Search

This study reviews advanced methods for corrosion detection and characterization in pipes using thermography, with a focus on addressing the limitations posed by small datasets. Thermography captures temperature distributions on the surface of pipes to identify subsurface defects. The challenges of sequential data processing, neural network performance, feature extraction, and dataset size are discussed, with proposed solutions such as advanced algorithms, feature selection techniques, and data augmentation. Given the significant gap in the current literature, there is a need for larger, more diverse datasets to train more robust and accurate machine learning models. A case study combining experimental data with Finite Element Method (FEM) simulations demonstrates that augmenting datasets with synthetic data significantly improves defect detection accuracy. These findings highlight the potential of integrating thermography with machine learning to enhance defect detection, providing insights for future research and practical applications. [ABSTRACT FROM AUTHOR]

In contrast to conventional non-destructive testing (NDT) and non-destructive evaluation (NDE) methodologies, including radiography, ultrasound, and eddy current analysis, coplanar capacitive sensing technique emerges as a novel and promising avenue within the field. This paper endeavors to elucidate the efficacy of coplanar capacitive sensing, also referred to as capacitive imaging (CI), within the realm of NDT. Leveraging extant scholarly discourse, this review offers a comprehensive and methodical examination of the coplanar capacitive technique, encompassing its fundamental principles, factors influencing sensor efficacy, and diverse applications for defect identification across various NDT domains. Furthermore, this review deliberates on extant challenges and anticipates future trajectories for the technique. The manifold advantages inherent to coplanar capacitive sensing vis-à-vis traditional NDT methodologies not only afford its versatility in application but also underscore its potential for pioneering advancements in forthcoming applications. [ABSTRACT FROM AUTHOR]

A novel nonsubsampled contourlet transform (NSCT) based image fusion approach, implementing an adaptive-Gaussian (AG) fuzzy membership method, compressed sensing (CS) technique, total variation (TV) based gradient descent reconstruction algorithm, is proposed for the fusion computation of infrared and visible images. Compared with wavelet, contourlet, or any other multi-resolution analysis method, NSCT has many evident advantages, such as multi-scale, multi-direction, and translation invariance. As is known, a fuzzy set is characterized by its membership function (MF), while the commonly known Gaussian fuzzy membership degree can be introduced to establish an adaptive control of the fusion processing. The compressed sensing technique can sparsely sample the image information in a certain sampling rate, and the sparse signal can be recovered by solving a convex problem employing gradient descent based iterative algorithm(s). In the proposed fusion process, the pre-enhanced infrared image and the visible image are decomposed into low-frequency subbands and high-frequency subbands, respectively, via the NSCT method as a first step. The low-frequency coefficients are fused using the adaptive regional average energy rule; the highest-frequency coefficients are fused using the maximum absolute selection rule; the other high-frequency coefficients are sparsely sampled, fused using the adaptive-Gaussian regional standard deviation rule, and then recovered by employing the total variation based gradient descent recovery algorithm. Experimental results and human visual perception illustrate the effectiveness and advantages of the proposed fusion approach. The efficiency and robustness are also analyzed and discussed through different evaluation methods, such as the standard deviation, Shannon entropy, root-mean-square error, mutual information and edge-based similarity index. [ABSTRACT FROM AUTHOR]

Pulsed thermography is a technique of significant interest in non-destructive testing, particularly in defect detection and depth characterisation of composite materials. This study presents an innovative methodology for simultaneously detecting defects and estimating depth using a combination of sequenced thermal signal encoding and a two-dimensional convolution neural network (CNN) model. We compare the results of the proposed method with those obtained from the feed-forward neural network (FFNN), a one-dimensional CNN, and a long short-term memory recurrent neural network (LSTM-RNN). The findings demonstrate that the proposed approach exhibits superior accuracy and robustness compared to the benchmarks. [ABSTRACT FROM AUTHOR]

• Demonstrate the feasibility of optical pulsed thermography for crack detection in Tang murals. • Propose a novel attention U-Net that combines an attention and a U-Net for enhanced crack feature extraction. • Evaluate attention U-Net's effectiveness and generalizability compared to wavelet edge detection and traditional U-Net. Ancient mural degradation and destruction may result from various natural causes, resulting in cracks, peeling, or bulging. As such, regular testing and evaluation of ancient murals are indispensable for protecting and preserving cultural relics. In many scenarios, the acquisition of detection data can be expedited through the use of mechanical arms and imaging equipment. However, the subsequent data analysis relies on experienced human inspectors, resulting in a laborious and time-consuming process. This study focuses on automated analysis of cracks in ancient murals using optical pulsed thermography. A technique that combines an attention mechanism and the U-Net neural network is proposed for refined crack feature extraction. Concerning the identification of ancient mural cracks based on limited training data, U-Net with the attention mechanism demonstrates superior performance over both the conventional U-Net and a traditional image segmentation algorithm. [Display omitted] [ABSTRACT FROM AUTHOR]

We address the challenge of determining a valid set of parameters for a dynamic line scan thermography setup. Traditionally, this optimization process is labor- and time-intensive work, even for an expert skilled in the art. Nowadays, simulations in software can reduce some of that burden. However, when faced with many parameters to optimize, all of which cover a large range of values, this is still a time-consuming endeavor. A large number of simulations are needed to adequately capture the underlying physical reality. We propose to emulate the simulator by means of a Gaussian process. This statistical model serves as a surrogate for the simulations. To some extent, this can be thought of as a "model of the model". Once trained on a relative low amount of data points, this surrogate model can be queried to answer various engineering design questions. Moreover, the underlying model, a Gaussian process, is stochastic in nature. This allows for uncertainty quantification in the outcomes of the queried model, which plays an important role in decision making or risk assessment. We provide several real-world examples that demonstrate the usefulness of this method. [ABSTRACT FROM AUTHOR]

Combinative methodologies have the potential to address the drawbacks of unimodal non-destructive testing and evaluation (NDT & E) when inspecting multilayer structures. The aim of this study is to investigate the integration of information gathered via phased-array ultrasonic testing (PAUT) and pulsed thermography (PT), addressing the challenges posed by surface-level anomalies in PAUT and the limited deep penetration in PT. A center-of-mass-based registration method was proposed to align shapeless inspection results in consecutive insertions. Subsequently, the aligned inspection images were merged using complementary techniques, including maximum, weighted-averaging, depth-driven combination (DDC), and wavelet decomposition. The results indicated that although individual inspections may have lower mean absolute error (MAE) ratings than fused images, the use of complementary fusion improved defect identification in the total number of detections across numerous layers of the structure. Detection errors are analyzed, and a tendency to overestimate defect sizes is revealed with individual inspection methods. This study concludes that complementary fusion provides a more comprehensive understanding of overall defect detection throughout the thickness, highlighting the importance of leveraging multiple modalities for improved inspection outcomes in structural analysis. [ABSTRACT FROM AUTHOR]

