Sparse Locations Research Articles

Radar-Based Shape and Reflectivity Reconstruction Using Active Surfaces and the Level Set Method.

We investigate a multiview shape reconstruction problem based on an active surface model whose geometric evolution is driven by radar measurements acquired at sparse locations. Building on our previous work in the context of variational methods for the reconstruction of a scene conceptualized as the graph of a function, we generalize this inversion approach for a general geometry, now described by an active surface, strongly motivated by prior variational computer vision approaches to multiview stereo reconstruction from camera images. While conceptually similar, use of radar echoes within a variational scheme to drive the active surface evolution requires significant changes in regularization strategies compared to prior image based methodologies for the active surface evolution to work effectively. We describe all of these aspects and how we addressed them. While our long term objective is to develop a framework capable of fusing radar as well as other image based information, in which the active surface becomes an explicit shared reference for data fusion. In this paper, we explore the reconstruction using radar as a single modality, demonstrating that the presented approach can provide reconstructions of quality comparable to those from image based methods showing great potential for further development toward data fusion.

Frank-van der Merwe growth in bilayer graphene

Frank-van der Merwe growth in bilayer graphene

Three-dimensional spatiotemporal wind field reconstruction based on physics-informed deep learning

In this work, a physics-informed deep learning model is developed to achieve the reconstruction of the three-dimensional (3-D) spatiotemporal wind field in front of a wind turbine, by combining the 3-D Navier–Stokes equations and the scanning LIDAR measurements. To the best of the authors’ knowledge, this is for the first time that the full 3-D spatiotemporal wind field reconstruction is achieved based on real-time measurements and flow physics. The proposed method is evaluated using high-fidelity large eddy simulations. The results show that the wind vector field in the whole 3-D domain is predicted very accurately based on only scalar line-of-sight LIDAR measurements at sparse locations. Specifically, at the baseline case, the prediction errors for the streamwise, spanwise and vertical velocity fields are 0.263 m/s, 0.397 m/s and 0.361 m/s, respectively. The prediction errors for the horizontal and vertical direction fields are 2.84° and 2.58° which are important in tackling yaw misalignment and turbine tilt control, respectively. Further analysis shows that the 3-D wind features are captured clearly, including the evolutions of flow structures, the wind shear in vertical direction, the blade-level speed variations due to turbine rotation, and the speed variations modulated by the turbulent wind. Also, the developed model achieves short-term wind forecasting without the commonly-used Taylor’s frozen turbulence hypothesis. Furthermore it is very useful in advancing other wind energy research fields e.g. wind turbine control & monitoring, power forecasting, and resource assessments because the 3-D spatiotemporal information is important for them but not available with current sensor and prediction technologies.

A new methodology based on finite element method (FEM) for generation of the probability field of rock types from subsurface

In this paper, we present a novel methodology by employing the finite element method (FEM) for creating probability fields of rock types (facies) present in a reservoir. The method is developed for a hydrocarbon reservoir, which, situated thousands of meters underground, has a high uncertainty regarding its geology. However, from the exploration phase, are collected facies observations (hard data) about the rock types present at some sparse locations of the reservoir domain. These observations are projected into a continuous probabilistic space, which is the environment of the FEM. The novelty consists of the fact that the probability field of a facies type is viewed as a deformation field of a plate. For each facies type, the FEM constructs a deformation profile which will be a probability field. The fields generated with the FEM verify the conditions of the probability theory without post-regularization techniques. We test the methodology for a two-dimensional reservoir model with three facies types. We also show the applicability of the new method by introducing the probability fields in the truncated pluri-Gaussian simulation method for simulating facies fields which are the inputs in the assisted history matching of the reservoir.

