TY - INPR A1 - Ahmadi, Samim A1 - Lecompagnon, Julien A1 - Hirsch, Philipp Daniel A1 - Burgholzer, P. A1 - Jung, P. A1 - Caire, G. T1 - Laser excited super resolution thermal imaging for nondestructive inspection of internal defects N2 - A photothermal super resolution technique is proposed for an improved inspection of internal defects. To evaluate the potential of the laser-based thermographic technique, an additively manufactured stainless steel specimen with closely spaced internal cavities is used. Four different experimental configurations in transmission, reflection, stepwise and continuous scanning are investigated. The applied image post-processing method is based on compressed sensing and makes use of the block sparsity from multiple measurement events. This concerted approach of experimental measurement strategy and numerical optimization enables the resolution of internal defects and outperforms conventional thermographic inspection techniques. KW - Super resolution KW - Photothermal KW - Imaging KW - Compressed sensing KW - Internal defects KW - Nondestructive testing PY - 2020 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-518866 DO - https://doi.org/10.48550/arXiv.2007.03341 SN - 2331-8422 SP - 1 EP - 9 PB - Cornell University CY - Ithaca, NY AN - OPUS4-51886 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - INPR A1 - Ahmadi, Samim A1 - Hauffen, Jan Christian A1 - Kästner, L. A1 - Jung, P. A1 - Caire, G. A1 - Ziegler, Mathias T1 - Learned block iterative shrinkage thresholding algorithm for photothermal super resolution imaging N2 - Block-sparse regularization is already well-known in active thermal imaging and is used for multiple measurement based inverse problems. The main bottleneck of this method is the choice of regularization parameters which differs for each experiment. To avoid time-consuming manually selected regularization parameters, we propose a learned block-sparse optimization approach using an iterative algorithm unfolded into a deep neural network. More precisely, we show the benefits of using a learned block iterative shrinkage thresholding algorithm that is able to learn the choice of regularization parameters. In addition, this algorithm enables the determination of a suitable weight matrix to solve the underlying inverse problem. Therefore, in this paper we present the algorithm and compare it with state of the art block iterative shrinkage thresholding using synthetically generated test data and experimental test data from active thermography for defect reconstruction. Our results show that the use of the learned block-sparse optimization approach provides smaller normalized mean square errors for a small fixed number of iterations than without learning. Thus, this new approach allows to improve the convergence speed and only needs a few iterations to generate accurate defect reconstruction in photothermal super resolution imaging. KW - Iterative shrinkage thresholding algorithm KW - Neural network KW - Deep learning KW - Active thermography KW - Photothermal super resolution PY - 2020 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-525364 DO - https://doi.org/10.48550/arXiv.2012.03547 SN - 2331-8422 SP - 1 EP - 11 PB - Cornell University CY - Ithaca, NY AN - OPUS4-52536 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - INPR A1 - Ahmadi, Samim A1 - Kästner, L. A1 - Hauffen, Jan Christian A1 - Jung, P. A1 - Ziegler, Mathias T1 - Photothermal-SR-Net: A customized deep unfolding neural network for photothermal super resolution imaging N2 - This paper presents deep unfolding neural networks to handle inverse problems in photothermal radiometry enabling super resolution (SR) imaging. Photothermal imaging is a well-known technique in active thermography for nondestructive inspection of defects in materials such as metals or composites. A grand challenge of active thermography is to overcome the spatial resolution limitation imposed by heat diffusion in order to accurately resolve each defect. The photothermal SR approach enables to extract high-frequency spatial components based on the deconvolution with the thermal point spread function. However, stable deconvolution can only be achieved by using the sparse structure of defect patterns, which often requires tedious, hand-crafted tuning of hyperparameters and results in computationally intensive algorithms. On this account, Photothermal-SR-Net is proposed in this paper, which performs deconvolution by deep unfolding considering the underlying physics. This enables to super resolve 2D thermal images for nondestructive testing with a substantially improved convergence rate. Since defects appear sparsely in materials, Photothermal-SR-Net applies trained blocksparsity thresholding to the acquired thermal images in each convolutional layer. The performance of the proposed approach is evaluated and discussed using various deep unfolding and thresholding approaches applied to 2D thermal images. Subsequently, studies are conducted on how to increase the reconstruction quality and the computational performance of Photothermal-SR-Net is evaluated. Thereby, it was found that the computing