Deep-Learning-Based 3-D Surface Reconstruction—A Survey

Ultrareliable low-latency communication (URLLC) constitutes a key service class of the fifth generation (5G) and beyond cellular networks. Notably, designing and supporting URLLC pose a herculean task due to the fundamental need to identify and accurately characterize the underlying statistical models in which the system operates, e.g., interference statistics, channel conditions, and the behavior

The lattice spring model (LSM) is a novel method to simulate seismic wave propagation from a micromechanical perspective, which describes the elastic dynamics in complex media comprehensively and delineates the dynamic characteristics of the wavefield; however, it remains a great challenge to simulate wavefields at irregular interfaces with high accuracy. This work presents an adaptive location LSM

There are thousands of observation satellites orbiting the Earth, each of which captures massive-scale photographs covering millions of square kilometers everyday. In practice, these aerial photographs are with ultrahigh definitions (UHDs) and may contain tens to hundreds of ground objects (e.g., vehicles and rooftops). Understanding the multiple categories of a rich variety of UHD aerial photographs

Underwater acoustic signal recognition (UASR) systems face challenges in achieving high accuracy when processing complex data with low signal-to-noise ratio (SNR) in underwater environments, leading to limited noise robustness. Conventional approaches typically employ pre-trained denoising models for preprocessing noisy signals. However, due to disparate optimization goals between denoising and recognition

Exploring the relationship between spike phenotypes and wheat yield is crucial for selecting wheat ideotypes, but remains a subject of ongoing debate, primarily due to the lack of efficient spike phenotyping methods, particularly in field environments with complex light conditions. Light detection and ranging (lidar) can precisely capture 3-D plant information, minimally affected by light conditions

In the study of subsurface seismic imaging, solving the acoustic wave equation is a pivotal component in existing models. The advancement of deep learning (DL) enables solving partial differential equations (PDEs), including wave equations, by applying neural networks to identify the mapping between the inputs and the solution. This approach can be faster than traditional numerical methods when numerous

Amplitude variation with offset (AVO) inversion is an important method for estimating elastic parameters in geosciences. The inversion results are highly affected by the initial low-frequency model. Deep learning, as a data mining algorithm, has the potential to capture more reliable low-frequency components from seismic data and well logs. In order to fully utilize limited seismic data and recover

Onboard image data management and sharing are the foundations for achieving cooperative data processing and analysis in multiple unmanned aerial vehicles (multi-UAVs). However, various challenges, such as the lack of efficient onboard data indices, restrict the development of multi-UAV cooperative applications. Here, we propose a novel and versatile cooperative data management framework based on a

High-quality 3-D point cloud maps are essential for precise indoor environment modeling. However, constructing such maps in multistory indoor environments is challenging due to the presence of narrow nonstructural spaces, such as staircases, corners, and corridors with similar textures. Simultaneous localization and mapping (SLAM) in these scenes is particularly difficult, as cumulative errors can

Hyperspectral image classification (HSIC) has attracted widespread attention due to its important application in environment alterations and geophysical disaster monitoring. However, surface cultivation is not static as time passes, which leads to different hyperspectral image (HSI) information collected from the same area at different time periods. Therefore, researchers are currently eager to construct

Object detection in remote-sensing images (RSIs) relies on a large amount of labeled data for training. However, the increasing number of new categories and class imbalance make exhaustive annotation impractical. Few-shot object detection (FSOD) addresses this issue by leveraging meta-learning on seen base classes and fine-tuning on novel classes with limited labeled samples. Nonetheless, the substantial

The polar format algorithm (PFA) is a popular choice for general bistatic synthetic aperture radar (BiSAR) imaging due to its computational efficiency and adaptability to situations with complicated geometries or arbitrary flight trajectories. However, efficient and accurate compensation of 2-D residual phase errors induced by the wavefront curvature remains challenging when obtaining high-quality

In recent years, optical remote sensing image (ORSI) scene analysis has attracted increasing interest. However, existing networks show a trend of bifurcation. Lightweight networks have very high inference speed but poor inference of contextual information in highly complex backgrounds. In contrast, networks with high-performance contextual information reasoning capability require many parameters and

Compared to models pretrained on ImageNet, the segment anything model (SAM) has been trained on a massive segmentation corpus, excelling in both generalization ability and boundary localization. However, these strengths are still insufficient to enhance shadow detection without additional training, and it raises the question: do we still need precise manual annotations to fine-tune SAM for high detection

We deduce a novel interferometric tropospheric error (NITE) formula for ground-based global navigation satellite system interferometric reflectometry (GNSS-IR). This formula contains two parts: a geometric displacement error that accounts for the reflection point change due to the atmosphere and Earth’s curvature, and a path delay derived following the definition of the mapping function (with the small