Spatiotemporal wind field prediction based on physics-informed deep learning and LIDAR measurements

Spatiotemporal wind field information is of great interest in wind industry e.g. for wind resource assessment and wind turbine/farm monitoring & control. However, its measurement is not feasible because only sparse point measurements are available with the current sensor technology such as LIDAR. This work fills the gap by developing a method that can achieve spatiotemporal wind field predictions by combining LIDAR measurements and flow physics. Specifically, a deep neural network is constructed and the Navier–Stokes equations, which provide a good description of atmospheric flows, are incorporated in the deep neural network by employing the physics-informed deep learning technique. The training of this physics-incorporated deep learning model only requires the sparse LIDAR measurement data while the spatiotemporal wind field in the whole domain (which cannot be measured) can be predicted after training. This study, which can discover complex wind patterns that do not present in the training dataset, is totally distinct from previous machine learning based wind prediction studies which treat machine learning models as “black-box” and require the corresponding input and target values to learn complex relations. The numerical results on the prediction of the wind field in front of a wind turbine show that the proposed method predicts the spatiotemporal flow velocity (including both downwind and crosswind components) in the whole domain very well for a wide range of scenarios (including various measurement noises, resolutions, LIDAR look directions, and turbulence levels), which is promising given that only line-of-sight wind speed measurements at sparse locations are used.

A feasibility study of radar-based shape and reflectivity reconstruction using variational methods

Remote sensing radar techniques provide highly detailed imaging. Nevertheless, radar images do not offer directly retrievable representations of shape within the scene. Therefore, shape reconstruction from radar typically relies on applying post-processing computer vision techniques, originally designed for optical images, to radar imaging products. Shape reconstruction directly from raw data would be desirable in many applications, e.g. in computer vision and robotics. In this perspective, inversion seems an attractive approach. Nevertheless, inversion has seldom been attempted in the radar context, as high frequency signals lead to energy functionals dominated by tightly packed narrow local minima. In this paper, we take the first step in developing a framework in which radar signals and images can be jointly used for shape reconstruction. In particular, we investigate the feasibility of shape reconstruction by inversion of pulse-compressed radar signals alone, collected at sparse locations. Motivated by geometric methods that have matured within the fields of image processing and computer vision, we pose the problem in a variational context obtaining a partial differential equation for the evolution of an initial shape towards the shape-reflectivity combination that best reproduces the data. While doing so, we highlight several non-obvious difficulties encountered and discuss how to surpass them. We illustrate the potential of this approach through three simulated examples and discuss several implementation choices, including boundary conditions, reflectivity estimation, and radiative models. The success of our simulations shows that this variational approach can naturally accommodate radar inversion and has the potential for further expansion towards active surfaces and level set applications, where we believe it will naturally complement current applications with optical images.

Latitudinal patterns of alien plant invasions

AbstractLatitudinal patterns of biodiversity have long been a central topic in ecology and evolutionary biology. However, while most previous studies have focused on native species, little effort has been devoted to latitudinal patterns of plant invasions (with a few exceptions based on data from sparse locations). Using the most up‐to‐date worldwide native and alien plant distribution data from 801 regions (including islands), we compared invasion levels (i.e. alien richness/total richness) in the Northern and Southern Hemispheres and across continental regions and islands around the globe. Results from quantile regressions using B‐splines to model nonlinearity showed (1) declining richness with increasing latitude, although the highest alien richness occurs at around 40 degrees in both hemispheres, (2) decreasing invasion levels towards higher latitudes on islands but a unimodal pattern in invasion level in continental regions in each hemisphere, (3) significantly higher invasion levels on islands than in continental regions and (4) a greater variability in invasion levels on islands at low latitudes than on high‐latitude islands. In continental regions, only the mid‐latitudes had high variability with both low and high invasion levels. Our findings identified latitudes with invasion hotspots where management is urgently needed, and latitudes with many areas of low invasions but high conservation potential where prevention of future invasions should be the priority.