time for creating high-resolution images could be significantly reduced without decreasing the reconstruction quality by using pixel binning as a preprocessing step. KW - Photothermal super resolution KW - Nondestructive testing KW - Deep unfolding KW - Deep learning KW - Deep imaging KW - Physics-based deep learning KW - Laser thermography KW - Elastic net KW - Iterative shrinkage thresholding algorithm PY - 2021 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-525371 DO - https://doi.org/10.48550/arXiv.2104.10563 SN - 2331-8422 SP - 1 EP - 10 PB - Cornell University CY - Ithaca, NY AN - OPUS4-52537 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - INPR A1 - Yagdjian, H. A1 - Lecompagnon, Julien A1 - Hirsch, Philipp Daniel A1 - Gurka, M. T1 - Optimization of thermal shock response spectrum as infrared thermography post-processing methodology using Latin hypercube sampling and analytical thermal N-layer model N2 - In this work, we continue to develop and investigate the Thermal Shock Response Spectrum (TSRS) method as an alternative data processing method for infrared thermography (IRT). We focus on improving the current TSRS algorithm and present an optimization methodology for finding the optimal thermal Q-factor and characteristic frequency pair, which is based on the widely applied random sampling method. We show the qualitative relationship between the determined optimal characteristic frequency and the corresponding maximum difference in diffusion length between reference and defective models, as calculated by selecting a specific one-dimensional thermal N-layer model. The investigations were performed on an inhomogeneous plate made of carbon fiber reinforced polymer (CFRP) with artificial square defects at different depths. Furthermore, two different heat sources were used: a xenon flash lamp and a laser. These sources are not only distinct by their underlying physics but also generate inherently different pulse shapes. To quantitatively estimate the contrast between defect and non-defect areas, and to compare these results with commonly used infrared thermography (IRT) data post-processing methods such as Pulse Phase Thermography (PPT) and Thermographic Signal Reconstruction (TSR), the Tanimoto criterion (TC) and signal-tonoise ratio (SNR) were used. KW - Infrared thermography KW - Composite materials KW - TSRS optimization KW - Defect identification KW - Heat source shape KW - N-layers model KW - Latin hypercube sampling PY - 2024 UR - https://ssrn.com/abstract=4910240 SP - 1 EP - 21 PB - Elsevier CY - New York, NY AN - OPUS4-60734 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - INPR A1 - Kunji Purayil, Sruthi Krishna A1 - Aroliveetil, S. A1 - Chaturvedi, A. A1 - Balasubramaniam, K. T1 - Simulation-assisted Multimodal Deep Learning (Sim-MDL) Fusion Models for the Evaluation of Thermal Barrier Coatings using Infrared Thermography and Terahertz Imaging N2 - Thermal Barrier Coatings (TBCs) are critical components in high-temperature applications, such as gas turbines and aerospace engines, where they protect the underlying substrate metals from extreme thermal stress and extend component life. Accurate evaluation of TBCs is essential to improve operational efficiency, optimize predictive maintenance strategies, and extend component life. Popular non-destructive evaluation (NDE) techniques such as infrared thermography (IRT) and terahertz (THz) imaging have been widely used for TBC inspection with limitations when used independently, including sensitivity to surface conditions, limited penetration depth, and challenges in inspecting multi-layer coatings and detecting subsurface defects. To address these challenges, our study proposes a novel framework called simulation-assisted multimodal deep learning (Sim-MDL) that integrates the strengths of IRT and THz imaging for a comprehensive evaluation of TBCs. To generalize the study to a range of thermophysical properties of TBCs, the study uses simulation-generated data along with experimental data for training deep learning models. The data from IRT and THz modalities are fused in the Sim-MDL models to enable characterization of the TBC topcoat layer. IRT and THz experimental data, together with simulations, form a large dataset that is used to train deep learning models. The framework is tested and optimized for multimodal data fusion using two DL architectures based on convolutional neural networks (CNN) and long short-term memory (LSTM), allowing the model to learn correlations and complex patterns across the IRT and THz modalities. The study is conducted on four newly coated samples ranging in thickness from 24 to 120 µm. An attention-based LSTM model trained on both simulation and experimental data shows high prediction accuracy with MAPE values ranging from 2.06–4.43% for thermal conductivity, 2.05–3.57% for heat capacity, 11.53–1.75% for topcoat thickness, and 0.27–1.05% for refractive index, respectively, for the topcoat layers of four samples. Our model outperformed the single-modality models and conventional parameter estimation methods in terms of accuracy and robustness, highlighting the potential