Thermography has been used extensively as a complementary diagnostic tool in breast cancer detection. Among thermographic methods, matrix factorization (MF) techniques show an unequivocal capability to detect thermal patterns corresponding to vasodilation in the cancer cases. One of the biggest challenges in such techniques is selecting the best representation of the thermal basis. In this study, an embedding method is proposed to address this problem and deep-semi-non-negative MF (Deep-SemiNMF) for thermography is introduced, then tested for 208 breast cancer screening cases. First, we apply Deep-SemiNMF to infrared images to extract low-rank thermal representations for each case. Then, we embed low-rank bases to obtain one basis for each patient. After that, we extract 300 thermal imaging features, called thermomics, to decode imaging information for the automatic diagnostic model. We reduced the dimensionality of thermomics by spanning them on to Hilbert space using radial-based function (RBF) kernel and select the three most efficient features using the block Hilbert–Schmidt Independence Criterion (HSIC) Lasso. The preserved thermal heterogeneity successfully classified asymptomatic versus symptomatic patients applying a random forest model [cross-validated accuracy of 71.36% (69.42%–73.3%)]. [ABSTRACT FROM AUTHOR]

ActiveU and passive thermographies are two efficient techniques extensively used to measure heterogeneous thermal patterns, leading to subsurface defects for diagnostic evaluations. This study conducts a comparative analysis on low-rank matrix approximation methods in thermography with applications of semi-, convex-, and sparse-nonnegative matrix factorization (NMF) methods for detecting subsurface thermal patterns. These methods inherit the advantages of principal component thermography (PCT) and sparse PCT and tackle negative bases in sparse PCT with nonnegative constraints and exhibit clustering property in processing data. The practicality and efficiency of these methods are demonstrated by the experimental results for subsurface defect detection in three specimens and preserving thermal heterogeneity for distinguishing breast abnormality in breast cancer screening data set (accuracy of 74.1%, 75.9%, and 77.8%). [ABSTRACT FROM AUTHOR]

Efficient multi-modal image fusion plays an important role in the non-destructive evaluation (NDE) of infrastructures, where an essential challenge is the precise visualizing of defects. While automatic defect detection represents a significant advancement, the determination of the precise location of both surface and subsurface defects simultaneously is crucial. Hence, visible and infrared data fusion strategies are essential for acquiring comprehensive and complementary information to detect defects across vast structures. This paper proposes an infrared and visible image registration method based on Euclidean evaluation together with a trade-off between key-point threshold and non-maximum suppression. Moreover, we employ a multi-modal fusion strategy to investigate the robustness of our image registration results. [ABSTRACT FROM AUTHOR]

Efficient image stitching plays a vital role in the Non-Destructive Evaluation (NDE) of infrastructures. An essential challenge in the NDE of infrastructures is precisely visualizing defects within large structures. The existing literature predominantly relies on high-resolution close-distance images to detect surface or subsurface defects. While the automatic detection of all defect types represents a significant advancement, understanding the location and continuity of defects is imperative. It is worth noting that some defects may be too small to capture from a considerable distance. Consequently, multiple image sequences are captured and processed using image stitching techniques. Additionally, visible and infrared data fusion strategies prove essential for acquiring comprehensive information to detect defects across vast structures. Hence, there is a need for an effective image stitching method appropriate for infrared and visible images of structures and industrial assets, facilitating enhanced visualization and automated inspection for structural maintenance. This paper proposes an advanced image stitching method appropriate for dual-sensor inspections. The proposed image stitching technique employs self-supervised feature detection to enhance the quality and quantity of feature detection. Subsequently, a graph neural network is employed for robust feature matching. Ultimately, the proposed method results in image stitching that effectively eliminates perspective distortion in both infrared and visible images, a prerequisite for subsequent multi-modal fusion strategies. Our results substantially enhance the visualization capabilities for infrastructure inspection. Comparative analysis with popular state-of-the-art methods confirms the effectiveness of the proposed approach. [ABSTRACT FROM AUTHOR]

Industrial radiography is a pivotal non-destructive testing (NDT) method that ensures quality and safety in a wide range of industrial sectors. Conventional human-based approaches, however, are prone to challenges in defect detection accuracy and efficiency, primarily due to the high inspection demand from manufacturing industries with high production throughput. To solve this challenge, numerous computer-based alternatives have been developed, including Automated Defect Recognition (ADR) using deep learning algorithms. At the core of training, these algorithms demand large volumes of data that should be representative of real-world cases. However, the availability of digital X-ray radiography data for open research is limited by non-disclosure contractual terms in the industry. This study presents a pipeline that is capable of modeling synthetic images based on statistical information acquired from X-ray intensity distribution from real digital X-ray radiography images. Through meticulous analysis of the intensity distribution in digital X-ray images, the unique statistical patterns associated with the exposure conditions used during image acquisition, type of component, thickness variations, beam divergence, anode heel effect, etc., are extracted. The realized synthetic images were utilized to train deep learning models, yielding an impressive model performance with a mean intersection over union (IoU) of 0.93 and a mean dice coefficient of 0.96 on real unseen digital X-ray radiography images. This methodology is scalable and adaptable, making it suitable for diverse industrial applications. [ABSTRACT FROM AUTHOR]

Abstract In order to quantitatively compare the reliability of pulsed thermography and ultrasonic testing techniques, a set of thirty-five Carbon Fiber Reinforced Plastic (CFRP) composite panels with impact damages are inspected by pulsed thermography and ultrasonic C-scan. The comparative experimental results and Probability of Detection (PoD) analysis results are presented. The quantitative comparison shows that pulsed thermography testing has smaller defect size at 90% PoD with 95% confidence level, i.e. a 90/95 values than ultrasonic testing for the parameters and setup used in the inspections of these thirty-five CFRP composite panels. [ABSTRACT FROM AUTHOR]