Localized Subsidence Zones in Gävle City Detected by Sentinel-1 PSI and Leveling Data

Among different sets of constraints and hazards that have to be considered in the management of cities and land use, land surface subsidence is one of the important issues that can lead to many problems, and its economic consequences cannot be ignored. In this study, the ground surface deformation of Gävle city in Sweden is investigated using the Persistent Scatterer Interferometry (PSI) technique as well as analyzing the historical leveling data. The PSI technique is used to map the location of hazard zones and their ongoing subsidence rate. Two ascending and descending Sentinel-1 datasets, collected between January 2015 and May 2020, covering the Gävle city, were processed and analyzed. In addition, a long record of a leveling dataset, covering the period from 1974 to 2019, was used to detect the rate of subsidence in some locations which were not reported before. Our PSI analysis reveals that the center of Gävle is relatively stable with minor deformation ranged between −2 ± 0.5 mm/yr to +2 ± 0.5 mm/yr in vertical and east–west components. However, the land surface toward the northeast of the city is relatively subsiding with a higher annual rate of up to −6 ± 0.46 mm/yr. The comparison at sparse locations shows a close agreement between the subsidence rates obtained from precise leveling and PSI results. The regional quaternary deposits map was overlaid with PSI results and it shows the subsidence areas are mostly located in zones where the subsurface layer is marked by artificial fill materials. The knowledge of the spatio-temporal extents of land surface subsidence for undergoing urban areas can help to develop and establish models to mitigate hazards associated with such land settlement.

Robust fusion algorithms for unsupervised change detection between multi-band optical images — A comprehensive case study

Unsupervised change detection techniques are generally constrained to two multi-band optical images acquired at different times through sensors sharing the same spatial and spectral resolution. In the case of the optical modality, largely studied in the remote sensing community, a straight comparison of homologous pixels such as pixel-wise differencing is suitable. However, in some specific cases such as emergency situations, punctual missions, defense and security, the only available images may be those acquired through different kinds of sensors with different resolutions. Recently some change detection techniques, dealing with images with different spatial and spectral resolutions, have been proposed. Nevertheless, they are focused on a specific scenario where one image has a high spatial and low spectral resolution while the other has a low spatial and high spectral resolution. This paper addresses the problem of detecting changes between any two multi-band optical images disregarding their spatial and spectral resolution disparities. To overcome resolution disparity, state-of-the art methods apply conventional change detection methods after preprocessing steps applied independently on the two images, e.g. resampling operations intended to reach the same spatial and spectral resolutions. Nevertheless, these preprocessing steps may waste relevant information since they do not take into account the strong interplay existing between the two images. Conversely, in this paper, we propose a method that more effectively uses the available information by modeling the two observed images as spatially and spectrally degraded versions of two (unobserved) latent images characterized by the same high spatial and high spectral resolutions. Covering the same scene, the latent images are expected to be globally similar except for possible changes in spatially sparse locations. Thus, the change detection task is envisioned through a robust fusion task which enforces the differences between the estimated latent images to be spatially sparse. We show that this robust fusion can be formulated as an inverse problem which is iteratively solved using an alternating minimization strategy. The proposed framework is implemented for an exhaustive list of applicative scenarios and applied to real multi-band optical images. A comparison with state-of-the-art change detection methods evidences the accuracy and the versatility of the proposed robust fusion-based strategy.

Construction of an indoor radio environment map using gradient boosting decision tree

Radio environment maps represent a signal strength map or a coverage area of radio networks. Constructing such maps involves gathering signal coverage information in sparse locations, which can be conventionally performed by measurement methods such as the manual drive test. Nevertheless, as this process is large-scale, time-consuming, and costly, several methods for minimization of drive tests have been introduced. Machine learning is commonly used in solving regression or classification problems; in several studies, its performance even surpassed human abilities. In this study, we applied the gradient boosting algorithm to construct radio environment maps from sparse data gathered by user equipments. XGBoost and light gradient boosting machine were experimentally evaluated in constructing base station coverage, reference signal received power, reference signal received quality, and signal-to-noise ratio heatmaps, under various configuration settings. Results validated the superior performance of the two approaches against existing baseline methods k-nearest neighbor and support vector machine. Furthermore, we also assessed our model’s ability to construct radio environment maps based on unseen configuration settings, which confirmed reliable results even if they were trained using completely different sets of configuration settings.