of multimodal data for automated analysis of TBC in industrial settings. KW - Infrared Thermography KW - Non-destructive Evaluation (NDE) KW - Multimodal Fusion KW - Terahertz Imaging KW - Deep Learning PY - 2025 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-638960 DO - https://doi.org/10.21203/rs.3.rs-7206285/v1 SP - 1 EP - 27 PB - Springer Science and Business Media LLC AN - OPUS4-63896 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - INPR A1 - Zhu, Pengfei A1 - Maldague, Xavier T1 - Principal Component Analysis-Based Terahertz Self-Supervised Denoising and Deblurring Deep Neural Networks N2 - Terahertz (THz) systems inherently introduce frequency-dependent degradation effects, resulting in low-frequency blurring and high-frequency noise in amplitude images. Conventional image processing techniques cannot simultaneously address both issues, and manual intervention is often required due to the unknown boundary between denoising and deblurring. To tackle this challenge, we propose a principal component analysis (PCA)-based THz self-supervised denoising and deblurring network (THz-SSDD). The network employs a Recorrupted-to-Recorrupted self-supervised learning strategy to capture the intrinsic features of noise by exploiting invariance under repeated corruption. PCA decomposition and reconstruction are then applied to restore images across both low and high frequencies. The performance of the THz-SSDD network was evaluated on four types of samples. Training requires only a small set of unlabeled noisy images, and testing across samples with different material properties and measurement modes demonstrates effective denoising and deblurring. Quantitative analysis further validates the network’s feasibility, showing improvements in image quality while preserving the physical characteristics of the original signals. KW - Non-destructive testing (NDT) KW - Terahertz KW - Denoising KW - Self-supervised learning KW - Deblurring PY - 2026 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-654043 DO - https://doi.org/10.48550/arXiv.2601.12149 SN - 2331-8422 SP - 1 EP - 9 PB - Cornell University CY - Ithaca, NY AN - OPUS4-65404 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - INPR A1 - Nerger, Tino A1 - Neumann, Patrick P. A1 - Weller, Michael G. T1 - Drone-Based Localization of Hazardous Chemicals by Passive Smart Dust N2 - The distribution of tiny sensors over a specific area was first proposed in the late 1990s as a concept known as Smart Dust. Several efforts focused primarily on computing and networking capabilities but quickly ran into problems related to power supply, cost, data transmission, and environmental pollution. To overcome these limitations, we propose using paper-based (confetti-like) chemosensors that exploit the inherent selectivity of chemical reagents, such as colorimetric indicators. In this work, cheap and biodegradable passive sensors made from cellulose could successfully indicate the presence of hazardous chemicals, e.g., strong acids, by a significant color change. A conventional color digital camera attached to a drone could easily detect this from a safe distance. The collected data was processed to define the hazardous area. Our work presents a combination of the smart dust concept, chemosensing, paper-based sensor technology, and low-cost drones for flexible, sensitive, economical, and rapid detection of hazardous chemicals in high-risk scenarios. N2 - Die Verteilung winziger Sensoren über ein bestimmtes Gebiet wurde erstmals Ende der 1990er Jahre als Konzept namens „Smart Dust“ vorgeschlagen. Mehrere Bemühungen konzentrierten sich hauptsächlich auf Rechen- und Netzwerkfähigkeiten, stießen jedoch schnell auf Probleme im Zusammenhang mit der Stromversorgung, den Kosten, der Datenübertragung und der Umweltverschmutzung. Um diese Einschränkungen zu überwinden, schlagen wir die Verwendung von papierbasierten (konfettiartigen) Chemosensoren vor, die die inhärente Selektivität chemischer Reagenzien, wie z. B. kolorimetrischer Indikatoren, nutzen. In dieser Arbeit konnten günstige und biologisch abbaubare passive Sensoren aus Zellulose erfolgreich das Vorhandensein gefährlicher Chemikalien, z.B. starker Säuren, durch eine deutliche Farbänderung anzeigen. Eine herkömmliche Farb-Digitalkamera, die an einer Drohne befestigt ist, konnte dies aus sicherer Entfernung leicht erkennen. Die gesammelten Daten wurden verarbeitet, um den Gefahrenbereich zu abzugrenzen. Unsere Arbeit stellt eine Kombination aus dem Smart-Dust-Konzept, Chemosensorik, papierbasierter Sensortechnologie und kostengünstigen Drohnen für eine flexible, empfindliche, wirtschaftliche und schnelle Erkennung gefährlicher Chemikalien in Hochrisikoszenarien dar. KW - Confetti KW - Remote sensing KW - Drones KW - UAV KW - Optical detection KW - Chemosensor KW - pH indicator KW - Paper-based sensors KW - Harmful chemicals KW - Chemical desaster PY - 2024 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-621284 DO - https://doi.org/10.20944/preprints202408.0030.v1 SP - 1 EP - 17 PB - MDPI CY - Basle, Switzerland AN - OPUS4-62128 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER -