A State-of-the-Art Review on FRF-Based Structural Damage Detection: Development in Last Two Decades and Way Forward

Vibration-based damage detection techniques receive wide attention of the research community in recent years to overcome the limitations of conventional structural health monitoring methods. The modal parameters, namely, natural frequencies, mode shapes, transmissibility, frequency response function (FRF), and other damage sensitive features are usually employed to identify damage in a structure. The main objective of this review is to generate a detailed understanding of FRF-based techniques and to study their performance in terms of advantage, accuracy, and limitations in structural damage detection. This paper also reviews various approaches to develop methodologies in terms of efficiency and computational time. The study observed that excitation frequency, location of application of excitation, type of sensor, number of measurement locations, noise contamination in FRF data, selection of frequency range for simulation, weighting and numerical techniques to solve the over-determined set of equations influence the effectiveness of damage identification procedure. Limitations and future prospects have also been addressed in this paper. The content of this paper aims to guide researchers in developing formulations, updating models, and improving results in the field of FRF-based damage identification.

Advanced Bioelectrical Signal Processing Methods: Past, Present and Future Approach—Part II: Brain Signals

As it was mentioned in the previous part of this work (Part I)—the advanced signal processing methods are one of the quickest and the most dynamically developing scientific areas of biomedical engineering with their increasing usage in current clinical practice. In this paper, which is a Part II work—various innovative methods for the analysis of brain bioelectrical signals were presented and compared. It also describes both classical and advanced approaches for noise contamination removal such as among the others digital adaptive and non-adaptive filtering, signal decomposition methods based on blind source separation, and wavelet transform.

Aircraft interior acoustics - background noise contamination

Aircraft interior noise is an important factor to be considered for cabin comfort. In a cruising condition this noise source is mostly broadband in nature and is coming from the exterior, primarily the turbulent boundary layer (TBL) of the flow around the moving aircraft. Capturing this noise to a high frequency is critical for designing the sound packaging. Also, this becomes important in the design of public announcement (PA) system for the aircraft cabin, i.e. the correct placement of speakers. One of the metrics used for this acoustic design is speech transmission index. Deterministic techniques like finite or boundary element techniques for low frequencies and ray tracing method to reach higher frequencies are better suited for getting the narrow band responses. On the other hand, to characterize the background noise due to the TBL loads, statistical energy analysis (SEA) route is pursued. In this paper the authors combine different techniques to capture the background noise and use them with PA sources and eventually capture the sound perceived at points of interest. The articulation metrics are compared for different operating conditions of the aircraft. In the presentation attempts will be made to play the auralized sounds.

Performance evaluation on multi-channel Wiener filter based speech enhancement for unmanned aerial vehicles recordings

Recording speech from unmanned aerial vehicles has been attracting interest due to its broad application including filming, search and rescue, and surveillance. One of the challenges in this problem is the quality of the speech recorded due to contamination by various interfering noise. In particular, noise contamination due to those radiated by the unmanned aerial vehicles rotors significantly impacts the overall quality of the audio recordings. Multi-channel Wiener filter has been a commonly used technique for speech enhancement because of its robustness under practical setup. Existing studies have also utilised such techniques in speech enhancement for unmanned aerial vehicle recordings, such as the well-known beamformer with postfiltering framework. However, many variants of the multi-channel Wiener filter have also been developed over recent years such as the speech distortion weighted multi-channel Wiener filter. To address these recent advancements, in this study we compare the performance of these variants of techniques. In particular, we explore the benefits these techniques may bring forth in the setting of audio recordings from an unmanned aerial vehicle.

Advanced Bioelectrical Signal Processing Methods: Past, Present and Future Approach—Part I: Cardiac Signals

Advanced signal processing methods are one of the fastest developing scientific and technical areas of biomedical engineering with increasing usage in current clinical practice. This paper presents an extensive literature review of the methods for the digital signal processing of cardiac bioelectrical signals that are commonly applied in today’s clinical practice. This work covers the definition of bioelectrical signals. It also covers to the extreme extent of classical and advanced approaches to the alleviation of noise contamination such as digital adaptive and non-adaptive filtering, signal decomposition methods based on blind source separation and wavelet transform.

