Vertical Images Research Articles

Three-dimensional histological explanation of the dermoscopy patterns in acral melanocytic lesions.

Dermoscopic images of pigmented lesions have distinct features on the sole where skin ridges and furrows exist. Pigmentation of benign nevus usually locates on the skin furrow, while the malignant melanoma is pigmented on the skin ridge. Correspondence between dermoscopy and pathology in the pigmented lesions on soles have been studied based on conventional vertical pathological images. However, for the full understanding of the correspondence, observation of horizontal histological images would be required, because the epidermis constructs unique horizontal structures, namely crista profunda limitans, crista profunda intermedia, and transverse ridge. In this study, we analyzed basic dermoscopic images of the representative acral melanocytic lesions (nevus, lentigo, and malignant melanoma) by horizonal histological images. We created serial horizontal pathological images by digital reconstruction of a hundred of serial vertical images. We could show that parallel furrow pattern is created by the pigmentation of crista profunda limitans, parallel ridge pattern by the pigmentation of both of crista profunda limitans and crista profunda intermediate, and lattice-like pattern by the pigmentation of transverse ridge. Our results would be useful for the intuitive histological understanding of dermoscopy.

A robust method of rectifying a tilted image to truly vertical

Unmanned Aerial Vehicle (UAV) is burgeoning technology for conducting aerial photography ranging from recreational purposes to engineering works. Due to its physical design, a mounted camera onto the UAV body is afflicted by engine and other vibrations during flight data acquisitions. In such condition, it is impossible to maintain the camera optical axis truly vertical while it is directed downward to the ground. Unavoidable aircraft tilts causes captured images to be exposed with the camera axis tilted slightly from vertical. In survey and mapping activity when vertical images are intended, the amount by which the optical axis deviates from vertical of more than 3 degrees is not met the work specification. Truly vertical images are necessary for producing high quality map from aerial images. This paper offers robust solution to rectify tilted images to generate equivalent truly vertical ones. We use exterior orientation parameters of the tilted images to perform double steps of transformation while resampling grid tessellations of the vertical images. To visualize the result, the transformation processes are coded by C++ programing language. The program results show a feasibility of our approach.

Photogrammetry Using UAV-Mounted GNSS RTK: Georeferencing Strategies without GCPs

Georeferencing using ground control points (GCPs) is the most common strategy in photogrammetry modeling using unmanned aerial vehicle (UAV)-acquired imagery. With the increased availability of UAVs with onboard global navigation satellite system–real-time kinematic (GNSS RTK), georeferencing without GCPs is becoming a promising alternative. However, systematic elevation error remains a problem with this technique. We aimed to analyze the reasons for this systematic error and propose strategies for its elimination. Multiple flights differing in the flight altitude and image acquisition axis were performed at two real-world sites. A flight height of 100 m with a vertical (nadiral) image acquisition axis was considered primary, supplemented with flight altitudes of 75 m and 125 m with a vertical image acquisition axis and two flights at 100 m with oblique image acquisition axes (30° and 15°). Each of these flights was performed twice to produce a full double grid. Models were reconstructed from individual flights and their combinations. The elevation error from individual flights or even combinations yielded systematic elevation errors of up to several decimeters. This error was linearly dependent on the deviation of the focal length from the reference value. A combination of two flights at the same altitude (with nadiral and oblique image acquisition) was capable of reducing the systematic elevation error to less than 0.03 m. This study is the first to demonstrate the linear dependence between the systematic elevation error of the models based only on the onboard GNSS RTK data and the deviation in the determined internal orientation parameters (focal length). In addition, we have shown that a combination of two flights with different image acquisition axes can eliminate this systematic error even in real-world conditions and that georeferencing without GCPs is, therefore, a feasible alternative to the use of GCPs.