Constraining magnetic amplitude inversion with magnetotelluric data to image volcanic units: A case study

We present a case study on imaging volcanic units in gas exploration by constraining magnetic amplitude inversions using magnetotelluric (MT) sounding data at sparse locations. Magnetic data can be effective in mapping volcanic units because they have remanent magnetization and significant susceptibility contrast with surrounding rocks. Although magnetic data can identify the lateral distribution of volcanic units, they often have difficulties in defining the depth extent. For this reason, additional structural constraints from other geophysical methods can often help improve the vertical resolution. Among the independent geophysical methods, MT data can provide the needed structural information at a low cost. We have investigated an approach to combine a set of sparse MT soundings with magnetic amplitude data to image the distribution of volcanics in a basin environment. We first use a blocky 1D MT inversion based on Ekblom norm to obtain the structural constraint, and then we perform a constrained 3D magnetic amplitude inversion to recover the distribution of effective susceptibility by incorporating the structural information from MT soundings. We determine that even a small number of MT stations (e.g., 20) in a [Formula: see text] area is sufficient to drastically improve the magnetic amplitude inversion. Our results indicate that magnetic amplitude inversion with structural constraint from MT soundings form a practical and cost-effective means to map the lateral and vertical distribution of volcanics.

Low-Power Embedded System for Gait Classification Using Neural Networks

Abnormal foot postures can be measured during the march by plantar pressures in both dynamic and static conditions. These detections may prevent possible injuries to the lower limbs like fractures, ankle sprain or plantar fasciitis. This information can be obtained by an embedded instrumented insole with pressure sensors and a low-power microcontroller. However, these sensors are placed in sparse locations inside the insole, so it is not easy to correlate manually its values with the gait type; that is why a machine learning system is needed. In this work, we analyse the feasibility of integrating a machine learning classifier inside a low-power embedded system in order to obtain information from the user’s gait in real-time and prevent future injuries. Moreover, we analyse the execution times, the power consumption and the model effectiveness. The machine learning classifier is trained using an acquired dataset of 3000+ steps from 6 different users. Results prove that this system provides an accuracy over 99% and the power consumption tests obtains a battery autonomy over 25 days.

The past and future of global river ice.

More than one-third of Earth's landmass is drained by rivers that seasonally freeze over. Ice transforms the hydrologic1,2, ecologic3,4, climatic5 and socio-economic6-8 functions of river corridors. Although river ice extent has been shown to be declining in many regions of the world1, the seasonality, historical change and predicted future changes in river ice extent and duration have not yet been quantified globally. Previous studies of river ice, which suggested that declines in extent and duration could be attributed to warming temperatures9,10, were based on data from sparse locations. Furthermore, existing projections of future ice extent are based solely on the location of the 0-°C isotherm11. Here, using satellite observations, we show that the global extent of river ice is declining, and we project a mean decrease in seasonal ice duration of 6.10±0.08 days per 1-°C increase in global mean surface air temperature. We tracked the extent of river ice using over 400,000 clear-sky Landsat images spanning 1984-2018 and observed a mean decline of 2.5percentage points globally in the past three decades. To project future changes in river ice extent, we developed an observationally calibrated and validated model, based on temperature and season, which reduced the mean bias by 87per cent compared with the 0-degree-Celsius isotherm approach. We applied this model to future climate projections for 2080-2100: compared with 2009-2029, the average river ice duration declines by 16.7days under Representative Concentration Pathway (RCP) 8.5, whereas under RCP 4.5 it declines on average by 7.3days. Our results show that, globally, river ice is measurably declining and will continue to decline linearly with projected increases in surface air temperaturetowards the end ofthis century.

OctreeNet: A Novel Sparse 3-D Convolutional Neural Network for Real-Time 3-D Outdoor Scene Analysis

