A novel IR-SRGAN assisted super-resolution evaluation of photothermal coherence tomography for impact damage in toughened thermoplastic CFRP laminates under room and low temperature

Electromagnetic induced infrared thermography for fast detection of interfacial debonding in coating-metal multilayer structure

Operando Infrared Thermography and Quantitative Insights into Lithium Plating Dynamics on Electrodes during Fast Charging

ABSTRACT Assessing heat loss in existing building envelopes is essential for implementing energy-efficient retrofitting projects. However, evaluating envelope thermal performance using infrared thermography (IRT) remains challenging because of limited detection range and the lack of standardized metrics and comparative methods. This study proposes an unmanned aerial vehicle (UAV)-based IRT method for heat-loss assessment. First, infrared digital orthophoto maps (IR-DOMs) of building facades are generated using photogrammetric techniques, enabling extraction of infrared (IR) temperatures from external walls and windows. A dynamic prediction model, referred to as the ‘standard temperature’, is then developed from numerically generated reference samples and regression analysis to provide reference temperatures for comparison with measured IR data. Finally, by analyzing the difference between the measured average IR temperature and the standard temperature, together with the average heat-loss discrepancy, the method introduces quantitative indicators for grading and evaluating envelope thermal performance. The methodology was examined using heating energy consumption data from five buildings in a cold region of China. The resulting heat-loss grades were generally consistent with both annual and 24-hour heating use, indicating that the proposed indicators can support comparative evaluation across multiple buildings and help prioritize retrofit candidates.

Context Residual feed intake (RFI) and residual intake and gain (RIG) are widely used metrics to assess feed efficiency in sheep. However, their practical implementation in breeding programs remains limited owing to the high cost and complexity of individual feed intake measurements. Consequently, alternative indicators, such as physiological responses and infrared thermography (IRT), have been proposed as potential proxies for identifying animals with superior feed efficiency under different environmental conditions. Aims This study aimed to evaluate the relationships between RFI and RIG classifications and physiological parameters, as well as surface body temperatures obtained via IRT, in Texel ewes exposed to natural heat stress. Methods Thirty-nine young Texel ewes were monitored for 57 days in a covered facility equipped with an automated feeding and watering system with individual intake recording. Animals were classified into low, medium, and high-efficiency groups according to their RFI and RIG values. Physiological responses, including respiratory rate (RR), heart rate (HR), rectal temperature (RT), and the heat tolerance coefficient (HTC), were recorded. Additionally, surface temperatures of the eye, muzzle, hooves, and vulva were measured using infrared thermography (IRT). Statistical analyses included ANOVA and principal component analysis (PCA) to explore associations among traits. Key results No significant differences were detected among RFI or RIG classes for RR, HR, RT, or HTC. Similarly, IRT-derived surface temperatures did not differ across efficiency classifications. PCA showed that RR and HTC explained the greatest proportion of total variance, whereas RFI and RIG contributed to other independent components. Conclusions Neither physiological parameters nor IRT-based surface temperatures were effective indicators of feed efficiency in Texel ewes under natural heat stress. Implications The results indicated that RFI and RIG cannot be accurately inferred from physiological or IRT variables under field conditions. Future research should integrate additional phenotypic and behavioral indicators to identify reliable, low-cost biomarkers for metabolic efficiency in sheep.

This contribution presents a combined approach for in-situ experimental characterisation and numerical modelling of thermo-mechanical behaviour in directed energy deposition (DED). Full-field temperature and substrate deformation are measured simultaneously using infrared (IR) thermography and stereo digital image correlation (DIC) during laser-beam powder deposition on a thin substrate. The experimental data are used to calibrate thermal boundary conditions and to validate a macroscopic finite-element model. The validated framework is then applied to compare different deposition strategies, demonstrating the capability of the coupled measurements and simulations to capture transient thermal fields, deformation evolution and toolpath-dependent effects relevant for process optimisation.

