Filtern
Dokumenttyp
- Zeitschriftenartikel (4)
- Vortrag (4)
- Posterpräsentation (4)
- Beitrag zu einem Tagungsband (3)
- Sonstiges (1)
Sprache
- Englisch (16)
Schlagworte
- Laser thermography (16) (entfernen)
Organisationseinheit der BAM
Eingeladener Vortrag
- nein (4)
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.
Learned block iterative shrinkage thresholding algorithm for photothermal super resolution imaging
(2022)
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. We show the benefits of using a learned block iterative shrinkage thresholding algorithm (LBISTA) that is able to learn the choice of regularization parameters, without the need to manually select them. In addition, LBISTA enables the determination of a suitable weight matrix to solve the underlying inverse problem. Therefore, in this paper we present LBISTA and compare it with state-of-the-art block iterative shrinkage thresholding using synthetically generated 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. Thus, this allows us to improve the convergence speed and only needs a few iterations to generate accurate defect reconstruction in photothermal super-resolution imaging.
Using an infrared camera for radiometric imaging allows the contactless temperature measurement of multiple surface pixels simultaneously. From the measured surface data, a sub-surface structure, embedded inside a sample or tissue, can be reconstructed and imaged when heated by an excitation light pulse. The main drawback in radiometric imaging is the degradation of the spatial resolution with increasing depth, which results in blurred images for deeper lying structures. We circumvent this degradation with blind structured illumination, combined with a non-linear joint sparsity reconstruction algorithm. The ground-breaking concept of super-resolution can be transferred from optics to thermographic imaging.
In this paper we propose super resolution measurement and post-processing strategies that can be applied in thermography using laser line scanning. The implementation of these techniques facilitates the separation of two closely spaced defects and avoids the expected deterioration of spatial resolution due to heat diffusion. The experimental studies were performed using a high-power laser as heat source in combination with pulsed thermography measurements (step scanning) or with continuous heating measurements (continuous scanning). Our work shows that laser line step scanning as well as continuous scanning both can be used within our developed super resolution (SR) techniques. Our SR techniques make use of a compressed sensing based algorithm in post- processing, the so-called iterative joint sparsity (IJOSP) approach. The IJOSP method benefits from both - the sparse nature of defects in space as well as from the similarity of each measurement. In addition, we show further methods to improve the reconstruction quality e.g. by simple manipulations in thermal image processing such as by considering the effect of the scanning motion or by using different optimization algorithms within the IJOSP approach. These super resolution image processing methods are discussed so that the advantages and disadvantages of each method can be extracted. Our contribution thus provides new approaches for the implementation of super resolution techniques in laser line scanning thermography and informs about which experimental and post-processing parameters should be chosen to better separate two closely spaced defects.
This paper presents different super resolution reconstruction techniques to overcome the spatial resolution limits in thermography. Pseudo-random blind structured illumination from a onedimensional laser array is used as heat source for super resolution thermography. Pulsed thermography measurements using an infrared camera with a high frame rate sampling lead to a huge amount of data. To handle this large data set, thermographic reconstruction techniques are an essential step of the overall reconstruction process. Four different thermographic reconstruction techniques are analyzed based on the Fourier transform amplitude, principal component analysis, virtual wave reconstruction and the maximum thermogram. The application of those methods results in a sparse basis representation of the measured data and serves as input for a compressed sensing based algorithm called iterative joint sparsity (IJOSP). Since the thermographic reconstruction techniques have a high influence on the result of the IJOSP algorithm, this paper Highlights their Advantages and disadvantages.
Classification of Spot-Welded Joints in Laser Thermography Data Using Convolutional Neural Networks
(2021)
Spot welding is a crucial process step in various industries. However, classification of spot welding quality is still a tedious process due to the complexity and sensitivity of the test material, which drain conventional approaches to its limits. In this article, we propose an approach for quality inspection of spot weldings using images from laser thermography data. We propose data preparation approaches based on the underlying physics of spot-welded joints, heated with pulsed laser thermography by analyzing the intensity over time and derive dedicated data filters to generate training datasets. Subsequently, we utilize convolutional neural networks to classify weld quality and compare the performance of different models against each other. We achieve competitive results in terms of classifying the different welding quality classes compared to traditional approaches, reaching an accuracy of more than 95 percent. Finally, we explore the effect of different augmentation methods.