No-reference light field image quality assessment based on depth, structural and angular information

Light field imaging can capture more abundant scene information, including angular and spatial information, compared with traditional imaging technologies. However, light field image (LFI) processing will inevitably introduce distortion which is different from traditional imaging process. Therefore, LFI quality assessment has become one of important research issues on LFI processing. In this paper, we propose a depth, structural and angular information based no-reference LFI quality assessment method. Firstly, the horizontal and vertical mean difference images (MDIs) are defined to integrate the difference images of LFI's sub-aperture images (SAIs) and reflect the horizontal and vertical depth and structural information of LFI. Considering the multi-channel characteristics of human visual system, an effective feature extraction scheme with Curvelet decomposition is designed for MDIs and SAIs of the distorted LFI. As one of the representations of LFI, epipolar plane images (EPIs) contain LFI's angular and depth information, which reflect the angular consistency of LFI more intuitively. Therefore, an algorithm, namely local maximum similarity index statistics, is designed to extract the directional features from horizontal and vertical EPIs. Finally, the quality of distorted LFI is predicted by pooling these extracted features. Experimental results on three LFI datasets show that the proposed method can achieve better performance, compared with the representative traditional image quality assessment methods as well as the state-of-the-art LFI quality assessment methods.

Macular irregularities of optical coherence tomographic vertical cross sectional images in school age children

The purpose of this study was to compare the incidences of macular irregularities of elementary school (ES) and junior high school (JHS) students. This was a prospective cross-sectional observational study of 122 right eyes of 122 ES students (8–9 years) and 173 right eyes of 173 JHS students (12–13 years). Vertical cross-sectional images of the macula were obtained by optical coherence tomography. The eyes were classified based on the vertical symmetry of the posterior pole, and then sub-classified as convex-, flat-, concave-, or dome-shaped based on the direction of the curvature of the retinal pigment epithelium. One hundred and two eyes (83.6%) were placed in the symmetrical group in the ES students and 149 eyes (86.1%) in the JHS students. Twenty eyes (16.4%) were placed in the asymmetric groups in the ES students and 24 eyes (13.9%) in the JHS students. In symmetrical group, 78 and 118 eyes had a convex-shape (76.4 and 79.2%), followed by 22 and 29 eyes of dome-shaped group (21.6 and 19.4%) in ES and JHS students respectively. Because the incidences of the posterior pole shapes were not significantly different between the groups, it is likely that the macular irregularities develop before the age of ES.

The microglia response to electrical overstimulation of the retina imaged under a transparent stimulus electrode

Objective. We investigated using the morphological response of retinal microglia as indicators of tissue damage from electrical overstimulation by imaging them through an optically transparent stimulus electrode. Approach. To track the microglia, we used a transgenic mouse where the microglia expressed a water soluble green fluorescent protein. The clear stimulus electrode was placed epiretinally on the inner limiting membrane and the microglia layers were imaged using time-lapse confocal microscopy. We examined how the microglia responded both temporally and spatially to local overstimulation of the retinal tissue. Using confocal microscope vertical image stacks, the microglia under the electrode were imaged at 2.5 min intervals. The retina was overstimulated for a 5 min period using 1 ms 749 μC cm−2 ph−1 biphasic current pulses and changes in the microglia morphology were followed for 1 h post stimulation. After the imaging period, a label for cellular damage was applied to the retina. Main results. The microglia response to overstimulation depended on their spatial location relative to the electrode lumen and could result in three different morphological responses. Some microglia were severely injured and became a series of immotile ball-like fluorescent processes. Other microglia survived, and reacted rapidly to the injury by extending filopodia oriented toward the damage zone. This response was seen in inner retinal microglia outside the stimulus electrode edge. A third effect, seen with the deeper outer microglia under the electrode, was a fading of their fluorescent image which appeared to be due to optical scatter caused by overstimulation-induced retinal edema. Significance. The microglial morphological responses to electrical overstimulation injury occur rapidly and can show both direct and indirect effects of the stimulus electrode injury. The microglia injury pattern closely follows models of the electric field distribution under thinly insulated disc electrodes.

Development of Motion Artifact Generator for Deep Learning in Brain MRI

We focused on deep learning for a reduction of motion artifacts in MRI. It is difficult to collect a large number of images with and without motion artifacts from clinical images. The purpose of this study was to create motion artifact images in MRI by simulation. We created motion artifact images by computer simulation. First, 20 different types of vertical pixel-shifted images were created with different shifts, and the amount of pixel shift was set from -10 to 10 pixels. The same method was used to create pixel-shifted images for horizontal shift, diagonal shift, and rotational shift, and a total of 80 types of pixel-shifted images were prepared. These images were Fourier transformed to create 80 types of k-space data. Then, phase encodings in these k-space data were randomly sampled and Fourier transformed to create artifact images. The reproducibility of the simulation images was verified using the deep learning network model of U-net. In this study, the evaluation indices used were the structural similarity index measure (SSIM) and peak signal-to-noise ratio (PSNR). The average SSIM and PSNR for the simulation images were 0.95 and 31.5, respectively; those for the clinical images were 0.96 and 31.1, respectively. Our simulation method enables us to create a large number of artifact images in a short time, equivalent to clinical artifact images.