Abstract Non-Destructive Testing (NDT) has become an essential technology for modern industrial maintenance and the inspection of various materials. Amongst different NDT technologies, Infrared Thermography (IRT) offers a robust framework to reveal the subsurface anomalies inside the sample. However, conventional Infrared Imaging (IR) methods are often time consuming, manual and incapable of automatic distinguishing defects from non-defective regions. To address the issue, an automatic defect detection framework is proposed, which integrates an unsupervised autoencoder learning approach with image fusion. Autoencoder is trained on non-defective regions to model the heat distribution patterns. During testing, defective frames show large reconstruction errors and obtained frames are further used for segmentation. The defective frames are segmented using adaptive thresholding to highlight regions with high reconstruction error and generate binary masks that localize potential defect areas. Subsequently, Signal to Noise Ratio (SNR) is computed across each defective region to estimate the thermal contrast. This SNR directs the fusion across all the defective frames of the given defect at given spatial location. The proposed approach estimated the reconstructed and accurate frame with automatic identification of defective frames, achieving high detection sensitivity across varying defect depths. The integration of infrared imaging with neural network based autoencoder substantially enhances the efficiency, reliability, and depth resolving capability of thermal-based defect detection system.

ABSTRACT Accurate online detection and quantitative evaluation of subsurface defects are essential for improving the reliability of wind energy systems. To overcome the limitations of conventional non-destructive testing techniques, which are typically confined to laboratory or manufacturing settings, this study proposes an inverse problem-solving framework that integrates drone-mounted active infrared thermography with Physics-Informed Neural Networks (PINNs). Transient surface temperature data of defect-containing glass fiber reinforced polymer (GFRP) specimens are obtained via an infrared imaging system deployed on an Unmanned Aerial Vehicle (UAV) platform. A PINN-based forward model is established to simulate heat conduction within the material, and Bayesian Optimization (BO) is integrated into the PINN framework (BO-PINN) to construct an inverse solver that simultaneously estimates defect depth and thickness from surface temperature data. Experimental results demonstrate that the proposed model achieves a depth estimation error below 2% and a relative thickness error below 10%, significantly outperforming conventional methods. This work provides a novel predictive framework and technical route for the non-destructive evaluation and service life prediction of GFRP composite structures.

One of the most common consequences in individuals with diabetes is the diabetic foot, which can cause foot ulcers and even lead to limb amputation. Since an increase of the temperature in the plantar region is directly correlated with an increased risk of ulceration, infrared thermography (IRT) has been used in multiple studies as an automatic tool for detecting problems in diabetic foot. Artificial intelligence-based computer-aided diagnosis systems are being more frequently used to improve decision-making and minimize errors. These technologies are designed to increase examination accuracy, consistency in image interpretation, prognosis evaluation support, and examination accuracy. They also have the ability to offer insightful information and help medical professionals to manage diabetic foot issues successfully. In this work, 37 papers that used thermography and artificial intelligence (AI) to identify diabetic foot complications and/or predict the risk of developing diabetic foot are analyzed. The results demonstrate the potential of IRT imaging implementation with AI for the identification and prediction of diabetic foot complications. The combination of IRT and AI shows significant potential for diabetic foot assessment; however, the great majority of these studies show that the research is confined to classification of foot thermograms using pre-prepared data sets. In particular, there is limited research on segmentation methods and constraints in the use of deep learning due to the lack of large and diverse datasets.

ABSTRACT A multi‐source monitoring approach integrating acoustic emission (AE), infrared thermography (IRT), and digital image correlation (DIC) was employed to investigate the compression‐after‐impact (CAI) failure mechanisms and damage tolerance of T800 carbon fiber reinforced polymer (CFRP) composites. CAI failure modes at representative impact energies were identified through load–displacement responses, out‐of‐plane displacement fields, and fracture morphologies. Damage modes were classified using principal component analysis combined with a Gaussian mixture model based on AE time‐frequency parameters. Thermodynamic responses during failure were characterized by IRT, while full‐field strain evolution and out‐of‐plane displacement distributions under different impact energies were analyzed using DIC. Pearson correlation analysis was further conducted to evaluate the relationships between key monitoring parameters and CAI damage tolerance. The results show that the displacement at peak load exhibits a non‐monotonic trend with increasing impact energy, reflecting alternating stiffness‐ and strength‐dominated behaviors. AE results reveal progressive damage evolution from matrix cracking to delamination and fiber fracture. At high impact energies, dual‐zone thermal hotspots correspond well with partitioned out‐of‐plane displacement fields, confirming localized buckling instability. Strong correlations are observed between CAI damage tolerance and AE energy, maximum average temperature rise, and axial strain. This study, validated through laboratory CAI tests, establishes an AE‐IRT‐DIC multi‐source synergistic monitoring and feature‐parameter screening methodology. It not only provides multi‐physics experimental evidence for a deeper understanding of CAI failure in composites, but also targets future in‐service structural health monitoring by informing the selection of key features required for post‐impact risk assessment.