In recent years, the conservation and protection of ancient cultural heritage have received increasing attention, and non-destructive testing (NDT), which can minimize the damage done to the test subject, plays an integral role therein. For instance, NDT through active infrared thermal imaging can be applied to ancient polyptychs, which can realize accurate detection of damage and defects existing on the surface and interior of the polyptychs. In this study, infrared thermography is used for non-invasive investigation and evaluation of two polyptych samples with different pigments and artificial defects, but both reproduced based on a painting by Pietro Lorenzetti (1280/85–1348) using the typical tempera technique of the century. It is noted that, to avoid as far as possible secondary damages done to the ancient cultural heritages, repeated damage-detection experiments are rarely carried out on the test subjects. To that end, numerical simulation is used to reveal the heat transfer properties and temperature distributions, as to perform procedural verification and reduce the number of experiments that need to be conducted on actual samples. Technique-wise, to improve the observability of the experimental results, a total variation regularized low-rank tensor decomposition algorithm is implemented to reduce the background noise and improve the contrast of the images. Furthermore, the efficacy of image processing is quantified through the structural-similarity evaluation. [ABSTRACT FROM AUTHOR]

• An improved YOLOX algorithm, by adding a convolutional block attention module (CBAM) mechanism to the network, is employed to improve the status quo. • A part of the famous "The Birth of Venus" by Botticelli (∼1485 ca.) was reproduced by a professional restorer, and three different defects were artificially introduced. • The simulated data were used for training, whereas the replica was used for testing. • The performance of the improved network herein proposed outperforms the benchmarks, with an average accuracy of 96.67%. Early defects in artworks, which are unnoticeable by the naked eye, can evolve in severity if left unattended; therefore, their timely and accurate detection is of vital importance. Infrared-based non-destructive testing (NDT) methods are widely seen as an effective means to detect defects in artworks. However, most existing methods still rely on the human-based judgment of defects after processing the thermal images, which would inevitably lead to opinion bias and inconsistency. Deep-learning algorithms for computer vision are thought adequate to address the situation. On this point, an improved YOLOX algorithm, with the convolutional block attention module (CBAM) mechanism added to the network, is employed to improve the status quo. The method works with thermal imaging data under flash lamps for tempera paintings on canvas. A part of the famous "The Birth of Venus" by Botticelli (∼1485 ca.) was reproduced by a professional restorer, and three different defects were artificially introduced. To overcome the challenge pertaining to having too few defect samples, the dataset is expanded using finite-element simulation methods. The simulated data are used for training, whereas the replica is used for testing. Experimental results show that adding the CBAM mechanism after upsampling and downsampling of the path aggregation network can maximize network performance. The performance of the improved network herein proposed outperforms the benchmarks, with an average accuracy of 96.67%, a detection speed of 20.0 fps, and a total of 5.08 million parameters. [ABSTRACT FROM AUTHOR]

With increasing attention paid to the protection of cultural relics, non-destructive testing (NDT) technologies are thought to be profoundly rewarding, resulting in a widespread uptake of feature-extraction algorithms and defect-detection techniques. Among various alternatives, infrared thermography (IRT) and Terahertz time-domain spectroscopy (THz-TDS) are non-invasive in nature and thus are appropriate for applications involving ancient buildings and artworks. The present study is motivated by the fact that online/offline background segmentation algorithms based on the Gaussian mixture model and widely used in video processing to distinguish between foreground and background, can be successfully integrated as a feature extraction tool with NDT. Since IRT and THz-TDS image sequences resemble a video, the image length and width can be taken as the first two dimensions, and time can serve as the third one. Such sequences can be processed effectively by the background segmentation algorithms, thus can help detect defects of different types at varying depths. The experimental section of the paper considers a tempera painting (a replica of Botticelli's "The Birth of Venus") with artificiallyintroduced defects. For benchmarking purposes, the background segmentation algorithms (based on a mixture of Gaussian models) are compared with the fast Fourier transform and principal component analysis to demonstrate the superior performance of the proposed novel algorithms. [ABSTRACT FROM AUTHOR]

Infrared thermography (IRT), is one of the most interesting techniques to identify different kinds of defects, such as delamination and damage existing for quality management of material. Objective detection and segmentation algorithms in deep learning have been widely applied in image processing, although very rarely in the IRT field. In this paper, spatial deep-learning image processing methods for defect detection and identification were discussed and investigated. The aim in this work is to integrate such deep-learning (DL) models to enable interpretations of thermal images automatically for quality management (QM). That requires achieving a high enough accuracy for each deep-learning method so that they can be used to assist human inspectors based on the training. There are several alternatives of deep Convolutional Neural Networks for detecting the images that were employed in this work. These included: 1. The instance segmentation methods Mask–RCNN (Mask Region-based Convolutional Neural Networks) and Center–Mask; 2. The independent semantic segmentation methods: U-net and Resnet–U-net; 3. The objective localization methods: You Only Look Once (YOLO-v3) and Faster Region-based Convolutional Neural Networks (Fast-er-RCNN). In addition, a regular infrared image segmentation processing combination method (Absolute thermal contrast (ATC) and global threshold) was introduced for comparison. A series of academic samples composed of different materials and containing artificial defects of different shapes and nature (flat-bottom holes, Teflon inserts) were evaluated, and all results were studied to evaluate the efficacy and performance of the proposed algorithms. [ABSTRACT FROM AUTHOR]

In response to the growing inspection demand exerted by process automation in component manufacturing, non-destructive testing (NDT) continues to explore automated approaches that utilize deep-learning algorithms for defect identification, including within digital X-ray radiography images. This necessitates a thorough understanding of the implication of image quality parameters on the performance of these deep-learning models. This study investigated the influence of two image-quality parameters, namely signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR), on the performance of a U-net deep-learning semantic segmentation model. Input images were acquired with varying combinations of exposure factors, such as kilovoltage, milli-ampere, and exposure time, which altered the resultant radiographic image quality. The data were sorted into five different datasets according to their measured SNR and CNR values. The deep-learning model was trained five distinct times, utilizing a unique dataset for each training session. Training the model with high CNR values yielded an intersection-over-union (IoU) metric of 0.9594 on test data of the same category but dropped to 0.5875 when tested on lower CNR test data. The result of this study emphasizes the importance of achieving a balance in training dataset according to the investigated quality parameters in order to enhance the performance of deep-learning segmentation models for NDT digital X-ray radiography applications. [ABSTRACT FROM AUTHOR]