An Epipolar Resampling Method for Multi-View High Resolution Satellite Images Based on Block

As the basis of stereo observation, epipolar resampling is a critical technology for 3D reconstruction. With the development of the 3D reconstruction using multi-view satellite images, the epipolar resampling of the multi-view satellite images has become a new challenging problem. In this paper, we propose an epipolar resampling method for multi-view high resolution satellite images based on block. Firstly, we establish the relationship between the vertical parallax and image block size for the epipolar resampling based on block. Using this relationship, we can choose the block size to limit the maximum vertical parallax in epipolar resampling. Secondly, we use the rational function model (RFM) to generate virtual corresponding points for fundamental matrix estimation, so image pairs with large stereo angles and poor consistency in multi-view satellite images which are difficult to extract feature matching point can be rectified. Experiments with multi-view satellite images demonstrated the universality and effectiveness of the proposed method. All the image pairs in the multi-view satellite image dataset were rectified successfully. The average epipolar resampling error is only 0.25 pixels when we rectify the multi-view satellite images acquired by WorldView-3 with the block size of 3000*3000 pixels, significantly less than the similar method.

The Integration of Planetary Science Simulation and Virtual Reality Geographic Information System

A method for constructing a measurable virtual reality environment based on the ground image of the planetary exploration patrol device and the high-resolution satellite image is proposed. According to the three-dimensional images acquired by the patrol at different azimuth angles and height angles, a 360° panoramic image is generated through cylindrical projection, automatic matching and seamless stitching. At the same time, the inverse calculation function from the panoramic image to the original image has been developed, allowing users to directly measure on the panoramic image, and obtain 3D terrain information from the original stereo pair. The three-dimensional measurable panoramic image is seamlessly integrated with the planet browser based on NASA's World Wind, which provides a measurable virtual reality environment for planetary exploration applications, and realizes the integration from the vertical view satellite image to the ground horizontal view patroller imagemeasurement. At the same time, combined with the current popular computer research fields of virtual reality and three-dimensional technology, it focuses on the application of virtual reality and three-dimensional visualization simulation technology in geographic information systems, and analyzes virtual dynamic simulation, three-dimensional space analysis, virtual geographic environment, and three-dimensional data acquisition. Basic issues such as processing, three-dimensional data structure and models, and discuss the development trend of VR-GIS and 3D-GIS.

Drone Mapping Versus Terrestrial Laser Scanning for Building Mapping andAssesement

Drone mapping and terrestrial laser scanning are two advanced techniques that have been used around the world in recent years for documentation and mapping of buildings with historic value, construction project/infrastructure monitoring, mapping of cities, and a lot more survey. A well-defined methodology for data extraction, modelling, and monitoring is yet to evolve. This study compares and demonstrates terrestrial laser scanning and drone survey for building mapping with validation and accuracy checks. A terrestrial scanner was used to perform three scans of a building and drone was used to capture high-performance vertical images and oblique images. A three dimensional (3D) model was generated using advanced software techniques. Building facade elevation and plans were extracted from the 3D model after validation and accuracy check. The drone model produced high-quality visualization while terrestrial scanner output data allowed to extracting minute details after removing noise. Drone mapping and terrestrial laser scanning are two unique stand-alone technologies for serving different purposes individually, but they greatly complement each other when used together. Visualization of digital real-time 3D model helps in scrolling around at different nooks and corners of the building for a survey which is not possible through conventional mapping techniques; this is one of the greatest advantages of these techniques.