Preterm infants are particularly burdened by wired monitoring systems. Contactless methods such as infrared thermography (IRT) and photoplethysmography imaging (PPGI) may offer a less invasive alternative. This study presents the feasibility and iterative development of these techniques during routine neonatal care.IRT was used to assess respiratory activity and skin temperature across various care settings (incubator, kangaroo care, heated bed) from 2011 to 2023. PPGI was applied between 2016 and 2021 to measure cardiac activity and perfusion, accompanied by technical refinement. Accuracy was evaluated through comparative studies using established reference standards, including ECG.Both IRT and PPGI reliably detected respiratory rate, skin temperature (including stress-related facial changes), cardiac activity, and perfusion in preterm infants.Advances in sensor miniaturization and data processing suggest that, following further prospective evaluation, contactless vital sign monitoring could be feasibly integrated into routine neonatal care.

Knowledge PET3D: An interpretable framework for 3D near-miss detection in thermal traffic video

Comparative assessment of quantitative infrared thermography approaches for experimental thermal transmittance determination using UAVs

Rheumatoid arthritis (RA) is a chronic inflammatory disease characterized by pain, swelling, stiffness, and loss of joint function, making early diagnosis challenging. The study aims to assess the differences between RA patients (n = 70) and healthy individuals (n = 30) while classifying Ritchie Articular Index (RAI) values (0-3) based on inflammation levels using artificial intelligence algorithms. Metacarpophalangeal (MCP), and proximal-interphalangeal (PIP) joints were analyzed for the degree of inflammation. Static thermal data was collected from individuals at rest in a controlled environment. Then, alcohol was applied to the participants' hand regions, followed by a 180-second thermal video recording of the same region. In the pre-processing step, background noise cleaning and alignment were performed. Background was eliminated using Snake algorithm. Thermal video recordings were aligned using Scale Invariant Feature Transform (SIFT) algorithm. The Skeletonization algorithm was employed to detect fingers and joint regions in the images. For static thermal analysis, initial temperature ([Formula: see text]) values were extracted from the resting thermogram data. In dynamic thermal analysis, the temperature parameters [Formula: see text], [Formula: see text], [Formula: see text], and [Formula: see text] were calculated. A statistical analysis of the four temperature parameters across different RAI values revealed that [Formula: see text] (p = 0.025) and [Formula: see text] (p = 0.042) exhibited statistically significant differences among the four RAI levels. Machine learning models were trained using the resting temperature values of patient and healthy groups, and the SVM achieved the highest success rate of 93%. It is believed that the proposed system may help diagnose RA in clinical settings and contribute to determining the severity of inflammation.

Abstract Computer vision has become indispensable to precision agriculture; however, the scarcity and imbalance of labeled data across crops, livestock, and aquaculture domains continue to limit the robustness and generalization of conventional convolutional neural networks (CNNs). We present AgriGen-XL, a physics-aware generative augmentation framework that integrates generative modeling with task-specific CNN backbones, including EfficientNet, DeepLabV3 +, and YOLOv5. The framework combines conditional diffusion models informed by physical and sensor priors with efficient adversarial sampling and couples these generators with domain-adaptive CNNs for classification, detection, and segmentation. AgriGen-XL is evaluated across three representative agricultural domains: (i) crop disease diagnosis and phenotyping from field and UAV imagery, (ii) livestock activity and welfare monitoring using RGB and thermal video, and (iii) aquaculture fish segmentation, counting, and biomass estimation under variable turbidity. The proposed system introduces three key contributions: a cross-domain synthetic data engine that encodes physical priors (e.g., illumination, canopy geometry, and turbidity) and sensor characteristics (e.g., spectral response and thermal noise); a rare-class up-sampling and scene-composition curriculum to address long-tail data distributions; and a precision analytics layer that converts model outputs into parcel-level disease risk, animal-level welfare indices, and cage-level biomass forecasts. Relative to real-only training with standard augmentations and non-physics generative baselines, synthetic augmentation with AgriGen-XL improves downstream performance by + 3–11 absolute points across classification, detection, segmentation, and counting tasks, reduces rare/long-tail class error by 22–38% (e.g., early disease stages and lameness), and improves cross-season crop generalization by + 7.4 accuracy points. Ablation studies show that gains saturate at a 2 × synthetic-to-real data ratio and that physics-informed generators consistently outperform style-based alternatives. Code, configuration files, and example synthetic asset recipes will be made available upon publication. All real datasets are sourced from publicly accessible repositories to support reproducibility.