Pulsed thermography is a nondestructive method commonly used to explore anomalies in composite materials. This paper presents a procedure for the automated detection of defects in thermal images of composite materials obtained with pulsed thermography experiments. The proposed methodology is simple and novel as it is reliable in low-contrast and nonuniform heating conditions and does not require data preprocessing. Nonuniform heating correction and the gradient direction information combined with a local and global segmentation phase are used to analyze carbon fiber-reinforced plastic (CFRP) thermal images with Teflon inserts with different length/depth ratios. Additionally, a comparison between the actual depths and estimated depths of detected defects is performed. The performance of the nonuniform heating correction proposed method is superior to that obtained on the same CFRP sample analyzed with a deep learning algorithm and the background thermal compensation by filtering strategy. [ABSTRACT FROM AUTHOR]

This paper presents a framework that integrates passive infrared thermography (IRT) results with building information modeling (BIM) to improve subsurface delamination inspection in concrete infrastructures. The paper combines solar analysis with BIM for better thermography inspection planning and documents thermographic data on delamination within BIM environment using a semi-automatic AI procedure for delamination identification. The framework was tested on two case studies: one laboratory sample with inserted delaminations and one real-scale concrete structure. As a result, besides the increase in communication related to data visualization and centralization, the volume of the delivered data and the analysis time was reduced through the application of the proposed method. The findings promote better collaboration between the civil engineering and inspection industries, utilizing advanced nondestructive techniques for detecting concrete delamination. This approach improves planning, analysis, visualization, reporting, and data storage for passive IRT inspections, benefiting infrastructure stakeholders and streamlining maintenance management. • Framework for integrating passive IRT and BIM to monitor delamination in concrete. • BIM-based solar load analysis for passive IRT inspection planning. • Proposed framework was demonstrated through laboratory and real-scale case studies. • The framework outperforms existing tools in data communication and analysis time. [ABSTRACT FROM AUTHOR]

Fiber orientation in composite materials is an important feature since the arrangement or orientation of the fibers relative to one another has a significant influence on the strength and other properties of fiber reinforced composites. In this paper we present a method to assess the fiber orientation on the surface of carbon fiber reinforced polymer (CFRP) laminates. More specifically, a diode-laser beam is used to locally heat a small spot on the surface of the sample. Observation of the heat pattern in the infrared spectrum enables the assessment of the fiber orientation. Different samples and different regions on the surface of the samples are tested in order to estimate the precision of the method. [ABSTRACT FROM AUTHOR]

The accuracy of identifying defects using specialized deep learning models can be affected by the circumstances in which the training data is curated. This is especially evident in digital X-ray radiography, where the depiction of flaws is significantly impacted by the exposure conditions. This study examines the effect of curating high-quality data on deep learning models. The variation in contrast-to-noise ratio (CNR), which is a crucial metric of image quality between features of interest and an adjacent normal background, has been found to be a key factor in model generalization in digital X-ray radiography applications. By making systematic alterations to exposure conditions during data curation, it was possible to obtain several representations of flaws in each test component with varying contrast-to-noise ratios (CNR) in the resultant radiographs. To evaluate the efficacy of the model under various conditions, two distinct datasets were curated. Dataset 1 was obtained by acquiring images with a consistent exposure setting on 140 test samples. The samples contained 4 morphologically distinct classes of flat bottom holes with seven different depths and sizes. The contrast-to-noise ratio (CNR) representations of flaws in this dataset can be attributed only to differences in depth in Dataset 1. Additionally, Dataset 2 was curated with an expanded range of CNR values by methodically adjusting exposure settings during image acquisitions. Hence, only 42 % of the test pieces from Dataset 1, which had three distinct depths of flat bottom holes, were used. Each of the two datasets was used to separately train YOLOv8 for instance segmentation and U-net for multi-class semantic segmentation. Each model was trained under the same conditions, and their performances were assessed using test sets from both dataset groups. The model trained on Dataset 1 exhibited a notable decline in performance when evaluated on test sets from Dataset 2, suggesting a lack of generalization ability. Conversely, the model that was trained using Dataset 2 consistently achieved high accuracy on both test sets, demonstrating impressive performance and successful generalization. This work shows that the generalization abilities of deep learning models may be improved by varying the contrast-to-noise ratio (CNR) of features in the training data. This finding paves the way for practical applications in digital X-ray radiography. [ABSTRACT FROM AUTHOR]

This paper introduces a computational approach for inferring the minimum requirements for the nondestructive inspection of subsurface delamination in outdoor concrete structures using passive infrared thermography (IRT). The non-linear numerical system was solved using the Finite Element Method (FEM). Complete verification and validation of the numerical model were performed through the analysis of experimental and computational errors, as well as through the comparison of computational outputs of thermal gradients with the contrast values measured in an experiment with solar radiation and passive IRT. The results of accuracy and precision of the computational simulation approach were found to be adequate, from a practical perspective, for the intended use of the model, with the thermal gradient values having an uncertainty of 0.080 ± 0.91 °C and -0.016 ± 0.91 °C for the concrete slab and column sample, respectively. Furthermore, the developed model was used to perform a one-year analysis of the studied case, in order to determine the approximate radiative heat flux required to identify defects with different size-to-depth (S/D) ratios in various concrete components with distinct solar exposures. Finally, the relationship between the calculated radiative heat flux and thermal contrast with the respective environmental variables in place was analyzed graphically. • Multiphysics modeling of passive infrared thermography (IRT) for concrete inspection. • Comprehensive verification and validation of the computational model. • Inference on the minimum radiative heat flux required for detecting delamination. • Required heat flux varied with defect size-to-depth (S/D) ratio and solar orientation. • Limited detection for defects with unfavorable solar orientation and small S/D ratio. [ABSTRACT FROM AUTHOR]