Cost‐Effective and Ultraportable Smartphone‐Based Vision System for Structural Deflection Monitoring

This work demonstrates the viability of using a smartphone‐based vision system to monitor the deflection of engineering structures subjected to external loadings. The video images of a test structure recorded by a smartphone camera are processed using the state‐of‐the‐art subset‐based digital image correlation (DIC) algorithm to extract the vertical image displacement of discrete calculation points defined on the test object. The measured vertical image displacement can be converted to deflection (vertical displacement) by easy‐to‐implement scaling factor determination approaches. For accuracy validation, laboratory experiments using a cantilever beam subjected to external loadings were performed. The deflection and inherent frequency of the test cantilever beam measured by the proposed smartphone‐based vision system were compared with those measured by conventional dial gauges and a dynamic strain gauge. The relative errors were estimated as 1% and 0.15% for deflection and inherent frequency, respectively. Outdoor real bridge deflection monitoring tests were also carried out on an overpass with subway passing by, and the measured deflection‐time curves agree well with actual situations. The cost‐effective, ultraportable, and easy‐to‐use smartphone‐based vision system not only greatly decreases the hardware investment and complexity in deflection measurement system construction, but also increases the convenience and efficiency of deflection monitoring of engineering structures.

A Comparison between Unmanned Aerial Vehicle And Aerial Survey Acquired in Separate Dates for The Production of Orthophotos

The Unmanned aerial vehicles (UAV) have become convenient for many applications such as the production of 3D ground models, orthophotos, road maintenance and dam monitoring, etc. The aim of this paper is to investigate the accuracy of the orthophoto obtained from UAV images compared to the existing orthophoto generated from aerial survey conducted by Vossing German Company in Duhok city, 2011 using GPS ground control point as a reference base. A digital elevation model (DEM) of the university campus was created with the UAV vertical images. Low flight height of 80m was used with the rate of 70% forward overlap and 40% side lap. An orthophoto is produced based on this DEM. To assess the accuracy of UAV orthophotos and the existing orthophotos produced from aerial survey, 7 ground control points (GCPs) were distributed and used for processing the orthophoto mosaic and DEM, 12 GCPs were used as check points. The 19 GCPs were accurately measured using GPS-RTK. Discrepancy in horizontal position of the GCPs measured on both orthophotos was obtained relative to the reference GPS base control points.The UAV data were processed and analyzed using two softwares pix4D and Agisoft Photoscan. The RMS errors obtained from both orthophotos is presented relative to the GPS base. Sub-centimetre accuracy for horizontal position was achieved from the UAV orthophoto at low flight altitude.

Structure-From-Motion Photogrammetry of Antarctic Historical Aerial Photographs in Conjunction with Ground Control Derived from Satellite Data

A longer temporal scale of Antarctic observations is vital to better understanding glacier dynamics and improving ice sheet model projections. One underutilized data source that expands the temporal scale is aerial photography, specifically imagery collected prior to 1990. However, processing Antarctic historical aerial imagery using modern photogrammetry software is difficult, as it requires precise information about the data collection process and extensive in situ ground control is required. Often, the necessary orientation metadata for older aerial imagery is lost and in situ data collection in regions like Antarctica is extremely difficult to obtain, limiting the use of traditional photogrammetric methods. Here, we test an alternative methodology to generate elevations from historical Antarctic aerial imagery. Instead of relying on pre-existing ground control, we use structure-from-motion photogrammetry techniques to process the imagery with manually derived ground control from high-resolution satellite imagery. This case study is based on vertical aerial image sets collected over Byrd Glacier, East Antarctica in December 1978 and January 1979. Our results are the oldest, highest resolution digital elevation models (DEMs) ever generated for an Antarctic glacier. We use these DEMs to estimate glacier dynamics and show that surface elevation of Byrd Glacier has been constant for the past ∼40 years.

Collapsed Building Detection Using 3D Point Clouds and Deep Learning

Collapsed buildings should be detected with the highest priority during earthquake emergency response, due to the associated fatality rates. Although deep learning-based damage detection using vertical aerial images can achieve high performance, as depth information cannot be obtained, it is difficult to detect collapsed buildings when their roofs are not heavily damaged. Airborne LiDAR can efficiently obtain the 3D geometries of buildings (in the form of point clouds) and thus has greater potential to detect various collapsed buildings. However, there have been few previous studies on deep learning-based damage detection using point cloud data, due to a lack of large-scale datasets. Therefore, in this paper, we aim to develop a dataset tailored to point cloud-based building damage detection, in order to investigate the potential of point cloud data in collapsed building detection. Two types of building data are created: building roof and building patch, which contains the building and its surroundings. Comprehensive experiments are conducted under various data availability scenarios (pre–post-building patch, post-building roof, and post-building patch) with varying reference data. The pre–post scenario tries to detect damage using pre-event and post-event data, whereas post-building patch and roof only use post-event data. Damage detection is implemented using both basic and modern 3D point cloud-based deep learning algorithms. To adapt a single-input network, which can only accept one building’s data for a prediction, to the pre–post (double-input) scenario, a general extension framework is proposed. Moreover, a simple visual explanation method is proposed, in order to conduct sensitivity analyses for validating the reliability of model decisions under the post-only scenario. Finally, the generalization ability of the proposed approach is tested using buildings with different architectural styles acquired by a distinct sensor. The results show that point cloud-based methods can achieve high accuracy and are robust under training data reduction. The sensitivity analysis reveals that the trained models are able to locate roof deformations precisely, but have difficulty recognizing global damage, such as that relating to the roof inclination. Additionally, it is revealed that the model decisions are overly dependent on debris-like objects when surroundings information is available, which leads to misclassifications. By training on the developed dataset, the model can achieve moderate accuracy on another dataset with different architectural styles without additional training.