Infrared thermography (IRT) has recently gained attention in the field of exercise physiology, due to its ability to monitor thermoregulatory and cardiopulmonary responses non-invasively and in real time during physical exercise. However, the reproducibility of intra-individual measurement and standardization of region-of-interest selection in relation to the acute exercise response remain inconclusive. This study aimed to examine the reproducibility and physiological relevance of specific skin temperature (TSK) metrics processed automatically using deep learning-assisted IRT during running, and to synchronize these metrics with cardiopulmonary and thermoregulatory parameters. Eleven endurance-trained individuals performed three 46-min running sessions over 2days, with the same average external load but different intensity distributions. Individual anaerobic threshold velocity (vIAT), previously determined by cardiopulmonary exercise testing, was used to prescribe running intensity. During exercise, oxygen consumption (VO2), core temperature (TCORE), heart rate (HR) and different TSK metrics, including non-vessel (TNV), cutaneous arterial perforator (TP), and superficial vein patterns, were continuously measured. All TSK metrics displayed consistent temporal dynamics aligned with external load, but their absolute temperature levels differed systematically. During intermittent running and recovery, TP exhibited robust correlations with HR and VO2 (r = - 0.63 to - 0.9, p < 0.001), and TP entropy showed consistent associations with TCORE during the warm-up (r = 0.59-0.83, p < 0.001). This indicates uniform response patterns across the cohort. In contrast, TNV demonstrated heterogeneous correlations with TCORE, depending on individual exercise capacity. A strong inverse correlation was identified between ∆TNV and vIAT (r = - 0.74 to - 0.88, p ≤ 0.009) and individuals with higher vIAT demonstrated greater TCORE-TNV gradients during running. Measurements of ∆TNV demonstrated high reproducibility, with intra-individual ICC(3,1) values of 0.89 for recovery and 0.76 for warm-up, and no statistically significant differences between the three sessions. Deep learning-assisted IRT provides reproducible, physiologically consistent metrics across repeated exercise sessions, regardless of the day or prior load. Distinct TSK metrics capture both uniform and individual-specific thermoregulatory responses. Variability in peripheral temperature regulation is more strongly associated with running velocity at the individual anaerobic threshold than with maximal cardiorespiratory fitness.

The aim of this study was to determine whether a low-intensity exercise may influence outcomes of skin temperature measured by infrared thermography (IRT) in the presence or not of delayed-onset muscle soreness (DOMS). Seventeen participants were recruited (12 men, age 24 ± 6 years, body mass 72.6 ± 12.5 kg, and height 1.8 ± 0.1 m). They visited the laboratory on two days with 48 h in between. On day 1, participants performed a 10-minute low-intensity treadmill walk followed by a squat protocol to induce DOMS (10 sets of 10 repetitions at 70% of body mass). On day 2, when participants reported DOMS, they returned to the laboratory and completed another 10-minute low-intensity treadmill walk. On each day, quadriceps DOMS was assessed using a numerical pain rating scale, and skin temperature was measured using IRT on both days. The skin temperature was higher before (p = 0.01; d=-0.44) and after walking (p = 0.03; d=-0.39) on the second day when significant DOMS was reported by the participants. However, skin temperature was not altered by light walking prior to the measurements on day 2. In conclusion, low-intensity walking did not induce detectable changes in skin temperature in the quadriceps before or after DOMS, suggesting that that acute, low-intensity exercise is unlikely to be a relevant confounder or enhancer of IRT-based assessment of DOMS.