• A new Faster R-CNN-CA network is proposed using coordinate attention. • Faster R-CNN-CA is compared with four other attention mechanism modules. • R-CNN-CA has the highest mean average precision among them. • 3D imaging using advanced processing stacking is presented with higher SNR than the previous work. The demand for non-invasive inspection (NII) is ever-increasing in the field of cultural heritage conservation. NII is a two-step procedure, first of data acquisition and second of defect detection. Stand-alone imaging techniques such as infrared thermography (IRT) are often insufficient for performing a complete remote analysis and diagnosis of historic structures and art pieces that are of very high cultural value. On this point, an emerging optical inspection method, terahertz time-domain spectroscopy (THz-TDS), is herein employed to provide more details of deeper defects. The imaging results from THz-TDS and IRT are compared and analyzed by employing advanced image processing methods. Next, to achieve automatic inspection of the test sample, which is an ancient marquetry, a Faster R-CNN with coordinate attention (Faster R-CNN-CA) is proposed and fitted with data from two different sources. Worth noting is that, in order to populate sufficient data for training, samples are simulated using finite element analysis and finite difference time domain method. The experiments demonstrate that the mean average precision of the Faster R-CNN-CA model improves by 6.09% over the traditional Faster R-CNN model. [ABSTRACT FROM AUTHOR]

Recent advancements in dimensionality reduction techniques have significantly contributed to the field of active infrared thermography (AIRT) for defect detection, aiding in data processing and feature extraction. Among these techniques, principal component thermography (PCT) and deep autoencoder thermography (DAT) are particularly notable. PCT is based on conventional linear multivariate analysis, while DAT leverages deep learning paradigms to better handle nonlinearity. These methods consolidate defect information from multiple thermograms into a concise set of feature images, enhancing the visibility of subsurface material defects. However, these feature images often suffer from disturbances, particularly non-uniform backgrounds caused by uneven heating in AIRT experiments. Such interferences can obscure defect information, necessitating further post-processing. In our research, we explore the efficacy of Adaptive Iteratively Reweighted Penalized Least Squares (AIR-PELS) as a refinement technique for PCT and DAT, focusing on background suppression. The adaptive iterative weighting with PELS smoothing effectively reduces noise and removes background disturbances. Case studies involving carbon fiber-reinforced polymer samples with inherent defects demonstrate the effectiveness of this post-processing approach. [ABSTRACT FROM AUTHOR]

This paper proposes an approach for the reliable identification of subsurface damages in thermal images of concrete structures. The work explores how to mitigate false positives in subsurface delamination segmentation using thermal and visible images. The methodology employs a few-shot learning method, specifically the Siamese Neural Network (SNN), to assess the similarity between corresponding multimodal regions. The findings indicate that leveraging similarities between visible and thermal images reduces false positives and improves the segmentation model's precision by 3.6%, eliminating 351 false positives. These results enhance the reliability of semi-automatic models for detecting subsurface delamination using infrared thermography, benefiting infrastructure maintenance and encouraging the research and development of compact and reliable automation models that integrate civil engineering, nondestructive testing, and artificial intelligence domains. • Few-shot learning approach to mitigate false positives in delamination segmentation. • Evaluation with 500 pairs of infrared and visible images from concrete infrastructure. • Multimodal SNN integration and filtering enhance the precision of segmentation model. [ABSTRACT FROM AUTHOR]

In the present study, a relatively novel non-destructive testing (NDT) method called the coplanar capacitive sensing technique was applied in order to detect different sizes of rebars in a reinforced concrete (RC) structure. This technique effectively detects changes in the dielectric properties during scanning in various sections of concrete with and without rebars. Numerical simulations were carried out by three-dimensional (3D) finite element modelling (FEM) in COMSOL Multiphysics software to analyse the impact of the presence of rebars on the electric field generated by the coplanar capacitive probe. In addition, the effect of the presence of a surface defect on the rebar embedded in the concrete slab was demonstrated by the same software for the first time. Experiments were performed on a concrete slab containing rebars, and were compared with FEM results. The results showed that there is a good qualitative agreement between the numerical simulations and experimental results. [ABSTRACT FROM AUTHOR]

Thermography is a nondestructive testing (NDT) technique based on the principle that two dissimilar materials, i.e., possessing different thermo-physical properties, would produce two distinctive thermal signatures that can be revealed by an infrared sensor, such as a thermal camera. The fields of NDT applications are expanding from classical building or electronic components monitoring to more recent ones such as inspection of artworks or composite materials. Furthermore, thermography can be conveniently used as a didactic tool for physics education in universities given that it provides the possibility of visualizing fundamental principles, such as thermal physics and mechanics among others. [ABSTRACT FROM AUTHOR]

Abstract: Infrared thermography (IRT) body temperature readings were taken in the ocular region of 258 pigs immediately before slaughter. Levels of lactate were measured in blood taken in the restrainer. Meat quality was assessed in the longissimus dorsi (LD), semimembranosus (SM), and adductor muscles. Ocular IRT (IROT) temperature was correlated with blood lactate levels (r =0.20; P =0.001), with pH taken 1hour postmortem (pH1: r =−0.18; P =0.03) and drip loss (r =0.20; P =0.02) in the LD muscle, and with pH1 in the SM muscle (r =−0.20; P =0.02). Potentially, IROT may be a useful tool to assess the physiological conditions of pigs at slaughter and predict the variation of important meat quality traits. However, the magnitude of the correlations is rather low, so a further development of image capture technique and further studies under more variable preslaughter conditions ensuring a larger pork quality variation are needed. [Copyright &y& Elsevier]

The main objective of this work is to extract the hand vein network using a non-invasive technique in the near-infrared region (NIR). The visualization of the veins is based on a relevant feature of the blood in relation with certain wavelengths of the electromagnetic spectrum. In the present paper, we first introduce the image formation in the NIR spectral band. Then, the acquisition system will be presented as well as the method used for the image processing in order to extract the vein signature. Extractions of this pattern on the finger, on the wrist and on the dorsal hand are achieved after exposing the hand to an optical stimulation by reflection or transmission of light. We present meaningful results of the extracted vein pattern demonstrating the utility of the method for a clinical application like the diagnosis of vein disease, of primitive varicose vein and also for applications in vein biometrics. [ABSTRACT FROM AUTHOR]

Abstract: This paper analyzes two mechanisms applied on the human body in order to study the thermoregulatory system according to heat generation and heat loss. Two approaches are presented. The first approach is based on plethysmography, where an armband is placed on the forearm in order to modulate the blood flow. The second approach uses a cold stimulation. The visualization is achieved using infrared imaging devices. The resulting images reveal a temperature balance between the stimulated and the non-stimulated hands. The thermal behavior and typical thermographic recordings on each subject are discussed and analyzed in response to different stimulations. [Copyright &y& Elsevier]

