Pointwise Measurements Research Articles

Continuous mapping of deformation over a ship structure model based on pointwise measurements

The development of structural monitoring systems on-board vehicles aims to achieve a full-picture of the vehicle response under operating conditions. The use of different data sources is a key point for many SHM approaches which demand for extensive information and data fusion techniques to address the specific inverse problem in a robust way. Therefore, a procedure to transform pointwise measurements into a continuous representation of the monitored structure is presented in this paper. To achieve the desired monitoring goals, a numerical approach which combines experimental recorded data and modelled responses from a ship structural model is developed. Craig-Bampton truncation technique is applied on the FE vehicle model to reduce elastic dofs and obtain a computationally cost-effective model. The main advantage of Craig-Bampton modes is the capability to keep unchanged the information at the sensor nodes. A state observer is then built to evaluate the elastic deflection field and the loads over the structure in the points where no measurement is available. A natural second-order observer is exploited as it is more suited for the present application than the Kalman filter. To enhance the observer performances, an optimization process is carried out based on a learning phase. Following a previous paper about theoretical aspects of the virtual sensing approach on 1D structures. the developed techniques are here applied to the monitoring of the structural deflections and the ambient excitation relative to an elastically scaled model of a fast catamaran, tested in the towing-tank. Accelerations and strain-gage data are used to reconstruct the elastic deformations of the physical model metallic backbone scaling the reference ship behavior. Hydrodynamic load data relative to the demihull portions and the wetdeck are obtained as well.

Sparse Cholesky factorization for solving nonlinear PDEs via Gaussian processes

In recent years, there has been widespread adoption of machine learning-based approaches to automate the solving of partial differential equations (PDEs). Among these approaches, Gaussian processes (GPs) and kernel methods have garnered considerable interest due to their flexibility, robust theoretical guarantees, and close ties to traditional methods. They can transform the solving of general nonlinear PDEs into solving quadratic optimization problems with nonlinear, PDE-induced constraints. However, the complexity bottleneck lies in computing with dense kernel matrices obtained from pointwise evaluations of the covariance kernel, and its partial derivatives, a result of the PDE constraint and for which fast algorithms are scarce. The primary goal of this paper is to provide a near-linear complexity algorithm for working with such kernel matrices. We present a sparse Cholesky factorization algorithm for these matrices based on the near-sparsity of the Cholesky factor under a novel ordering of pointwise and derivative measurements. The near-sparsity is rigorously justified by directly connecting the factor to GP regression and exponential decay of basis functions in numerical homogenization. We then employ the Vecchia approximation of GPs, which is optimal in the Kullback-Leibler divergence, to compute the approximate factor. This enables us to compute ϵ \epsilon -approximate inverse Cholesky factors of the kernel matrices with complexity O ( N log d ⁡ ( N / ϵ ) ) O(N\log ^d(N/\epsilon )) in space and O ( N log 2 d ⁡ ( N / ϵ ) ) O(N\log ^{2d}(N/\epsilon )) in time. We integrate sparse Cholesky factorizations into optimization algorithms to obtain fast solvers of the nonlinear PDE. We numerically illustrate our algorithm’s near-linear space/time complexity for a broad class of nonlinear PDEs such as the nonlinear elliptic, Burgers, and Monge-Ampère equations. In summary, we provide a fast, scalable, and accurate method for solving general PDEs with GPs and kernel methods.

A robust computational framework for variational data assimilation of mean flows with sparse measurements corrupted by strong outliers

A variational framework for the reconstruction of time-averaged mean flows using a sparse set of observations with large magnitudes of noise (referred to as outliers), is presented. The observations constitute a set of point-wise measurements of the mean flow with outliers at certain measurement locations and are incorporated into a numerical simulation governed by the two-dimensional, incompressible Reynolds-averaged Navier-Stokes (RANS) equations with an unknown momentum forcing. This forcing, which corresponds to the divergence of the Reynolds stress tensor, is calculated from a direct-adjoint optimization procedure to reduce the deviation between the measured and estimated mean velocities. ℓ2, ℓ1, Huber, and hybrid loss functions are used to represent the discrepancy in the mean velocity field between the measurements and the predictions. A variety of algorithms are considered to solve the optimization problem with these loss functions and a performance comparison in terms of the quality and physical features of the recovered mean flow is presented. The Huber loss function performed best as it remained robust to strong outliers in the measurements with its ℓ1 contribution and also ensured the uniqueness of the optimal solution with its ℓ2 contribution. Huber loss functions restrict the effect of outliers at the local measurement locations, thereby not affecting the quality of the high-dimensional reconstructed mean flow field. The hybrid loss function, a modified form of the continuous Huber loss function, also recovered the mean flow with high accuracy. We demonstrate the performance of the data assimilation framework for the case of two-dimensional laminar flow around a circular cylinder at Re=100. We then extend the analysis to the case of two-dimensional laminar flow over a backward-facing step at Re=500, to further assess the efficacy and robustness of the data assimilation framework.