This study investigated the use of stress-related blood biomarkers (cortisol, creatine kinase - CK, and lactate dehydrogenase - LDH) and infrared thermography (IRT) to predict meat quality in Nellore cattle under tropical conditions. These biomarkers reflect metabolic and stress pathways that influence muscle glycogen depletion before slaughter, contributing to elevated pHu and increased dark, firm, and dry (DFD) incidence. A total of 389 bulls from seven batches on the same farm were assessed at slaughter. Carcasses were classified by ultimate pH (pHu) as normal (pHu < 5.8), atypical DFD (5.8 ≤ pHu < 6.0), or typical DFD (pHu ≥ 6.0). The incidence of typical and atypical DFD meat was 20.8%. The typical DFD group showed significantly higher levels of cortisol, LDH, and CK. IRT images revealed that animals in the typical DFD group exhibited a higher minimum eye temperature (P < 0.003). In comparison, the atypical DFD group showed a higher maximum eye temperature compared to the normal group. Regression models demonstrated a strong predictive relationship (R² > 0.8) between cortisol, glucose, lactate, and pHu. We conclude that integrating blood biomarker analysis (specifically cortisol, glucose, and lactate) and IRT offers a practical tool for the early identification of DFD carcasses, which could enhance quality management in the beef industry.

Abstract The conservation of historic masonry affected by rising damp requires accurate diagnosis and interventions compatible with the nature of the materials. This study presents a systematic literature review of advanced diagnostic and material characterization techniques applied to rising damp in heritage buildings. Searches were conducted in eight databases (Scopus, Web of Science, Wiley Online Library, Engineering Village, Springer Nature Link, Taylor &amp; Francis, MDPI and ScienceDirect) using structured Boolean strings adapted to each source. A total of 351 records were identified, and, after PRISMA-based screening, 44 studies were finally included. The most frequent techniques were Infrared Thermography (IRT), Ion Chromatography (IC), X-ray Diffraction (XRD), Scanning Electron Microscopy (SEM) and Mercury Intrusion Porosimetry (MIP), often combined with standardized moisture and salt tests such as EN 15,801, EN 13,755 and EN 16,322. The review organizes these methods across macroscopic, microscopic and molecular/elemental analytical levels, encompassing in situ, non-destructive and laboratory-based approaches, and highlights multimodal workflows in which preliminary screening (e.g. IRT and other in situ tools) is followed by confirmatory analyses (e.g. IC, XRD, protocol-based tests) and microstructural assessment (e.g. SEM, MIP). The findings show that integrated use of complementary techniques improves the understanding of moisture and salt dynamics, supports the selection of compatible materials and informs conservation strategies. The review also identifies methodological gaps, including the limited use of AI- or machine-learning-based data integration, the absence of standard calibration materials for historic substrates and the need for low-cost diagnostic tools in contexts with restricted access to advanced instrumentation.

The use of stationary bikes has increased significantly, making physical activity more accessible to the general population. However, the lack of precise, non-invasive monitoring during training sessions can mask muscle imbalances and inefficient biomechanical patterns. These factors, subclinical in their initial stages, can lead to a high incidence of musculoskeletal injuries. Currently, exercise evaluation relies on the subjective perception of effort or conventional performance metrics, creating a critical gap in the early detection of physiological issues before they progress to chronic conditions. Given the crucial role of exercise in public health, it is imperative to implement sustainable evaluation strategies applicable to various fitness levels. In this context, infrared thermography (IRT) emerges as a promising alternative due to its ability to quantify skin surface temperature, a direct indicator of metabolic activity, thermoregulation, and blood flow. This project establishes a standardized methodological protocol using IRT to monitor the physiological response during cycling sessions in a non-athletic population. The application of this protocol revealed that standardized IRT can effectively detect subclinical thermal asymmetries and localized "hotspots" in joints (e.g., the knee at 32.8 °C) that are otherwise undetectable through subjective effort perception (Borg scale) or pedaling cadence monitoring. By systematically controlling environmental and subject-related variables, this research analyzes surface temperature distribution in the lower body to evaluate biomechanics and muscle fatigue. Specifically, the protocol facilitates the correlation between thermal patterns, fatigue levels, and pedaling cadence. This approach not only supports technological innovation in preventive medicine and public health but also lays the groundwork for an early detection tool capable of mitigating injury risks and optimizing the well-being of the population engaged in this activity.