While other nondestructive testing methods hardly reveal microscopic open cracks, vibrothermography stimulated by ultrasound provides very promising results by converting mechanical waves into local heat by friction. This phenomenon enhances thermal gradients in temperature maps as compared to conventional techniques. To detect temperature gradients caused by hidden cracks, high thermal and spatial resolution infrared cameras are usually used. The aim of this work is to investigate the ability of the vibrothermography stimulated by ultrasound to detect such cracks and measure the gap between adjacent cracks. To do so, we investigated tungsten carbide coatings where cracks were artificially generated using a controlled bending test. Several samples were investigated during this study but only typical results from one sample are presented and discussed. [ABSTRACT FROM AUTHOR]

Simple Summary: The findings of predictive and diagnostic systems in cancer are an intriguing topic for physicians and the oncologic community. Computer-aided decision (CAD) is vital for breast cancer diagnosis. It aids in higher accuracy and early, reliable diagnosis. To achieve such aims, diverse imaging modalities have been used and decision-making was facilitated by artificial intelligence and machine learning models. High-fidelity automated breast lesion finding, along with their corresponding radiomic feature biomarkers, can be delivered by a trained model. In this study, the potential impact of a machine learning model for detecting breast lesions and various radiomic biomarkers are examined. This study presents a model that automatically segments and extracts radiomics and can enable the clinical practice to find breast lesions while performing diagnosis concurrently. Automated medical data analysis demonstrated a significant role in modern medicine, and cancer diagnosis/prognosis to achieve highly reliable and generalizable systems. In this study, an automated breast cancer screening method in ultrasound imaging is proposed. A convolutional deep autoencoder model is presented for simultaneous segmentation and radiomic extraction. The model segments the breast lesions while concurrently extracting radiomic features. With our deep model, we perform breast lesion segmentation, which is linked to low-dimensional deep-radiomic extraction (four features). Similarly, we used high dimensional conventional imaging throughputs and applied spectral embedding techniques to reduce its size from 354 to 12 radiomics. A total of 780 ultrasound images—437 benign, 210, malignant, and 133 normal—were used to train and validate the models in this study. To diagnose malignant lesions, we have performed training, hyperparameter tuning, cross-validation, and testing with a random forest model. This resulted in a binary classification accuracy of 78.5% (65.1–84.1%) for the maximal (full multivariate) cross-validated model for a combination of radiomic groups. [ABSTRACT FROM AUTHOR]

The non-uniformity of non-planar object inspection data makes their analysis challenging. This paper reports a study of the use of recurrent neural network and artificial feed-forward neural network in pulsed thermography during the automated inspection of non-planar carbon fiber reinforced plastic samples. The time series, including the raw temperature–time series and sequenced signals obtained from the first derivative after thermographic signal reconstruction was used to train and test the models respectively. Quantitative comparisons of testing results showed that the long short-term memory recurrent neural network model was more accurate in handling time dependent information compared to the artificial feed-forward neural network model. [ABSTRACT FROM AUTHOR]

With the recent developments in artificial intelligence, deep learning (convolutional neural networks) has been investigated for the quality management of materials by infrared thermography (IRT). In this work, we present an approach to efficiently process thermal data in order to improve the defect detection performance of IRT. A defect-enhanced algorithm based on Generative Adversarial Networks (GANs) has been designed and implemented qualitatively analysis and improve defects visibility based on data augmentation from a deep learning approach. We implemented a data dimension reduction method based on Partial Least Square Thermography (PLST) merged with GAN Networks (PLS-GANs) to achieve interpretable feature extraction and visualization, and compare result with thermal data of Pulsed Thermography in order to evaluate the efficacy of the proposed algorithm. By applying PLS-GANs, a small dataset of thermal data can be able to enlarge the diversity of data in order to improve the performance of the detection model. The experimental results were empirically illustrated over the benchmark specimens: Carbon Fiber Reinforced Polymers (CFRPs). Consequently, the experimental detection results on the CFRPs demonstrated its feasibility of the PLST-GANs method. [ABSTRACT FROM AUTHOR]

• Utilize finite element analysis to investigate the infrared and terahertz non-destructive inspection principle for artworks. • Propose a new terahertz feature extraction algorithm, which converts time-step data to 2D images and enhances the noise robustness. • Implement an unsupervised fusion algorithm, which integrates the features obtained from both infrared and terahertz imaging. This study utilizes infrared thermography (IRT) and terahertz (THz) imaging techniques to conduct a non-invasive inspection of marqueteries. Specifically, two experimental investigations and image processing methods have been carried out, and their advantages and disadvantages have been highlighted. To cross-reference with the experimental results, a finite element simulation was performed for both IRT and THz, which validate the experimental outcomes and providing instructions to the experimental set-up and parameters. Then, an ad-hoc fusion imaging method has been applied to the experimental results to identify defects and features. Such a reverse engineering procedure integrating numerical and experimental images acquired at different wavelengths may be considered an interesting step forward for the optimization of cultural heritage nondestructive testing (NDT). [ABSTRACT FROM AUTHOR]

• Three distinct non-invasive inspection methods are used, including infrared thermography, terahertz time-domain spectroscopy, and air-coupled ultrasound. • YOLOv7, as an automatic defect detection deep learning network, is applied. • To obtain a sufficient dataset to train the network, the finite element analysis and finite-difference time-domain method are introduced. Due to fungal growth and mishandling in the book, there are various types of defects as they age such as foxing, tears, and creases. It is important to develop novel non-invasive inspection techniques and defect recognition algorithms. In this work, three non-invasive inspection techniques, including infrared thermography (IRT), terahertz time-domain spectroscopy (THz-TDS), and air-coupled ultrasound (ACU), were employed for the detection of defects in an ancient book cover. To improve the image quality and defect contrast, principal component analysis, fast Fourier transform, and partial least squares regression algorithms are used as the post-processing methods. Furthermore, the YOLOv7 network is deployed for defect automatic detection. Finite element analysis and finite-difference time-domain methods were employed for generating training dataset of YOLOv7 network. Experimental results demonstrate that IRT and THz-TDS has excellent detection capability for surface and subsurface defects, respectively. By employing YOLOv7 network with simulation datasets, defects can be effectively identified. [ABSTRACT FROM AUTHOR]

