumer satisfaction. Traditional methods of assessing fruit quality rely on external characteristics, which may not accurately reflect internal attributes such as sweetness, acidity, and moisture content. The current study introduces non-destructive techniques, X-ray computed tomography (CT), and machine learning for automated quality evaluation of citrus (Satsuma). A dataset of citrus samples was developed under ambient conditions with 20–22 °C temper-ature and 50–60% humidity, and the other half citrus samples were placed under refrigerated conditions with 6–8 °C temperature and 65–75% humidity to analyze quality variations. The hypothesis of the”CitrusNetQuality” model integrated Random Forest Regressor, Gradient Boosting Regressor, Partial Least Squares Regression, K-Nearest Neighbors Regressor, and Support Vector Regression, com-bined with Wavelet Transforms and Principal Component Analysis for quality prediction. The model achieved high R2 values (0.97–0.98) and low MAE values (0.01–0.08) across key parameters, outperforming existing models. By integrat-ing X-ray CT and machine learning,”CitrusNetQuality” was enhanced by fruit quality evaluation, reducing assessment challenges and improving consumer sat-isfaction. This study provided a non-destructive tool for precise citrus quality assessment.</div>","PeriodicalId":655,"journal":{"name":"Journal of Nondestructive Evaluation","volume":"45 3","pages":""},"PeriodicalIF":3.0,"publicationDate":"2026-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148281006","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

FAS-YOLO: A Lightweight Network for Steel Defects Detection Based on Deep Learning 基于深度学习的轻量钢缺陷检测网络FAS-YOLO

IF 3 3区材料科学

Journal of Nondestructive Evaluation Pub Date : 2026-06-14 DOI: 10.1007/s10921-026-01373-3

Yuzhe Luo, Hang Liu, Haowei Feng, Hua Tian

{"title":"FAS-YOLO: A Lightweight Network for Steel Defects Detection Based on Deep Learning","authors":"Yuzhe Luo, Hang Liu, Haowei Feng, Hua Tian","doi":"10.1007/s10921-026-01373-3","DOIUrl":"10.1007/s10921-026-01373-3","url":null,"abstract":"<div>Timely defect detection is an important means to ensure product quality in the production process of steel products. Detection methods based on computer vision are currently widely popular due to its fast speed and high accuracy. However, computer defect detection methods face challenges of sufficient training within limited samples and balancing detection accuracy and model size. Hence, we propose a lightweight steel surface defects detection model based on YOLOv8, named FAS-YOLO, “FAS” is the abbreviation for our improvement method. Firstly, we introduce a novel convolution method called Falcon to replace the standard convolution of the head component, which can reduce the parameter count and the computational cost of the model. Secondly, we propose the Adaptive Weight Normalized Wasserstein Distance (AW-NWD) loss function. Building upon the Normalized Wasserstein Distance (NWD) loss function, we design a weight coefficients matrix using the Softmax function to establish the relationship between the shape information of the object and the error dimension. Finally, the attention module SimAM is integrated to enhance the model's feature extraction capability without increasing the computational complexity. Experimental results show that compared to the original YOLOv8, FAS-YOLO reduces the parameter count by 20% and the floating point operations by 31%, while achieving an improvement in mean Aver-age Precision (mAP) of 3.2% and 1.2% in NEU-DET and GC10-DET datasets. Furthermore, the proposed model demonstrates superior detection accuracy and reduced computational complexity compared to other YOLO series models, rendering it more suitable for industrial production.</div>","PeriodicalId":655,"journal":{"name":"Journal of Nondestructive Evaluation","volume":"45 3","pages":""},"PeriodicalIF":3.0,"publicationDate":"2026-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148281007","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Evaluation of Stereovision Tracking of Targets with Gaussian Pixel Intensity Distribution for Measuring 3D Displacements of Structures 结构三维位移测量中高斯像素强度分布目标的立体视觉跟踪评价

IF 2.4 3区材料科学

Journal of Nondestructive Evaluation Pub Date : 2026-06-13 DOI: 10.1007/s10921-026-01378-y

Daniel Alexander Pohoryles, Marco Peroni, Simone Peloso, Francisco Javier Molina

{"title":"Evaluation of Stereovision Tracking of Targets with Gaussian Pixel Intensity Distribution for Measuring 3D Displacements of Structures","authors":"Daniel Alexander Pohoryles, Marco Peroni, Simone Peloso, Francisco Javier Molina","doi":"10.1007/s10921-026-01378-y","DOIUrl":"10.1007/s10921-026-01378-y","url":null,"abstract":"<div>This study evaluates a method for tracking targets using stereovision-based structural monitoring for accurately measuring three-dimensional (3D) displacements of structures. The greyscale targets are printed with a Gaussian pixel intensity distribution, which allows to efficiently identify their centres using an optimisation algorithm. Compared to traditional sensors, such as linear transducers, this non-contact technique provides enhanced versatility, especially in situations where the application of contact-based sensors may be challenging and full-field vision-based measurements, such as Digital Image Correlation (DIC), may be impractical. A series of experimental tests on a calibration bench were conducted to evaluate the displacement measurements using the target tracking algorithm in comparison to traditional contact-based monitoring systems and DIC. The evaluation was carried out according to five criteria: resolution, accuracy, repeatability, reversibility, and standard error. Various parameters, including target size, colour schemes, and camera orientations, were analysed. The results obtained in this study demonstrate that tracking of the Gaussian greyscale targets provides precise measurements with low computational cost, achieving accuracies comparable to both contact-based sensors and DIC. It was shown that a minimum target size of 6 pixels is required and that target sizes between 9 and 36 pixels are preferable, as the impact of colour scheme or camera orientation is least significant. Overall, the tracking algorithm shows promise as a viable alternative for structural and mechanical testing, offering advantages in speed, flexibility, and the ability to measure 3D motion, addressing some limitations of existing techniques.</div>","PeriodicalId":655,"journal":{"name":"Journal of Nondestructive Evaluation","volume":"45 3","pages":""},"PeriodicalIF":2.4,"publicationDate":"2026-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10921-026-01378-y.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148237379","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0