On the low-frequency flapping motion in flow separation

Transitional separating flow induced by a rectangular plate subjected to uniform incoming flow at Reynolds number (based on the incoming velocity and half plate height) 2000 is investigated using direct numerical simulation. The objective is to unveil the long-lasting mystery of low-frequency flapping motion (FM) in flow separation. At a fixed streamwise-vertical plane or from the perspective of previous experimental studies using pointwise or planar measurements, FM manifests as a low-frequency periodic switching between low and high velocities covering the entire separation bubble. The results indicate that in three-dimensional space, FM reflects an intricate evolution of streamwise elongated streaky structures under the influence of separated shear layer and mean flow reversal. The FM is an absolute instability, and is initiated through a lift-up mechanism boosted by mean flow deceleration near the crest of the separating streamline. At this particular location, the shear bends the vortex filament abruptly, so that one end is vertically struck into the first half of the separation bubble, whereas the other end is extended in the streamwise direction in the second half of the separation bubble. These two ends of vortex filament are mutually sustained and also stretched by the vertical acceleration and streamwise acceleration in the first and second halves of the separation bubble, respectively. This process periodically switches the low-velocity (or high-velocity) streaky structure to a high-velocity (or low-velocity) streaky structure encompassing the entire separation bubble, and thus flaps the separated shear layer up and down in the vertical direction. A ‘large vortex’ shedding manifests when the streaky structure switches signs. This large vortex is fundamentally different from the spanwise vortex shedding residing in the shear layer originated from the Kelvin–Helmholtz instability and successive vortex amalgamation. It is also believed that the three-dimensional evolution of streaky structures in the form of FM is applicable for both geometry- and pressure-induced separating flows.

Nongradient-Based Tracking of Environmental Field Isolines in a Changing Medium by Dubins-Vehicle Type Robots

A non-holonomic robot of a Dubins-vehicle type moves in a changing resistant medium, like water or air. The objective is to exhibit the extension of a spatial phenomenon (e.g., a critically contaminated area) related to a dynamic environmental field (e.g., a distribution of the contaminant concentration). Specifically, the robot should arrive at a certain level set of the field and repeatedly sweep it afterwards. The field data comes exclusively from online and point-wise measurements of the field value at the current location of the robot. An additional problem is related to the significant uncertainty on the field and the medium, including their variation with time and the effect of the medium on the robot's motion. The conditions for the accomplishability of the mission objective are found under these uncertainties and constraints on the kinematics of the robot. A navigation algorithm is presented that solves the mission with a slight enhancement of these necessary conditions. The performance of the algorithm is rigorously justified by the non-local convergence result and validated by computer simulation tests with a 6-DOF catamaran-like vessel and real-world environmental data.

Hysteretic quantised security control for PDE systems under replay attacks

ABSTRACT This article designs a quantised security controller for partial differential equation (PDE) systems under replay attacks. Initially, the Takagi – Sugeno fuzzy model is utilised to represent the considered nonlinear system and combines pointwise measurement and pointwise control methods to save the whole system’s costs. Moreover, to reduce network transmission pressure and simulate the network, hysteretic quantiser and replay attacks are considered. Furthermore, based on the Lyapunov direct method and the inequality technique, sufficient conditions to assure the stability of the closed-loop system are achieved. Ultimately, the effectiveness and superiority of the submitted control strategy are confirmed through numerical examples and comparative analysis.

