14 July 2021 in IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control
Corrosion detection is a critical problem in many research areas. Guided wave tomography provides a powerful tool to estimate the remaining thickness of corroded structures. This article introduces an ultrasonic quantitative tomography method called fast inversion tomography (FIT) for corrosion mapping on plate-like structures. FIT consists of offline training and online inversion. The offline training stage uses the supervised descent method (SDM) to generate a series of average descent directions iteratively by minimizing the waveform misfit function between the fixed initial models and training examples. The minimization of the misfit function is equivalent to solving the linear least-squares problem. In the online inversion stage, we reconstruct the velocity map of testing examples using the learned descent directions in an iterative manner. Then, we convert the velocity map into the thickness map using the dispersive characteristics of a specific guided wave mode. The performance of this technique is evaluated using synthetic datasets which include both training and testing examples with different corrosion depths and shapes on an aluminum plate. We also compare the reconstruction accuracy and computation efficiency between FIT and time/frequency-domain full-waveform inversion. The results indicate that FIT exhibits great performance in the problem of quantitative corrosion imaging.
26 October 2022 in IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control
Noninvasive brain diagnosis is extremely important because of its efficiency, low cost and painless nature to the prediction of stroke, cerebral hemorrhage and other brain researches. At present, achieving full three-dimensional quantitative ultrasonic imaging of the human brain is a cutting-edge challenge due to complex structures of the human brain and strong scattering caused by the skulls. In this article, we achieved quantitative ultrasonic imaging of inside-brain anomalies with our proposed method, the decomposition descent learning based full waveform inversion (DDL-FWI). The proposed method adopts a linear residual decomposing technique to greatly alleviate the computation burden in fast inversion tomography (FIT) with enhanced convergence guaranteed by residual functions. Testing results in both simulation and laboratory experiments demonstrated that our method can achieve high quality quantitative imaging of brain soft tissues and skulls even starting from homogeneous water background in 2D, and this method is capable of reconstructing both complex brain tissues and clots in 2D and 3D cases using either clean or noisy signals, with a robust 3D clot resolution as small as 18 mm and 2D reconstruction speed in 11.20 seconds. Combined with advanced ultrasonic hardware, DDL-FWI can be easily trained and used for brain imaging efficiently that frees patients from harmful influences from traditional imaging techniques, e.g., ionization radiations from X-ray computed tomography.
16 August 2022 in IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control
As an established biometric authentication approach, fingerprint scanning has received considerable attention due to its high accuracy and reliability. In this article, the fingerprint reconstruction at any position is achieved in large physical domains, which monitors wavefield variations of plate-like structures within arrays through the ultrasonic guided wave. Accurate reconstruction and quantitative characterization of fingerprints are obtained using fast inversion tomography (FIT) based on the deep learning convolutional neural network (DLCNN). Parametric optimization is conducted to reveal submillimeter fingerprint minutiae, and a specific DLCNN model is proposed for the artifact removal in FIT reconstructions. The results prove that the FIT based on DLCNN restoration can significantly improve the imaging quality in terms of increased resolution, reduced reconstruction errors, and higher fingerprint matching confidence. The reconstruction also allows an exponential improvement in computational efficiency as a result of much-reduced sensor numbers. Several factors affecting the performance of the proposed reconstruction method are discussed at the end.
1 August 2021 in Phytochemistry