TY - GEN 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 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:b43-525364 UR - https://arxiv.org/abs/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 - GEN 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 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:b43-525371 UR - https://arxiv.org/abs/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 - GEN A1 - Strangfeld, Christoph A1 - Prinz, Carsten A1 - Hase, F. A1 - Kruschwitz, Sabine T1 - Data of embedded humidity sensors, sample weights, and measured pore volume distribution for eight screed types N2 - Four cement-based and four calcium-sulphate-based screed types are investigated. The samples have a diameter of 300 mm and a height of 35 or 70 mm. Up to ten humidity sensors are embedded directly during the concreting of the screed samples. Thus, the humidity over the sample height is monitored during hardening, hydration, evaporation, and oven drying. Furthermore, the screed samples are weighted during every measurement to determine the total mass and the corresponding moisture loss. To define the pore system precisely, mercury intrusion porosimetry as well as gas adsorption is performed. According to the data, the entire pore volume distribution is known. The measured pore diameters range from 0.8 nm to 100 μm and the total porosity of the examined screeds ranges between 11 % and 22 %. Based on these measurement data, moisture transport, pore saturation as well as sorption isotherms and their hysteresis may be calculated quantitatively as described by Strangfeld et al. KW - Concrete and screed KW - Material moisture KW - Pore volume distribution KW - Embedded humidity sensors KW - Data in brief PY - 2018 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:b43-448261 UR - https://data.4tu.nl/repository/uuid:d2ba436f-78c0-4105-8a1f-5422fcb37851 SP - 1 EP - 4 PB - 4TU.Centre for Research Data CY - Delft AN - OPUS4-44826 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - GEN 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 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:kobv:b43-518866 UR - https://arxiv.org/abs/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 -