Thermographic NDE is based on the interaction of thermal waves with inhomogeneities. These inhomogeneities are related to sample geometry or material composition. Although thermography is suitable for a wide range of inhomogeneities and materials, the fundamental limitation is the diffusive nature of thermal waves and the need to measure their effect radiometrically at the sample surface only. The propagation of the thermal waves from the heat source to the inhomogeneity and to the detection surface results in a degradation in the spatial resolution of the technique. A new concerted ansatz based on a spatially structured heating and a joint sparsity of the signal ensemble allows an improved reconstruction of inhomogeneities. As a first step to establish an improved thermographic NDE method, an experimental setup was built based on structured 1D illumination using a flash lamp behind a mechanical aperture. As a follow-up to this approach, we now use direct structured illumination using a 1D laser array. The individual emitter cells are driven by a random binary pattern and additionally shifted by fractions of the cell period. The repeated measurement of these different configurations with simultaneously constant inhomogeneity allows for a reconstruction that makes use of joint sparsity. With analytical-numerical modelling or numerical FEM simulations, we study the influence of the parameters on the result of non-linear reconstruction. For example, the influence of the illumination pattern as a variable heat flux density and Neumann boundary condition for convolution with the constant Green's function can be studied. These studies can be used to derive optimal conditions for a measurement technique.
Thermal waves are solutions of the heat diffusion equation for periodic boundary conditions and can be seen analogously to strongly damped waves. Although the underlying differential equation differs from the wave equation, the essential property for analogy between both equations is linearity such that superposition applies. This linearity is maintained even after a linear transformation, such as the Fourier transform from time to frequency domain. It follows that the temporal superposition principle is already used in active thermography, e.g. in pulsed thermography, as a superposition of many individual frequencies. However, the systematic spatial superposition has not yet been fully exploited, mainly due to a lack of suitable energy sources.
As a first step, we are investigating how thermal wave fields of arbitrary space-time structures can be engineered using structured laser illumination. The proof of principle was shown using a laser coupled projector. Unfortunately, the available optical output power was limited due to the thermal stress limit of the device. That is why we are working towards a more sophisticated moving 1D array of high-power diode lasers. We characterized the novel light source and believe that apart from the benefit of spatial and temporal illumination it can combine the temporal regimes of impulse and lock-in thermography.
In a second step, we investigate moving and oscillating line sources with different line shapes. We use a Green’s Function ansatz to analytically model the thermal wave propagation of structured 1D laser illumination in isotropic materials. Furthermore, we show some methods how they can be implemented. With this technique, we were able to accelerate our detection method firstly presented in for vertical narrow defects by factor three. Generally, we believe that this technique opens up similar opportunities than in other NDE methods. High-resolution ultrasound, for example, is also based on the superposition of single emitters and a recent concept suggests an option to deal with the diffusion wave character of the thermal waves.
The work to be presented focuses on our most recent studies to laser excited super resolution (SR) thermography. The goal of nondestructive testing with SR is to facilitate the separation of closely spaced defects. Photothermal SR can be realized by performing structured illumination measurements in com-bination with the use of deconvolution algorithms in post-processing. We explain that stepwise as well as continuous scanning techniques are applicable to generate structured illumination measurements. Finally, we discuss the effect of experimental parameters and image processing techniques to find the optimal SR technique which leads to the highest reconstruction quality within laser thermography.
Thermographic nondestructive evaluation (NDE) is based on the interaction of thermal waves with inhomogeneities. These inhomogeneities are related to sample geometry or material composition. Although thermography is suitable for a wide range of inhomogeneities and materials, the fundamental limitation is the diffusive nature of thermal waves and the need to measure their effect radiometrically at the sample surface only. The propagation of the thermal waves from the heat source to the inhomogeneity and to the detection surface results in a degradation in the spatial resolution of the technique. A new concerted ansatz based on a spatially structured heating and a joint sparsity of the signal ensemble allows an improved reconstruction of inhomogeneities. As a first step to establish an improved thermographic NDE method, an experimental setup was built based on structured 1D illumination using a flash lamp behind a mechanical aperture. As a follow-up to this approach, we now use direct structured illumination using a 1D laser array. The individual emitter cells are driven by a pseudo-random binary pattern and are additionally shifted by fractions of the cell period. The repeated measurement of these different configurations enables to illuminate each spot of the sample surface in lateral direction. This allows for a reconstruction that makes use of joint sparsity.
The measured data set is processed using super resolution image reconstruction algorithms such as the iterative joint sparsity (IJOSP) algorithm. Using this reconstruction technique and 150 different illumination patterns results in a spatial resolution enhancement of approximately a factor of four compared to the resolution of 5.9 mm for homogenously illuminated thermographic reconstruction.
Further, new data processing techniques have been studied before applying the IJOSP algorithm that are more performant or less prone to errors regarding image reconstruction. The choice of regularization parameters in data processing as well as experimental parameters such as the illumination pattern as a variable heat flux density (i.e., the Neumann boundary condition for convolution with the constant Green's function) have a big influence on the reconstruction goodness. With analytical-numerical modelling and numerical FEM simulations, we studied the influence of the experimental parameters on the result of the non-linear IJOSP reconstruction. This has also been investigated experimentally e.g. using different laser line widths or more measurements per position. These studies are used to derive optimal conditions for a certain measurement image reconstruction technique.