Effects of elevated pressure on thermochemical states of turbulent flame-wall interaction studied by multi-parameter laser diagnostics

Effects of increased pressure on the thermochemistry of turbulent flame-wall interaction (FWI) were investigated within this study. Quantitative, pointwise measurements of three state parameters, i.e., gas-phase temperature, CO2 and CO mole fraction, were performed by means of dual-pump coherent anti-Stokes Raman spectroscopy (CARS) and two-photon laser-induced fluorescence (LIF) of CO. The flame front was simultaneously tracked by planar OH-LIF imaging. The measurements were carried out in a fully premixed methane-air flame in an enclosed side-wall quenching burner test rig under atmospheric and pressurized conditions (3 bar absolute pressure). An evaluation of near wall flame front topologies showed that the interaction of the flame with the wall is predominantly characterized by one single, extended reaction zone, similar to a so-called head-on quenching configuration. The dominance of this quenching scenario occurred most likely due to the confined burner configuration preventing an unhindered expansion of the hot exhaust gases. Thermochemical states were studied using pairwise correlations of temperature, CO2 and CO mole fraction. Regarding the measurements at atmospheric pressure, an overall good agreement with other studies on atmospheric FWI in a similar unconfined burner was observed considering differences concerning the flow field and the burner configuration. For the thermochemistry measurements at elevated pressure, similar trends as for the atmospheric measurements were identified. A comparison of both operating cases suggested that FWI processes were more strongly affected by heat losses at elevated pressure, but differences are moderately pronounced. To the authors’ best knowledge, these experiments are the first attempt to explore the near-wall thermochemistry of FWI processes at elevated pressure using multi-parameter measurements. As the near-wall measurements were further complicated by effects arising with increasing pressure, e.g., intensified beam steering and reduced length scales, implications on the application of the laser diagnostics are reported.

SUITABILITY ASSESSMENT OF DIFFERENT SENSORS TO DETECT HIDDEN INSTALLATIONS FOR AS-BUILT BIM

Abstract. Knowledge on the utilities hidden in the wall, e.g., electric lines or water pipes, is indispensable for work safety and valuable for planning. Since most of the existing building stock originates from the pre-digital era, no models as understood for Building Information Modeling (BIM) exist. To generate these models often labor-intensive procedures are necessary; however, recent research has dealt with the efficient generation and verification of a building’s electric network. In this context, a reliable measurement method is a necessity. In this paper we test different measurement techniques, such as point-wise measurements with hand-held devices or area-based techniques utilizing thermal imaging. For this purpose, we designed and built a simulation environment that allows various parameters to be manipulated under controlled conditions. In this scenario the low-cost handheld devices show promising results, with a precision between 92% and 100% and a recall between 89% and 100%. The expensive thermal imaging camera is also able to detect electric lines and pipes if there is enough power on the line or if the temperature of the water in the pipe and the environment’s temperature are sufficiently different. Nevertheless, while point-wise measurements can directly yield results, the thermal camera requires post-processing in specific analysis software. The results reinforce the idea of using reasoning methods in both the do-it-yourself and commercial sector, to rapidly gather information about hidden installations in a building without prior technical knowledge. This paves the way for, e.g., exploring the possibilities of an implementation and presentation in augmented reality (AR).

Characterized environmental influences on autocollimator measurement uncertainty using an extended Allan variance

Autocollimators play an important role in the high-precision testing of the slope errors of X-ray and large flat mirrors. However, the time-varying errors caused by environmental changes exhibit different characteristics over time, such as low-frequency drift errors and high-frequency random errors. Time-varying errors can seriously affect the accuracy of the autocollimator (AC) and the read repeatability. They have become important factors limiting the continued improvement of nano-optic-measuring machines and scanning pentaprism system accuracy. Considering the difficulty in quantitatively analyzing the influence of time-varying errors on the measurement uncertainties of the AC, this paper proposes the extended Allan variance to characterize environmental influences on the AC measurement uncertainty based on a point-wise scanning measurement mode. Taking advantage of the extended Allan variance's characteristics, which allow the identification of various errors and provide stability information about the AC measurement, the characteristic curves of the AC measurement uncertainty in a specific environment were obtained. Based on the characteristic curve parameters, the ultimate accuracy of AC in the current environment was obtained by optimizing the measurement strategy. Experimental results show that the measurement uncertainty of the AC can reach 2.3 nrad RMS in the measurement time-domain window.

Dynamic event–triggered security control for stochastic partial differential equation systems via a unified pointwise measurement/control strategy

This article focuses on the security control for parabolic partial differential equation systems driven by fractional Brownian motion based on a unified pointwise measurement/control strategy. Initially, by employing the Takagi–Sugeno fuzzy rule, nonlinear stochastic systems are transformed into a series of linear subsystems. Then, based on the improved dynamic event-triggered mechanism and pointwise measurements method, a security controller is presented to resist deception attacks. In addition, the Lyapunov direct method and some novel inequalities are used to ensure the considered system is mean-square exponentially stable, and the gain of the point controller is obtained. Finally, an application example is provided to verify the effectiveness of the proposed method.