Line-field confocal optical coherence tomography for three-dimensional skin imaging.

This paper reports on the latest advances in line-field confocal optical coherence tomography (LC-OCT), a recently invented imaging technology that now allows the generation of either horizontal (x × y) section images at an adjustable depth or vertical (x × z) section images at an adjustable lateral position, as well as three-dimensional images. For both two-dimensional imaging modes, images are acquired in real-time, with real-time control of the depth and lateral positions. Three-dimensional (x × y × z) images are acquired from a stack of horizontal section images. The device is in the form of a portable probe. The handle of the probe has a button and a scroll wheel allowing the user to control the imaging modes. Using a supercontinuum laser as a broadband light source and a high numerical microscope objective, an isotropic spatial resolution of ∼1 µm is achieved. The field of view of the three-dimensional images is 1.2 mm × 0.5 mm × 0.5 mm (x × y × z). Images of skin tissues are presented to demonstrate the potential of the technology in dermatology.

Two-dimensional beam emission spectroscopy for hydrogen isotope negative neutral beam in Large Helical Device

A new beam emission spectroscopy system that has improved lines of sight is installed in the Large Helical Device (LHD), and routine measurement has been started in the 21st LHD experiment campaign in 2019–2020. The new system is optimized for hydrogen isotope experiments by equipping a rotatable large-diameter interference filter to be compatible with either the hydrogen or the deuterium beam emission component. An avalanche photo diode detector array having 8 × 8 pixels is used for obtaining a radial–vertical image of electron density fluctuation covering the mid-radius to the plasma periphery. Spatial resolution and wavenumber cutoff are derived from equilibrium reconstruction and plasma kinetic profiles. Obtained fluctuation data is presented for a low field high beta discharge. The spatiotemporal structure of the fluctuations is clearly shown by Fourier correlation analyses.

Line-field confocal optical coherence tomography-Practical applications in dermatology and comparison with established imaging methods.

Non-invasive diagnostic techniques in dermatology gained increasing popularity in the last decade. Reflectance confocal microscopy (RCM) and optical coherence tomography (OCT) are meanwhile established in research and clinical routine. While OCT is mainly indicated for detecting non-melanoma skin cancer, RCM has proven its usefulness additionally in distinguishing melanocytic lesions. Line-field confocal optical coherence tomography (LC-OCT) is an emerging tool combining the principles of both above-mentioned methods. Healthy skin at different body sites and exemplary skin lesions (basal cell carcinoma, malignant melanoma, actinic keratosis) were examined using dermoscopy, RCM, OCT and LC-OCT. Standard features for RCM and OCT and comparable features for LC-OCT were analysed. LC-OCT has a lower penetration depth but superior resolution compared to OCT. In comparison with RCM, which provides only horizontal sections, LC-OCT creates both vertical and horizontal images in real time and has nearly the same cellular resolution. Our preliminary experiences suggest that LC-OCT combines the advantages of RCM and OCT, with optimal resolution and penetration depth to diagnose all types of skin cancer. Larger systematic studies are needed to further characterize the field of use of this device and its sensitivity and specificity compared to histology.

