Deep learning approaches for thermographic imaging

Abstract

In this paper, we investigate two deep learning approaches to recovering initial temperature profiles from thermographic images in non-destructive material testing. First, we trained a deep neural network (DNN) in an end-to-end fashion by directly feeding the surface temperature measurements to the DNN. Second, we turned the surface temperature measurements into virtual waves (a recently developed concept in thermography), which we then fed to the DNN. To demonstrate the effectiveness of these methods, we implemented a data generator and created a dataset comprising a total of 100 000 simulated temperature measurement images. With the objective of determining a suitable baseline, we investigated several state-of-the-art model-based reconstruction methods, including Abel transformation, curvelet denoising, and time- and frequency-domain synthetic aperture focusing techniques. Additionally, a physical phantom was created to support evaluation on completely unseen real-world data. The results of several experiments suggest that both the end-to-end and the hybrid approach outperformed the baseline in terms of reconstruction accuracy. The end-to-end approach required the least amount of domain knowledge and was the most computationally efficient one. The hybrid approach required extensive domain knowledge and was more computationally expensive than the end-to-end approach. However, the virtual waves served as meaningful features that convert the complex task of the end-to-end reconstruction into a less demanding undertaking. This in turn yielded better reconstructions with the same number of training samples compared to the end-to-end approach. Additionally, it allowed more compact network architectures and use of prior knowledge, such as sparsity and non-negativity. The proposed method is suitable for non-destructive testing (NDT) in 2D where the amplitudes along the objects are considered to be constant (e.g., for metallic wires). To encourage the development of other deep-learning-based reconstruction techniques, we release both the synthetic and the real-world datasets along with the implementation of the deep learning methods to the research community.

Imaging approaches for monitoring three‐dimensional cell and tissue culture systems

“…However, some common limitations for PTI studies on tissues, especially quantitative biochemical PTI studies, include: the overlap of the water signal in the mid-IR region often resulting in low signal to noise values [189,215], and heat-induced photo-chemical reactions, which do not produce heat but rather may produce a new chemical species which may alter the photothermal properties of the sample [185]. The current ambitions of PTI imaging is focused towards overcoming these limitations, with studies focused towards heterodyne detection [216], digital holography and optical diffraction holography [17,191], VIPPS phase-sensitive lock-in detection Scheme [207,209,210] (using reconstruction methodologies from other fields such as high-order correlation reconstruction (where the thermal emitting processes are dominated by the thermal diffusion processes) from super-resolution microscopy [217], and machine learning [218,219]), and developing multimodal photothermal systems (such as epifluorescence using thermo-sensitive fluorescent probes [220], Raman [221,222], photoacoustic [223] and OCT [224,225]).…”

Pipeline In-Line Inspection Method, Instrumentation and Data Management

“…Therefore, this method is effective in achieving automatic defect inspection and identification. Two deep learning approaches for thermographic image reconstruction were studied by Kovács et al [ 168 ]. By comparing these with other methods, the hybrid deep learning approach has an outstanding performance.…”

Importance of Image Enhancement and CDF for Fault Assessment of Photovoltaic Module Using IR Thermal Image

“…In [32] the authors used a machine learning technique based on the extracted features of thermal images to classify defective and non-defective solar panels. However, the main limitation in the proposed approach was the insufficient number of available datasets, which affected the model's reliability.…”