How to estimate the 3D stress state for a deep geological repository

Abstract. When stresses yield a critical value, rock breaks or pre-existing faults are reactivated, generating pathways for fluid migration. Thus, the contemporary undisturbed stress state is a key parameter for assessing the stability of deep geological repositories and for quantifying whether stress changes induced by sub-surface constructions and storage of heat-generating waste lead to a critical state. The prediction of the contemporary 3D stress state is achieved with geomechanical–numerical models that are based on the static geological models. These are based on 3D seismic imaging and borehole data and describe the 3D structure and distribution of rock properties. Furthermore, pointwise measurements of the stress state are needed to fit the model to these data. The workshop is divided into three parts. Each part will have a short introduction to the specific topic, followed by dedicated hands-on exercise in small groups and discussions of the exercise outcome and its implications. 1. How do we formally describe the stress field? Here you will learn that the stress field is different from other physical field quantities such as temperature or displacement. The latter can be described with a scalar or a vector at a point. In contrast, the stress state at a point is described with the so-called Cauchy stress tensor that has nine components. We will give a quick tour through this concept, starting from very basic physics, and explain why the tensor is needed. In particular, you will learn in the exercises that stress cannot be measured directly with most methods but that strain is observed, from which stress is inferred.2. How do we actually measure stress? We present a range of methods that determine individual components of the stress state. In the exercise you will analyse borehole image logs to identify borehole breakouts and drilling-induced tensile fractures at the borehole wall. These are used to derive the orientation of the minimum and maximum horizontal stresses. Furthermore, we will present methods for the estimation of stress magnitudes. These are essential data for the calibration of a 3D geomechanical model. We will give a brief overview of different methods and in the exercise on the analysis of mini-hydraulic fracture tests as well as on the sleeve-fracturing and sleeve re-opening methods.3. From a viewpoint to a 3D description. If rock properties in the sub-surface were homogenous and isotropic and were to follow a simple linear behaviour between stress and strain, we could estimate the stress state analytically. However, this is not the case, and we will discuss three questions. a. How many pointwise stress data are needed to efficiently constrain a 3D model?b. How do I know whether a data point taken from less than a cubic metre is a good representation of a large rock volume, i.e. a whole lithological layer?c. How can we quantify the uncertainties of the model stress state, and what is the typical range? You will use a case study in northern Switzerland to explore these questions.

Improved Dynamic Event-Triggered Security Control for T–S Fuzzy LPV-PDE Systems via Pointwise Measurements and Point Control

Improved Dynamic Event-Triggered Security Control for T–S Fuzzy LPV-PDE Systems via Pointwise Measurements and Point Control

A novel geometry-based analysis of hippocampal morphometry in mesial temporal lobe epilepsy.

Hippocampal volumetry is an essential tool in researching and diagnosing mesial temporal lobe epilepsy (mTLE). However, it has a limited ability to detect subtle alterations in hippocampal morphometry. Here, we establish and apply a novel geometry-based tool that enables point-wise morphometric analysis based on an intrinsic coordinate system of the hippocampus. We hypothesized that this point-wise analysis uncovers structural alterations not measurable by volumetry, but associated with histological underpinnings and the neuropsychological profile of mTLE. We conducted a retrospective study in 204 individuals with mTLE and 57 age- and gender-matched healthy subjects. FreeSurfer-based segmentations of hippocampal subfields in 3T-MRI were subjected to a geometry-based analysis that resulted in a coordinate system of the hippocampal mid-surface and allowed for point-wise measurements of hippocampal thickness and other features. Using point-wise analysis, we found significantly lower thickness and higher FLAIR signal intensity in the entire affected hippocampus of individuals with hippocampal sclerosis (HS-mTLE). In the contralateral hippocampus of HS-mTLE and the affected hippocampus of MRI-negative mTLE, we observed significantly lower thickness in the presubiculum. Impaired verbal memory was associated with lower thickness in the left presubiculum. In HS-mTLE histological subtype 3, we observed higher curvature than in subtypes 1 and 2 (all p < .05). These findings could not be observed using conventional volumetry (Bonferroni-corrected p < .05). We show that point-wise measures of hippocampal morphometry can uncover structural alterations not measurable by volumetry while also reflecting histological underpinnings and verbal memory. This substantiates the prospect of their clinical application.