Liquid moulded polymer matrix composites (LM-PMCs) are increasingly used in aerospace, automotive and other industrial applications. Liquid moulding processes featuring dry fibre preforms provide flexibility and enable cost reductions for secondary load-bearing structures. However, preform variability and manufacturing reproducibility remain major obstacles to wider use in primary structures. The open literature records only marginal use of non-destructive inspection (NDI) methods for dry multilayer preforms due to technological limitations and cost of NDI methods. In this work, terahertz time-domain spectroscopy (THz-TDS) is used for inspecting three dry multilayer glass fibre preforms featuring different defects, for the first time. A novel time-domain enhancement method is compared with classical image processing methods, aiming at improving image contrast and detecting potential defects. Furthermore, THz B-Scan is used for verifying the accuracy of interply defect detection. Finite difference time domain is simulated for analyzing THz magnitude variation in time-domain. Finally, quantitative evaluation is applied to further illustrate the significant potential of THz-TDS for the inspection of dry fibre preforms. The results show that the errors on defects lengths, widths, angles, and diameters for THz-TDS are in the ranges 8.2 %–34.3 %, 18.67 %–75 %, 0.29 %–6.67 %, and 1.33 %–10 % respectively. [Display omitted] • Terahertz technology is used for multilayer dry glass fibre preforms inspection for the first time. • A novel time-domain image enhancement method is proposed. • A new method based on disturbances is used to verify the interply detection accuracy. • Quantitative evaluation is performed and the potential of industrial use is discussed. [ABSTRACT FROM AUTHOR]

• This study investigates the reliability of a machine learning system to identify eleven different mineral grains. • A Long Wave Infrared (LWIR, 7.7–11.8 μm) imaging system with a macro lens used to improve spatial resolution. • Micro X-ray Fluorescence (µXRF) represented the mineral aggregates while Scanning Electron Microscope (SEM) showed the grains' surface. • The system accuracy was determined using two sets of ground truth (i.e. rigid-GT, and observed-GT). • The results of µXRF and the automated mineral identification system were compared using ArcGIS. Application of hyperspectral infrared imagery for mineral grain identification suffers from a lack of prediction on the irregular grain's surface along with the mineral aggregates. Here, we present an investigation to determine the reliability of automatic mineral identification in the longwave Infrared (LWIR, 7.7–11.8 μ m) with an LWIR-macro lens having a spatial resolution of 100 μ m. We attempt to identify eleven different mineral grains (biotite, epidote, goethite, diopside, smithsonite, tourmaline, kyanite, scheelite, pyrope, olivine, and quartz). A machine learning-based algorithm (implemented by software) compares all of the pixel-spectra to the ASTER spectral library of JPL/NASA using spectral angle mapper (SAM) and normalized cross-correlation (NCC) to create false-color maps. Then a hue-saturation-value (HSV) principle component analysis (PCA) based K-means clustering approach groups the mineral regions in different categories. The results were compared to two different ground truths (GT) (i.e. rigid-GT and observed-GT) for quantitative calculation and as an integrated step for validating our approach. Observed-GT increased the accuracy up to 1.5 times higher than rigid-GT, from 45.67 % to 69.39 %. The samples were also examined by micro X-ray fluorescence (μ XRF) and scanning electron microscope (SEM) in order to retrieve information on the mineral aggregates and the grain's surface. The results of μ XRF imagery (aggregate map) were compared to the results of automatic mineral identification techniques, using ArcGIS software, and the results represent a promising performance for automatic identification. [ABSTRACT FROM AUTHOR]

Infrared thermal imager is a powerful instrument for temperature measurement and nondestructive testing, but the measurement accuracy of infrared thermal imagers is easily affected by other factors. Aiming at the influence of dust in the optical path between the measured object and the infrared thermal imager, this article proposes a compensation model based on the infrared temperature measurement principle. Then, dust transmittance and reflectance, two important parameters but difficult to determine in the compensation model, are determined based on the reference body. Furthermore, an approximate compensation method is derived to increase the applicability of the compensation method, and its application requirement is also studied. The effectiveness of the proposed compensation method is demonstrated in a blackbody furnace experiment and a water cup experiment. [ABSTRACT FROM AUTHOR]

This study investigates the effect of electrospun nylon nanofibres modification on the low-velocity impact resistance of carbon fibre reinforced polymer (CFRP) laminates. Integration of multiple excitation infrared thermography using advanced image processing techniques is used to characterize the damage as a function of impact energy. The results show an enhanced damage resistance for the modified specimens, which is also validated by force – displacement curves. The interlaminar cracks show a maximum reduction of 24.8%, while delaminated areas and defect depths experience reductions of up to 28.7% and 24.3%, respectively. Interestingly, thermal diffusivity is introduced to identify the damage shape successfully. Finally, a correlation study is conducted, and a novel damage prediction model based on thermal diffusivity is proposed for further quantitative analysis. [Display omitted] • Electrospun nylon nanofibre veils are applied for interlayer toughening to enhance impact resistance. • Infrared thermography is firstly used to quantitatively evaluate severity of impact damage in composite. • Force – displacement curve is used to validate experimental results. • Thermal diffusivity map is produced by a novel quantitative model to establish a link to the impact energy. [ABSTRACT FROM AUTHOR]

Carbon fibre reinforced polymer (CFRP) composites are commonly used in the aerospace industry due to their excellent mechanical properties. The extreme mechanical environments make CFRP to exceed its yield limit, which is associated with composite defects, such as delaminations and cracks. Therefore, it is important to develop a stable, non-contact and highly accurate stress measurement technique. In this work, a novel internal stress measurement technique is proposed being based on the elastic plate theory and thermal stress analysis. By analyzing the temperature distribution on the material external surface, one can evaluate composite mechanical behaviour and internal stresses. To validate the proposed novel method, the theoretical investigations, including numerical simulations and experiments, have been performed, and the experimental results were optoimized by applying the Levenberg-Marquardt. Theoretical and experimental results have been used to predict the force loaded on the sample surface. Furthermore, the effects of coating on the stress inversion problem have also been discussed. [Display omitted] • Infrared thermography has been first proposed to evaluate internal forces in CFRP composite. • The theory of elastic plate and thermoelasticity has been proposed to characterize internal forces in materials. • A force-temperature prediction model was constructed and optimized by the Levenberg-Marquardt algorithm. • The proposed model has been theoretically and experimentally validated on both isotropic and anisotropic materials. [ABSTRACT FROM AUTHOR]