Microwave radiometers are widely used in Earth observation and ocean monitoring for their strong penetration ability. Nevertheless, their utilization is somewhat constrained by their intrinsic low-resolution capability, particularly in complex applications such as coastal zone monitoring and analysis of typhoons. To overcome this limitation, resolution enhancements are necessary. Here, we present a

3-D line segments contain richer geometric and structural information than 3-D point clouds in man-made environments, which is beneficial for providing constraints to refine point-cloud-based mesh models or build accurate wireframes. However, the efficient reconstruction of 3-D line segments with high scene coverage from multiview images is still challenging. In this study, the depth maps obtained

With the increasing application of deep learning (DL) in various domains, salient object detection in optical remote sensing images (ORSIs-SOD) has attracted significant attention. However, most existing ORSI-SOD methods predominantly rely on local information from low-level features to infer salient boundary cues and supervise them using boundary ground truth (GT) but fail to sufficiently optimize

The multiple elevation beam (MEB) mode is an effective technique for enhancing imaging width in spaceborne synthetic aperture radar (SAR) systems. This mode combines intrapulse beam-steering during transmitting and digital beamforming (DBF) during receiving. By sequentially illuminating the far sub-swath followed by the near sub-swath, echoes from different sub-swaths can reach the antenna at the same

Super-resolution mapping (SRM) is an effective way to alleviate the mixed pixel problem of remotely sensed imagery by transforming the coarse-resolution fraction image originating from spectral unmixing into a fine-resolution land cover map. Deep learning has been widely used in SRM since it has a powerful ability to represent the complex heterogeneous spatial distribution patterns of land cover patches;

Unmanned aerial vehicles (UAVs) are widely used in various industries, and various visual tasks under the perspective of the UAV have been widely studied. In particular, the red-green-blue-T (RGB-T) detection method based on UAVs has shown significant advantages. However, the existing RGB-T methods are designed based on registration image pairs rather than detecting images directly acquired by UAVs

Multispectral and hyperspectral image fusion (MHIF) involves the fusion of high-spatial-resolution multispectral images (HR-MSIs) and low-spatial-resolution hyperspectral images (LR-HSIs) to generate high-spatial-resolution hyperspectral images (HR-HSIs) and has gained significant attention in the field of remote-sensing imaging. While CNN and Transformer models have shown effectiveness in MHIF, existing

Remote-sensing image–text retrieval (RSITR) has attracted widespread attention due to its great potential for rapid information mining ability on remote-sensing images. Although significant progress has been achieved, existing methods typically overlook the challenge posed by the extremely analogous descriptions, where the subtle differences remain largely unexploited or, in some cases, are entirely

Limited labeled training samples constitute a challenge in hyperspectral image classification, with much research devoted to cross-domain adaptation, where the classes of the source and target domains are different. Current cross-domain few-shot learning (FSL) methods only use a small number of sample pairs to learn the discriminant features, which limits their performance. To address this problem

Visible-near infrared (VIS-NIR) and thermal infrared (TIR) methods have long been used for ice cloud property retrievals based on satellite observations. Both retrieval methods are sensitive to the assumed ice particle models, which can significantly impact the accuracy of the retrieved microphysical and radiative properties of ice clouds. The two-habit model (THM) is considered a suitable ice particle

Polarimetric inverse synthetic aperture radar (ISAR) can operate all-day and all-weather, making it crucial for space surveillance. The compact polarimetric (CP) mode balances hardware complexity and polarimetric information, which is commonly equipped with ISAR systems. Given the constraints of limited physical conditions, exploring ISAR image super-resolution is worthwhile. Currently, deep learning

Contrastive learning (CL) with learnable examples performs outstandingly in data representation. However, when dealing with hard samples, instance-level alignment with excessive uniformity may descend into trivial clusters, especially when confronted with interclass similarity and intraclass diversity in hyperspectral images (HSIs). To solve this problem, we regard prototypical CL as tracing the potential

Self-supervised learning is attracting large attention in point cloud understanding. However, exploring discriminative and transferable features still remains challenging due to their nature of irregularity. We propose a geometrically and adaptively masked autoencoder on point clouds for self-supervised learning, termed PointGame. PointGame contains two core components: GATE and EAT. GATE stands for

Recently, the Transformer-based technique has emerged as a promising solution for modeling contextual information in remote sensing (RS) scenes and has found widespread applications in RS scene classification. However, how to make full use of intermediate features learned in Transformers is of crucial importance in the RS scene classification tasks. Therefore, this article proposes a hierarchical feature