Convolutional neural networks (CNNs) for 3-D data analyses require a large size of memory and fast computation power, making real-time applications difficult. This article proposes a novel OctreeNet (a sparse 3-D CNN) to analyze the sparse 3-D laser scanning data gathered from outdoor environments. It uses a collection of shallow octrees for 3-D scene representation to reduce the memory footprint of 3-D-CNNs and performs point cloud classification on every single octree. Furthermore, the smallest non-trivial and non-overlapped kernel (SNNK) implements convolution directly on the octree structure to reduce dense 3-D convolutions to matrix operations at sparse locations. The proposed neural network implements a depth-first search algorithm for real-time predictions. A conditional random field model is utilized for learning global semantic relationships and refining point cloud classification results. Two public data sets (Semantic3D.net and Oakland) are selected to test the classification performance in outdoor scenes with different spatial sparsity. The experiments and benchmark test results show that the proposed approach can be effectively used in real-time 3-D laser data analyses. Note to Practitioners —This article was motivated by the limitations of existing deep learning technologies for analyzing 3-D laser scanning data. This technology enables robots to infer what the surroundings are, which is closely linked to semantic mapping and navigation tasks. Previous deep neural networks have seldom been used in robotic systems since they require a large amount of memory and fast computation power to apply dense 3-D operations. This article presents a sparse 3-D-Convolutional neural network (CNN) for real-time point cloud classification by exploiting the sparsity of 3-D data. This framework requires no GPUs. The practicality of the proposed method is verified on data sets gathered from different platforms and sensors. The proposed network can be adopted for other classification tasks with laser sensors.

Interpolation of daily rainfall data using censored Bayesian spatially varying model

This paper illustrates a Bayesian spatially varying space-time model to analyse daily precipitation in the Murray Darling Basin (MDB) of Australia. The model is capable of addressing data censoring at zero rainfall days using a latent Gaussian spatial field. The model is also useful for analysing data obtained from sparse locations, where spatially varying parameters are used to interpolate the rainfall estimates in high-density grids across the MDB region. We find that the proposed censored method provides a better out-of-sample predictive performance for the daily rainfall measurements compared to a non-truncated method. Additionally, it identifies location specific trends over the study region. The model is constructed under the Bayesian paradigm and the Markov chain Monte Carlo algorithm is used to obtain inferences of the model parameters. The R software spTDyn is hence extended to version 2.0 to implement the model.

Areal prediction of survey data using Bayesian spatial generalised linear models

The conditional autoregressive approach is popular to analyse data with geocoded boundary. However, spatial prediction is often challenging when observed data are sparse. It becomes more challenging in predicting areal units with different areal boundaries. Hence, this paper develops a spatial generalised linear model for spatial predictions using data from spatially misaligned sparse locations. A spatial basis function associated with the conditional autoregressive models and the kriging method is considered. The proposed model demonstrates its better predictive performance through a simulation study and then is applied to understand the spatial pattern of undecided voting preferences in Australia.

The past and future of global river ice

More than one-third of Earth’s landmass is drained by rivers that seasonally freeze over. Ice transforms the hydrologic1,2, ecologic3,4, climatic5 and socio-economic6–8 functions of river corridors. Although river ice extent has been shown to be declining in many regions of the world1, the seasonality, historical change and predicted future changes in river ice extent and duration have not yet been quantified globally. Previous studies of river ice, which suggested that declines in extent and duration could be attributed to warming temperatures9,10, were based on data from sparse locations. Furthermore, existing projections of future ice extent are based solely on the location of the 0-°C isotherm11. Here, using satellite observations, we show that the global extent of river ice is declining, and we project a mean decrease in seasonal ice duration of 6.10 ± 0.08 days per 1-°C increase in global mean surface air temperature. We tracked the extent of river ice using over 400,000 clear-sky Landsat images spanning 1984–2018 and observed a mean decline of 2.5 percentage points globally in the past three decades. To project future changes in river ice extent, we developed an observationally calibrated and validated model, based on temperature and season, which reduced the mean bias by 87 per cent compared with the 0-degree-Celsius isotherm approach. We applied this model to future climate projections for 2080–2100: compared with 2009–2029, the average river ice duration declines by 16.7 days under Representative Concentration Pathway (RCP) 8.5, whereas under RCP 4.5 it declines on average by 7.3 days. Our results show that, globally, river ice is measurably declining and will continue to decline linearly with projected increases in surface air temperature towards the end of this century. An analysis based on Landsat imagery shows that the extent of river ice has declined extensively over past decades and that this trend will continue under future global warming.