영화 <기생충>에 나타난 집의 모티프 연구

The purpose of this study is to analyze the motif of the house shown in Parasite, focusing on the characteristics and values of the house of Gaston Bachelard, which are verticality, centrality, and protective value, and examine how this film uses it for class problems. First, this film vertically polarizes the secret underground space of Dong-Ik’s house and the rest of the ground space, and makes the ’irrationality of the basement’ and ’the rationality of the roof’ conflict. This vertical image that does not fit Korean culture, however, when Koreans think of their horizontal houses, not only utilizes the basement reminiscent of darkness, conspiracy, and the depths of human beings, but also expresses the seriousness of expanding polarization symbolized by the basement and the semi-underground and the difficulty of overcoming it. Therefore, it can be said that the verticality of the house in this film is class verticality. Next, in this movie, the ‘sunny garden’ of Dong-Ik’s house and the ‘toilet on the altar’ of Ki-Taek’s house are expressed as the center of the house as an intimate shelter. Unlike the former, the latter is an inadequate center in light of our traditional housing view, which considers a sunny house a good house and toilets to be unclean and to be avoided. The suggestion of the ‘toilet on the altar’, nevertheless, as the center of the semibasement house is to maximize the contrast with the garden, based on the paradox of the toilet, and to highlight as much as possible the harsh conditions with which the underground class is faced. Lastly, the fact that Ki-Taek’s semibasement house does not have a protective value unlike Dong-Ik’s house is revealed through ‘submergence’ which is the biggest disaster of the semibasement house, and through ’smell’ caused by moisture, poor lighting and mold. Floods caused by heavy rains engulf even humble shelters, threatening the survival of the underground people, and smells become a stigma to discriminate against their class. Parasite reveals the roots of serious class problems in our society by using various images of the house.

Longitudinal Changes of Macular Curvature in Patients with Retinitis Pigmentosa

PurposeTo investigate the longitudinal changes of the macular curvature in eyes with retinitis pigmentosa (RP) and to determine the factors associated with the changes.MethodsWe reviewed the medical charts of 107 RP patients, for whom the axial length of their right eyes ranged from 21.5 to 26.0 mm and who had had been followed by spectral-domain optical coherence tomography (OCT). The OCT images at the initial and the most recent examinations were compared. The mean curvature of Bruch's membrane within 6 mm of the central macula obtained from the OCT images was evaluated as the mean macular curvature index (MMCI). Changes in the MMCI and their relationships with other clinical factors, including the ellipsoid zone (EZ) width, were assessed.ResultsThe MMCI decreased significantly in the vertical OCT images, from –15.47 × 10−5 µm−1 to –16.36 × 10−5 µm−1 (P = 0.008) during the mean observation period of 3.4 ± 1.4 years (mean ± SD). This indicated that the macular shape became more concave. The change to a steeper shape was more prominent in eyes with less photoreceptor degeneration and for which the EZ width was preserved at >2000 µm. In three eyes, the MMCI increased markedly by >5 × 10−5 µm−1, and this was accompanied by absorption of the macular edema.ConclusionsThe macular curvature in RP eyes becomes more concave in eyes with preserved EZ width.Translational RelevanceLongitudinal changes of the macular curvature in RP should be considered in future therapies, such as the implantation of the retinal prosthesis.

Matrix‐free crosshole elliptical‐anisotropy tomography: parametrization analysis and ground‐penetrating radar applications in carbonates

ABSTRACTA novel traveltime tomographic approach is applied to anisotropic media, limited to 2D geometry at present. A general anisotropic Eikonal solver based on a discontinuous Galerkin method is combined with an efficient adjoint formulation for multiparameter least‐squares inversion. This new approach is tested considering synthetic crosshole ground‐penetrating radar data. The configuration of the ground‐penetrating radar survey is inspired by a real experiment done on layered carbonate media disturbed by the presence of a deep gallery, which induces a localized high‐electromagnetic contrast. This made it possible to define a well‐adapted general workflow in this context. We notably show that, under the elliptical anisotropic assumption, the parametrization based on vertical and horizontal velocities provides less biased results than those obtained by considering the vertical velocity and the relevant Thomsen parameter . The initial vertical and horizontal velocity models are identical and built from an isotropic inversion. The presence of the high‐contrast gallery generates a weak diffraction pattern, which is taken into account in our tomography approach. It also creates potential artefacts due to the model discretization, which are mitigated by a model regularization term within the definition of the misfit function. This general workflow is then applied to the real experiment dataset. The vertical and horizontal velocity images provide similar structures as those previously obtained by isotropic full waveform inversion, complemented by an image of a rather weak elliptical anisotropy.