Deep representation learning has improved automatic remote sensing change detection (RSCD) in recent years. Existing methods emphasize primarily convolutional neural networks (CNNs) or transformer-based networks. However, most of them neither effectively combine CNNs and transformers nor use prior geometric information to refine regions. In this article, a novel geometric representation transformer

This study developed an efficient 3-D inversion algorithm for the controlled-source electromagnetic method (CSEM). The spectral element method based on high-order Gauss-Lobatto-Legendre (GLL) basis functions with infinite element boundary conditions is used to quickly solve forward problems and adjoint forward problems for inversion, which can significantly reduce the computational cost while guarantee

Light detection and ranging (LiDAR)–camera fusion can enhance the performance of 3-D object detection by utilizing complementary information between depth-aware LiDAR points and semantically rich images. Existing voxel-based methods face significant challenges when fusing sparse voxel features with dense image features in a one-to-one manner, resulting in the loss of the advantages of images, including

Diffraction separation and imaging are important for subsurface discontinuities characterization. However, conventional diffraction separation methods may lose validity when the diffractions and reflections do not have discernible differences in data domain. Moreover, due to limited acquisition geometry and narrow frequency band of seismic data, the diffraction imaging methods that use conventional

This study uses data from January 2020 to July 2022 to develop an artificial neural network (ANN)-based model to retrieve atmospheric water vapor (Wv) pressure profiles from Constellation Observing System for Meteorology, Ionosphere (COSMIC-2) observations over the Indian region. Prior to using the COSMIC-2 (C2) refractivity, we evaluated the C2 refractivity’s quality by comparing it with radiosonde

As an effective manner to reduce data redundancy and processing inconvenience, hyperspectral band selection aims to select a subset of informative and discriminative bands from the original data cube. Although a large number of approaches have been proposed and obtained great success, they still face at least two issues. First, most of the previous methods only consider the redundancy between neighbor

Contrastive self-supervised learning (CSSL) is a promising method for extracting effective features from unlabeled data. It performs well in image-level tasks, such as image classification and retrieval. However, the existing CSSL methods are not suitable for pixel-level tasks, for example, change detection (CD), since they ignore the correlation between local patches or pixels. In this article, we

State-of-the-art (SOTA) phase linking (PL) methods for distributed scatterer (DS) interferometry (DSI) retrieve consistent phase histories from the sample coherence matrix or the one whose magnitudes are calibrated. To unify them, we first propose a framework consisting of sample coherence matrix estimation and Kullback–Leibler (KL) divergence minimization. Within such framework, we observe that the

This article presents a novel approach to inverse synthetic aperture radar (ISAR) image enhancement using time-frequency (TF) vertical synchrosqueezing (VSS). The authors proposed adding the signal extraction block to the ISAR image processing path. Since the VSS technique does not change the signal amplitude-phase dependencies but only performs its decomposition and denoising, the proposed method

Hyperspectral image (HSI) classification faces the inherent challenge of small sample learning, primarily due to the difficulty in labeling vast land covers. Meta-learning, with its ability to learn transferable meta-knowledge from existing HSIs, is seen as a promising solution. However, different HSIs usually have varying distributions manifested as differing spectral wavelengths and reflectance shifts

With the abundant emergence of remote sensing (RS) data sources, multimodal remote sensing observation has become an active field. Extracting valuable information from multimodal data has the potential to make a significant contribution to applications such as urban planning and monitoring. However, existing studies are deficient in extracting spectral and spatial features from hyperspectral (HS) remote

The data-driven inversion methods based on deep learning (DL) for gravity and magnetic data have been well studied in existing literature due to their advantages such as less constraint requirements and extremely high efficiency compared with various conventional model-driven inversion methods. However, some issues should be further addressed when these methods are applied to the joint inversion, such

Conventional resistivity inversion methodologies encounter constraints in perpetual monitoring owing to the necessity for recurrent measurements. In response, this research leverages a 3-D finite element method to formulate an approximate geometry imaging of cross-borehole resistivity during forward modeling, circumventing the direct computation of Jacobian matrix equations in the electric field. This

The space geodetic technologies [e.g., global navigation satellite system (GNSS) and Gravity Recovery and Climate Experiment (GRACE)/GRACE Follow-on (GFO)] provide effective tools to infer terrestrial water storage (TWS) change, which is an important indicator of the hydrological cycle and climate change. This study investigated the optimization approaches of the Slepian basis function (SBF) method

Interferometric measurements of very-high-frequency (VHF) radio frequency signals produced by lightning are one of the most effective techniques for studying lightning breakdown processes, so uncertainty and error analyses of interferometric location results have become important topics. Based on the plane wave approximation of lightning RF signal transmission for interferometric location, a geometric