Velocimeter Light-Detection-and-Ranging–Informed Terrain Relative Navigation

Velocimeter light-detection-and-ranging (LIDAR)–informed batch and sequential estimation approaches for terrain relative navigation applications are developed in this paper. State-of-the-art velocimeter LIDAR sensors are capable of delivering simultaneous three-dimensional relative position, relative Doppler velocity, and reflectivity measurements for every point in the field of view. The proposed batch estimation algorithm yields accurate and statistically consistent vehicle velocity estimates in unknown (uncooperative) environments. This work presents novel pointwise relative position and Doppler velocity measurement models that can be utilized in the presence of known (cooperative) environments. Rate estimates obtained by the batch estimator are fused with the pointwise measurement models and classical inertial measurement unit formulations in the form of a multiplicative extended Kalman filter. This Doppler LIDAR-based sensor fusion approach yields informative terrain relative navigation solutions suitable for both cooperative and uncooperative environments. Results from extensive emulation robotics testing performed at NASA Johnson Space Center and Texas A&M’s Land, Air, and Space Robotics laboratory are presented to validate the proposed methodologies.

Adaptive sampling for accelerating neutron diffraction-based strain mapping * *This manuscript has been authored by UT-Battelle, LLC, under Contract No. DE-AC05-00OR22725 with the U.S. Department of Energy. The United States Government and the publisher, by accepting the article for publication, acknowledges that the United States Government retains a non-exclusive,paid-up, irrevocable, world-wide license to publish or reproduce the published form of

Neutron diffraction is a useful technique for mapping residual strains in dense metal objects. The technique works by placing an object in the path of a neutron beam, measuring the diffracted signals and inferring the local lattice strain values from the measurement. In order to map the strains across the entire object, the object is stepped one position at a time in the path of the neutron beam, typically in raster order, and at each position a strain value is estimated. Typical dwell times at neutron diffraction instruments result in an overall measurement that can take several hours to map an object that is several tens of centimeters in each dimension at a resolution of a few millimeters, during which the end users do not have an estimate of the global strain features and are at risk of incomplete information in case of instruments outages. In this paper, we propose an object adaptive sampling strategy to measure the significant points first. We start with a small initial uniform set of measurement points across the object to be mapped, compute the strain in those positions and use a machine learning technique to predict the next position to measure in the object. Specifically, we use a Bayesian optimization based on a Gaussian process regression method to infer the underlying strain field from a sparse set of measurements and predict the next most informative positions to measure based on estimates of the mean and variance in the strain fields estimated from the previously measured points. We demonstrate our real-time measure-infer-predict workflow on additively manufactured steel parts—demonstrating that we can get an accurate strain estimate even with 30%–40% of the typical number of measurements—leading the path to faster strain mapping with useful real-time feedback. We emphasize that the proposed method is general and can be used for fast mapping of other material properties such as phase fractions from time-consuming point-wise neutron measurements.

A Gaussian process guided super resolution sampling strategy for the efficient recovery of complex surfaces

Accurate recovery of complex surfaces of manufactured artefacts frequently requires intensive sampling, resulting in inefficient measurements for some point-by-point probe instruments. To tackle this problem, we fully exploit Gaussian process (GP) to guide the super resolution (SR) model to perform efficient and accurate sampling. The model makes use of a kernel-based GP method to model these low-frequency geometric features, while a pretrained SR method with multiple residual attention blocks is used to focus on the high-frequency features and further improve the details of the surface. In addition to geometric errors and distance information, global uncertainty from the statistical properties of the GP and an additional feature error from the SR are combined as critical criteria to select the most informative points of the surface. The effectiveness of the proposed method was demonstrated through several experiments on synthetic and real-world data, showing that the proposed method achieves state-of-the-art performance for pointwise measurements.

Development of a retrofit layer with an embedded array of piezoelectric sensors for transient pressure measurement in maritime applications

Measurement of transient pressure distribution on maritime structures is important for the assessment of the hydrodynamic loads applied. The commonly used pressure sensors are mostly bulky, need to be bolted to the structure, and/or only provide point-wise measurements. In this paper, an elastic matrix layer with a network of embedded piezoelectric sensors is proposed to address these issues. For experimental validation, a 400 × 400 × 5 mm epoxy layer is fabricated embedding 25 piezoelectric sensors on a square grid in accordance with Gauss-Lobatto-Legendre points. A finite element based inverse procedure is developed to reconstruct the pressure field from the electric potentials measured by the piezoelectric transducers. Feasibility of the concept is evaluated by measuring and reconstructing the pressure field generated by a travelling wave in a water tank. Sensitivity of the layer is also investigated through the experiments. The results indicate that the retrofit layer is capable of pressure field reconstruction, and that the presence of disturbances on the sensing surface does not affect the measurements in a notable way, while non-ideal conditions of the mounting can have a significant impact on the accuracy of the measurements. The results highlight the potential of the concept in pressure distribution measurements.