High clarity speech separation using synchro extracting transform

Degenerate unmixing estimation technique (DUET) is the most ideal blind source separation (BSS) method for underdetermined conditions with number of sources exceeds number of mixtures. Estimation of mixing parameters which is the most critical step in the DUET algorithm, is developed based on the characteristic feature of sparseness of speech signals in time frequency (TF) domain. Hence, DUET relies on the clarity of time frequency representation (TFR) and even the slightest interference in the TF plane will be detrimental to the unmixing performance. In conventional DUET algorithm, short time Fourier transform (STFT) is utilized for extracting the TFR of speech signals. However, STFT can provide on limited sharpness to the TFR due to its inherent conceptual limitations, which worsens under noise contamination. This paper presents the application of post-processing techniques like synchro squeezed transform (SST) and synchro extracting transform (SET) to the DUET algorithm, to improve the TF resolution. The performance enhancement is evaluated both qualitatively and quantitatively by visual inspection, Renyi entropy of TFR and objective measures of speech signals. The results show enhancement in TF resolution and high clarity signal reconstruction. The method also provides adequate robustness to noise contamination.

Image Denoising Using Adaptive and Overlapped Average Filtering and Mixed-Pooling Attention Refinement Networks

Cameras are essential parts of portable devices, such as smartphones and tablets. Most people have a smartphone and can take pictures anywhere and anytime to record their lives. However, these pictures captured by cameras may suffer from noise contamination, causing issues for subsequent image analysis, such as image recognition, object tracking, and classification of an object in the image. This paper develops an effective combinational denoising framework based on the proposed Adaptive and Overlapped Average Filtering (AOAF) and Mixed-pooling Attention Refinement Networks (MARNs). First, we apply AOAF to the noisy input image to obtain a preliminarily denoised result, where noisy pixels are removed and recovered. Next, MARNs take the preliminary result as the input and output a refined image where details and edges are better reconstructed. The experimental results demonstrate that our method performs favorably against state-of-the-art denoising methods.

Towards On-Board Hyperspectral Satellite Image Segmentation: Understanding Robustness of Deep Learning through Simulating Acquisition Conditions

Although hyperspectral images capture very detailed information about the scanned objects, their efficient analysis, transfer, and storage are still important practical challenges due to their large volume. Classifying and segmenting such imagery are the pivotal steps in virtually all applications, hence developing new techniques for these tasks is a vital research area. Here, deep learning has established the current state of the art. However, deploying large-capacity deep models on-board an Earth observation satellite poses additional technological challenges concerned with their memory footprints, energy consumption requirements, and robustness against varying-quality image data, with the last problem being under-researched. In this paper, we tackle this issue, and propose a set of simulation scenarios that reflect a range of atmospheric conditions and noise contamination that may ultimately happen on-board an imaging satellite. We verify their impact on the generalization capabilities of spectral and spectral-spatial convolutional neural networks for hyperspectral image segmentation. Our experimental analysis, coupled with various visualizations, sheds more light on the robustness of the deep models and indicate that specific noise distributions can significantly deteriorate their performance. Additionally, we show that simulating atmospheric conditions is key to obtaining the learners that generalize well over image data acquired in different imaging settings.