Probabilistic guidance for catheter tip motion in cardiac ablation procedures.

Radiofrequency catheter ablation is one of the commonly available therapeutic methods for patients suffering from cardiac arrhythmias. The prerequisite of successful ablation is sufficient energy delivery at the target site. However, cardiac and respiratory motion, coupled with endocardial irregularities, can cause catheter drift and dispersion of the radiofrequency energy, thus prolonging procedure time, damaging adjacent tissue, and leading to electrical reconnection of temporarily ablated regions. Therefore, positional accuracy and stability of the catheter tip during energy delivery is of great importance for the outcome of the procedure. This paper presents an analytical scheme for assessing catheter tip stability, whereby a sequence of catheter tip motion recorded at sparse locations on the endocardium is decomposed. The spatial sliding component along the endocardial wall is extracted from the recording and maximal slippage and its associated probability are computed at each mapping point. Finally, a global map is generated, allowing the assessment of potential areas that are compromised by tip slippage. The proposed framework was applied to 40 retrospective studies of congenital heart disease patients and further validated on phantom data and simulations. The results show a good correlation with other intraoperative factors, such as catheter tip contact force amplitude and orientation, and with clinically documented anatomical areas of high catheter tip instability.

Cosmogenic signature of geomagnetic reversals and excursions from the Réunion event to the Matuyama–Brunhes transition (0.7–2.14 Ma interval)

Long-term variations of the geomagnetic dipole moment (GDM) during periods of stable polarity and in transitional states (reversals and excursions) provide key information for understanding the geodynamo regime. Following several studies dealing with the Brunhes chron and the Matuyama–Brunhes transition, this study presents a new authigenic 10Be/9Be ratio (Be-ratio) record obtained from the MD97-2143 core (western Pacific Ocean). This new Be-ratio series yields a record of GDM variations covering the early Brunhes and mid to late Matuyama time period (i.e. 700–2140 ka), independently from the relative paleointensity (RPI) record obtained from the same core, that can be compared with available RPI records and stacks. Stratigraphic offsets measured between the Be-ratio peaks and the corresponding RPI minima reach 2 to 14 cm. They can be assigned to (post-) detrital remanent magnetization (pDRM) effects leading to magnetization locking-in delays varying from 2 to 12 ka in the studied core. 10Be overproduction episodes triggered by geomagnetic dipole moment lows (GDL) linked to polarity reversals and excursions confirm the global control exerted by the GDM on cosmogenic radionuclides production. A dipole moment reconstruction derived from the Beryllium-10 (BeDiMo) was compiled and calibrated using absolute paleointensity data. This independent record complements the available paleomagnetic RPI records, permitting 1) to overcome the pDRM lock-in offsets induced below the mixing layer, 2) to confront and increase the robustness and precision of GDM reconstructions and, 3) to better constrain the chronology of geomagnetic field instabilities during the mid to late Matuyama chron. Our new 10Be derived inventory is fully compatible with the GDL series linked to geomagnetic polarity reversals and events (Matuyama–Brunhes transition, Jaramillo and Olduvai subchron boundaries, Cobb Mountain, Réunion) and it strengthens the occurrence of several excursions (Kamikatsura, Santa Rosa, Punaruu, Bjorn, Gilsa, Gardar) that were until now reported from only sparse locations.

Magnetic Subsurface Imaging Systems in a Smartphone Based on the Built-In Magnetometer

We present a magnetic subsurface imaging system on a smart-phone that employs the built-in magnetometer. The smart-phone’s sensor measures magnetic field strength at sparse locations on a user-defined grid of the surveyed area. Based on the collected data, magnetic field distribution of the entire area is then reconstructed by using an interpolation algorithm, which yields field values in all of three-axial directions. Since the field shows the influence of the buried objects within the surveyed area to the background magnetic fields, the shape and size of the objects can be derived from the image representing the field. An implementation of the system as an application software on Android smart-phone, which consists of data measurement and reconstruction subsystems, is also described. We demonstrate the capability of the proposed system by performing subsurface imaging of small hidden objects in a laboratory test-range.