Three-dimensional full-field velocity measurements in shock compression experiments using stereo digital image correlation.

Shock compression plate impact experiments conventionally rely on point-wise velocimetry measurements based on laser-based interferometric techniques. This study presents an experimental methodology to measure the free surface full-field particle velocity in shock compression experiments using high-speed imaging and three-dimensional (3D) digital image correlation (DIC). The experimental setup has a temporal resolution of 100ns with a spatial resolution varying from 90 to 200μm/pixel. Experiments were conducted under three different plate impact configurations to measure spatially resolved free surface velocity and validate the experimental technique. First, a normal impact experiment was conducted on polycarbonate to measure the macroscopic full-field normal free surface velocity. Second, an isentropic compression experiment on Y-cut quartz-tungsten carbide assembly is performed to measure the particle velocity for experiments involving ramp compression waves. To explore the capability of the technique in multiaxial loading conditions, a pressure shear plate impact experiment was conducted to measure both the normal and transverse free surface velocities under combined normal and shear loading. The velocities measured in the experiments using digital image correlation are validated against previous data obtained from laser interferometry. Numerical simulations were also performed using established material models to compare and validate the experimental velocity profiles for these different impact configurations. The novel ability of the employed experimental setup to measure full-field free surface velocities with high spatial resolutions in shock compression experiments is demonstrated for the first time in this work.

Validation of a Hyperspectral Imaging System for Color Measurement of In-Vivo Dental Structures.

A full comprehension of colorimetric relationships within and between teeth is key for aesthetic success of a dental restoration. In this sense, hyperspectral imaging can provide point-wise reliable measurements of the tooth surface, which can serve for this purpose. The aim of this study was to use a hyperspectral imaging system for the colorimetric characterization of 4 in-vivo maxillary anterior teeth and to cross-check the results with similar studies carried out with other measuring systems in order to validate the proposed capturing protocol. Hyperspectral reflectance images (Specim IQ), of the upper central (UCI) and lateral incisors (ULI), were captured on 30 participants. CIE-L*a*b* values were calculated for the incisal (I), middle (M) and cervical (C) third of each target tooth. ΔEab* and ΔE00 total color differences were computed between different tooth areas and adjacent teeth, and evaluated according to the perceptibility (PT) and acceptability (AT) thresholds for dentistry. Non-perceptible color differences were found between UCIs and ULIs. Mean color differences between UCI and ULI exceeded AT (ΔEab* = 7.39-7.42; ΔE00 = 5.71-5.74) in all cases. Large chromatic variations between I, M and C areas of the same tooth were registered (ΔEab* = 5.01-6.07 and ΔE00 = 4.07-5.03; ΔEab* = 5.80-8.16 and ΔE00 = 4.37-5.15; and ΔEab* = 5.42-5.92 and ΔE00 = 3.87-4.16 between C and M, C and I and M and I, respectively). The use of a hyperspectral camera has proven to be a reliable and effective method for color evaluation of in-vivo natural teeth.

Deep learning approach for delamination identification using animation of Lamb waves

The complex design of composite structures makes it very difficult for conventional visual inspection techniques to detect defects in these materials. Therefore, numerous types of nondestructive testing and structural health monitoring techniques have been developed for damage identification in composite structures, among which ultrasonic guided waves, particularly Lamb waves, have gained much popularity. In recent years, apart from piezoelectric pointwise measurements, laser Doppler vibrometry has been used for full wavefield measurements of propagating Lamb waves. Damage imaging can be done with animations of Lamb waves interacting with defects. In this work, two end-to-end deep learning-based models of many-to-one sequence prediction were developed to perform pixel-wise image segmentation. A large dataset of full wavefield images resulting from the interaction with delamination of random size, shape, and location was utilised to train the proposed deep learning models. The main goal of this work is to investigate the feasibility of implementing deep learning methods for delamination identification in composite laminates by only utilising the animations of Lamb waves. The elaborated models worked well on the numerically generated test data and have also shown good performance on the real-world experimental data. These methods can enable the automation of delamination identification and to create damage maps without the intervention of the user.