In this article, we consider the detection problem of marine radar targets embedded in correlated non-Gaussian sea clutter, which is modeled by a compound Gaussian model with lognormal texture (CG-LN) and unknown covariance matrices. To reduce the dependence of detectors on training data, the original radar data are transformed via exploiting the persymmetric structure of clutter covariance matrix

We propose a deep learning framework of semantic position probability distribution for synthetic aperture radar (SAR)-optical image matching, termed as SPPD. Unlike the pixel-by-pixel searching matching method, a correspondence is directly obtained by an outputted matching position probability distribution. First, multiscale pyramidal features are created for each pixel in the SAR and optical images

Supervised deep neural networks (DNNs) have proven to be powerful tools for synthetic aperture radar (SAR) image interpretation tasks. However, they present a formidable challenge in acquiring a substantial amount of labeled data. In this article, we investigate the promising technique of contrastive self-supervised learning for SAR image classification. This approach allows us to take advantage of

Rapid advances in remote sensing technology have allowed its extensive use in defense, land use planning, urban traffic monitoring, and natural disaster warning. Remote sensing technology has penetrated every aspect of modern life. However, some problems need to be solved in the use of remote sensing data, such as the presence of clouds in images. Efficient air–ground data transmission can be realized

Logging-while-drilling (LWD) sonic data are critical for marine gas hydrate reservoir evaluation and production prediction. However, acquiring complete acoustic logs, particularly shear wave, poses significant challenges and incurs high costs. To tackle this issue, we develop a two-branch hybrid framework for predicting LWD sonic logs of hydrate-bearing sediments from existing logging data. One branch

This study regards the assessment of surface gravity surveying for CO2 plume monitoring in the deep Johansen saline aquifer, a potential offshore site for CO2 geologic storage. We used the available benchmark model and geological information to simulate the injection and postinjection phases. We calculated the gravity response at the surface from the estimated models of reservoir density and saturation

Accurate prediction of the ship motion attitude in the future period is important to ensure the safety of offshore operations and sea navigation. Coherent S-band radar is a novel wave monitoring device that can be applied to advance the development of ship motion attitude prediction. In this study, we propose a method for realizing ship motion attitude prediction from radial velocity of water particle

The remote sensing (RS) image super-resolution (SR) algorithm aims to reconstruct a high-resolution (HR) image with rich texture details from a given low-resolution (LR) image, improving the spatial resolution. It has been widely concerned in RS image processing and application. Most current deep-learning-based methods rely on paired training datasets. However, most datasets are often based on bicubic

For fast solving the electromagnetic scattering of multiple moving targets above or on a rough surface, the multi-dynamic-octrees-based shooting and bouncing ray (SBR) algorithm is proposed in the article. First, a local coordinate system (LCS) for each object is established that moves with targets in the geodetic coordinate system (GCS). The dynamic octree structure based on LCS is applied to keep

Six Clouds and the Earth’s Radiant Energy System (CERES) instruments on four satellites are used to produce a global continuous multidecadal record of Earth’s radiation budget (ERB) at the top-of-atmosphere (TOA). Each CERES instrument was calibrated and characterized on the ground before launch, while postlaunch calibration was conducted using onboard calibration sources. The performance of the CERES

Microwave links as a novel method of rainfall monitoring have been extensively studied over the last two decades. Still, the applicability of $\gamma $ (rain-induced attenuation rate)– $R$ (rain rate) relationship in quantitative rainfall inversion with microwave link under tilted conditions has not been systematically investigated to date. With the establishment of rain-induced attenuation fields

Active radar instruments provide vertically resolved clouds and precipitation measurements that cannot be provided by the passive instruments. These active measurements are not conventionally assimilated into the data assimilation systems because of the lack of fast forward radiative transfer (RT) models and also difficulties in the error modeling of the measurements. This article describes the development

Lack of interpretability and weak generalization ability have become the major challenges with data-driven intelligent synthetic aperture radar-automatic target recognition (SAR-ATR) technology, especially in practical applications with azimuth-sparse training samples. A novel insight into SAR image representation with neural networks from a causal perspective is presented in this article. First, a

One-bit synthetic aperture radar (SAR) imaging, which collects echoes into one-bit quantized samples, is a highly promising technique for the faster SAR images acquisition and simplified implementations in numerous applications. Harmonics, accompanied with one-bit sampling operation, are one of the fundamental factors that trigger severe aliasing artifacts problem in frequency domain. For one-bit SAR

The accuracy of a classifier, whether it is an ensemble or not, is directly influenced by the training data used in learning. In remote sensing, training data mislabeling is inevitable and faces a major challenge. This article proposes a versatile data cleaning, which handles the mislabeling problem by exploiting the ensemble concepts for identifying and then eliminating or correcting the mislabeled