Simple Summary: The identification of the best anatomical site to ensure the efficient use of infrared thermography to measure body surface temperature variation in response to handling and transport stress may allow easy, non-invasive, real-time and practical animal welfare monitoring under commercial conditions. The objective of this study was to validate the anatomical sites for the measurement of infrared body surface temperature as a tool to monitor the pigs' response to handling and transport stress. Based on the greatest variation and the closer association with heart rate and salivary cortisol found in this study, the orbital and behind ear regions (in the head) appear to be reliable points for measuring body surface temperature through the technique of thermography in response to handling and transport stress in pigs. However, based on the low to moderate correlations with other physiological indicators, infrared thermography cannot be used as a stand-alone measurement of the physiological condition of pigs in response to stress. Therefore, an appropriate use of infrared technology combined with other physiological stress indicators, e.g., heart rate, blood lactate or salivary cortisol may provide the swine industry with a tool for a real-time evaluation of the physiological condition of pigs during handling and may help to monitor critical areas during the pre-slaughter process improving animal welfare and control meat quality variation. This study aimed at validating the anatomical sites for the measurement of infrared (IR) body surface temperature as a tool to monitor the pigs' response to handling and transport stress. The selected anatomical sites were the neck (infrared neck temperature—IRNT), rump (infrared rump temperature—IRRT), orbital (infrared orbital temperature—IROT) and behind ears (infrared behind ears temperature—IRBET) regions. A total of 120 pigs were handled from the finishing pen to the loading dock through a handling test course. Two handling types (gentle vs. rough) and number of laps (1 vs. 3) were applied according to a 2 × 2 factorial design. After loading, pigs were transported for 40 min and returned to their home pens. Animal behavior, heart rate, rectal temperature and salivary cortisol concentration were measured for validation. Increased IR body temperature, heart rate and salivary cortisol levels were observed in response to rough handling and longer distance walk (P < 0.05 for all). The greatest correlations were found between IROT and IRBET temperatures and salivary cortisol concentration at the end of the handling test (r = 0.49 and r = 0.50, respectively; P < 0.001 for both). Therefore, IR pig's head surface temperature may be useful for a comprehensive assessment of the physiological response to handling and transport stress. [ABSTRACT FROM AUTHOR]

• We have improved the calculation of the principal component analysis which is called Candid Covariance-Free Incremental Principal Component Thermography (CCIPCT). • We have tested the algorithm with CFRP, Plexiglas and Aluminum specimens. • The computational load of the algorithm compared to PCT and PCA. • CCIPCT provided faster performance in higher sequential acquisition. • A color based K-medoids clustering approach used for segmentation of defects. Infrared Non-destructive Testing (IRNDT) applications are unequivocally expanded and portend a commodity to improve the quality of defect detection in different fields such as aviation and industrial methods to arts and archaeology. The proposed approach focuses on the application of low-rank sparse principal component thermography (Sparse-PCT or SPCT) to assess the advantages and drawbacks of the method for non-destructive testing. For benchmarking the approach, two types of infrared image sets are tested: the Square Pulse Thermography (SPT) method for two hybrid composites (carbon and flax fiber reinforced epoxy), and passive infrared test of Bell Tower and the University of L'Aquila (AQ) faculty's wall infrared sets. The quantitative assessment of the approach is also compared for every method and indicate considerable segmentation performance where other similar approaches were not able to detect the defects. SPCT performance was compared to some popular decomposition methods such as principal component thermography (PCT), candid covariance-free incremental principal component thermography (CCIPCT), non-negative matrix factorization (NMF) using gradient descent (GD) or non-negative least square (NNLS). The comparative results demonstrate the considerable performance while the other methods failed. [ABSTRACT FROM AUTHOR]

The application of the thermal and infrared technology in different areas of research is considerably increasing. These applications involve nondestructive testing, medical analysis (computer aid diagnosis/detection—CAD), and arts and archeology, among many others. In the arts and archeology field, infrared technology provides significant contributions in terms of finding defects of possible impaired regions. This has been done through a wide range of different thermographic experiments and infrared methods. The proposed approach here focuses on application of some known factor analysis methods such as standard nonnegative matrix factorization (NMF) optimized by gradient-descent-based multiplicative rules (SNMF1) and standard NMF optimized by nonnegative least squares active-set algorithm (SNMF2) and eigen-decomposition approaches such as principal component analysis (PCA) in thermography, and candid covariance-free incremental principal component analysis in thermography to obtain the thermal features. On the one hand, these methods are usually applied as preprocessing before clustering for the purpose of segmentation of possible defects. On the other hand, a wavelet-based data fusion combines the data of each method with PCA to increase the accuracy of the algorithm. The quantitative assessment of these approaches indicates considerable segmentation along with the reasonable computational complexity. It shows the promising performance and demonstrated a confirmation for the outlined properties. In particular, a polychromatic wooden statue, a fresco, a painting on canvas, and a building were analyzed using the above-mentioned methods, and the accuracy of defect (or targeted) region segmentation up to 71.98%, 57.10%, 49.27%, and 68.53% was obtained, respectively. [ABSTRACT FROM AUTHOR]

Passive infrared thermography (IRT) has been introduced as a faster, safer, and contactless alternative for the nondestructive inspection of subsurface delamination in concrete infrastructure. However, some uncertainties remain, such as the absence of an inspection protocol to inspect multiple concrete components and the dependence of the technique's performance on environmental conditions and solar energy. This study presents a proof of concept about the capabilities of passive IRT in detecting subsurface damages in multiple concrete components under variate solar exposure. The passive IRT capabilities are explored by analyzing the thermal sequences obtained from samples of artificially damaged concrete structures inspected over 24 h in various environmental conditions over three seasons. As a result, damages with a size-to-depth (S/D) ratio between 1.0 and 2.7 were detected using the thermal contrast method. Furthermore, the implementation of signal processing techniques yielded an improvement in capabilities ranging from 15% to 52%. Finally, a procedure for acquiring data while inspecting multiple concrete components using passive IRT is proposed. • Infrared thermography detection of concrete delamination under varying solar exposure. • Procedure to optimize data collection in concrete under varying solar exposure. • Detection of damages up to 8.0 cm depth in concrete with a 1.0 width-to-depth ratio. • The specific capability varied based on the component type and solar orientation. • Signal processing mostly improved contrast-to-noise ratio and detection capabilities. [ABSTRACT FROM AUTHOR]