Energetic characteristics of High Frequency (HF) and Very High Frequency (VF) Martian events

<p>The InSight seismometer SEIS recorded tens of high-frequency (1.5-5Hz; HF) and Very-high frequency (1.5-15Hz, VF) Martian events. They are characterized by two temporally separated arrivals with a gradual beginning, a broad maximum and a very long decay. This observation is consistent with a long-range propagation of seismic P and S waves in a heterogeneous crust (Van Driel et al., accepted). To examine this hypothesis, first, we employ basic multiple-scattering concepts on the two groups of events. Then, we propose a full envelope modeling based on elastic radiative transport in a half-space. The model parametrization and the radiative transfer equations are presented in (Lognonné, P., et al. (2020) and Margerin, L., (2017)). We find that both HF and VF signals are depolarized and verify Gaussian statistics, at the exception of the ballistic primary and secondary arrivals. These properties agree with a multiple-scattering origin. For VF events, the energy partitioning ratio V<sup>2</sup>/H<sup>2</sup> between horizontal and vertical components is frequency dependent. We observe that V<sup>2</sup>/H<sup>2 </sup>is maximum at the so-called ‘2.4Hz resonance’ (~2) and decreases rapidly at frequencies higher than 5Hz (~0.1) then i remains relatively low up to frequencies of 15Hz at least. HF events do not exhibit a decrease of V<sup>2</sup>/H<sup>2 </sup>at high frequencies however further analysis reveals a strong correlation between energy partitioning and signal-to-noise (S/N) ratio for HF events. This observation suggests that a part of the difference between the HF and VF events can to some extent be explained by noise contamination. The generally low V<sup>2</sup>/H<sup>2 </sup>ratio of VF events is reminiscent of the response of unconsolidated layers, as observed at Pinyon Flats Observatory on Earth (Margerin, L., et al. (2009)). Unlike earthquakes and moonquakes observed in the same frequency band, the delay time measured from onset to peak of the secondary arrival of HF and VF events is frequency-independent. This suggests that the spectrum of heterogeneity of the Martian crust is smooth. We observe that, for HF and VF events, the delay time is weakly dependent on hypocentral distance. This observation cannot be reconciled with the predictions of multiple-scattering theories in a statistically homogeneous medium however it suggests a stratification of heterogeneity in the Martian lithosphere. The coda quality factor Q<sub>c</sub> of VF events is high and shows a linear increase with frequency. Q<sub>c</sub> of HF events is higher but it may be overestimated due to the noise contamination. The linear frequency dependence of Q<sub>c</sub> is strongly reminiscent of the leakage effect in a crustal scattering waveguide and suggests that part of the observed coda attenuation may be of structural origin. The full envelope modeling of the S0334a VF event results shows that the estimated value of the diffusivity (≃ 619 km<sup>2</sup>/s) is almost 6 times greater than for the S0128a VF event (≃ 90 km<sup>2</sup>/s). This observation again suggests a stratification of heterogeneity. In future works, we will perform the full envelope modeling of all the VF selected events at different frequencies to constrain a 1D attenuation and diffusion model of the Martian crust.</p>

Influence of the Distortion of Vertical Wavenumber Spectra on Estimates of Turbulent Dissipation using the Finescale Parameterization: Eikonal Calculations

The finescale parameterization, formulated on the basis of a weak nonlinear wave–wave interaction theory, is widely used to estimate the turbulent dissipation rate, ε. However, this parameterization has previously been found to overestimate ε in the Antarctic Circumpolar Current (ACC) region. One possible reason for this overestimation is that vertical wavenumber spectra of internal wave energy are distorted from the canonical Garrett-Munk spectrum and have a spectral “hump” at low vertical wavenumbers. Such distorted vertical wavenumber spectra were also observed in other mesoscale eddy-rich regions. In this study, using eikonal simulations, in which internal wave energy cascades are evaluated in the frequency-wavenumber space, we examine how the distortion of vertical wavenumber spectra impacts on the accuracy of the finescale parameterization. It is shown that the finescale parameterization overestimates ε for distorted spectra with a low-vertical-wavenumber hump because it incorrectly takes into account the breaking of these low-vertical-wavenumber internal waves. This issue is exacerbated by estimating internal wave energy spectral levels from the low-wavenumber band rather than from the high-wavenumber band, which is often contaminated by noise in observations. Thus, in order to accurately estimate the distribution of ε in eddy-rich regions like the ACC, high-vertical-wavenumber spectral information free from noise contamination is indispensable.