Wetlands in the Mississippi River Delta Plain (MRDP) have been suffering from a high rate of land loss. Sediment cores have been drilled into the wetlands to understand their growth and degradation ...and to provide subsurface information for the coastal protection and restoration projects. However, few three-dimensional (3D) stratigraphy models have been developed for the wetlands on a regional scale, due to difficulties in correlating large amount of spatial scattered subsurface data and integrated visualization of stratigraphic features and topobathymetric features. In this study, a 3D model was constructed in the Lower Barataria Bay (LBB) and the Lower Breton Sound (LBS), covering an area of 190 km
2
and extending from 0.5 to − 4 m in elevation. Sediment composition (sand%, silt%, and clay%) was spatially interpolated, using a compositional kriging method, extended from ordinary kriging by a log-ratio transformation. Instead of visualizing three composition components independently, sediment composition was translated into sediment texture to be visualized as sediment types. Modeling results intuitively show spatial distribution of stratigraphic features and their spatial relationships with topobathymetric features such as marsh surface, river channel, and dredging channels. Results show a silty depositional package, which consists of crevasse splays and periodic overbank flooding deposits, made up the largest portion of the wetlands. A clayey blanket is observed to cover most part of the LBB and landward side of the LBS. A large area of clayey blanket in the seaward part of the LBS has apparently been eroded away, which is likely caused by coastal reworking processes.
We used seven 210Pb‐dated sediment cores from the Gaer Arm in the Doubtful Sound fjord complex, Fiordland, New Zealand to evaluate organic carbon (OC) dynamics in a temperate fjord‐head delta. The ...highly dynamic spatial features of this delta were clearly evident in the observed sediment properties such as mass accumulation rates that varied by a factor of 14, sediment grain size by a factor 5, and sedimentary OC content by a factor 6. Low lignin concentrations (e.g., 2.95 mg (100 mg OC)−1) and syringic/vanillic ratios of lignin phenols (S/V; e.g., 0.44) at the upper deltaic stations were representative of substantial autochthonous OC contributions to delta sediments. Significantly higher acid/aldehyde ratios of vanillic phenols (Ad/Al)v at the deltaic stations (0.45–0.82) than the surface grabs (0.26–0.30) indicated rapid degradation of OC within the delta. Despite being a “hot spot” for OC oxidation, the delta likely improves OC preservation in the adjacent fjord by filtering out coarse‐grained particles and exporting fine‐grained particles to fjord sediments. Our results showed that fjord‐head deltas can influence sedimentation and OC dynamics in select regions of fjords and thus warrant more examination of fjord‐head processes, particularly in areas where they are expanding. In particular, as Earth warms and glaciers retreat, the newly exposed fjord‐head platforms in high‐latitude environments may evolve into similar “hot spots” of OC oxidation, thereby altering the dynamics of OC burial in these systems.
Key Points
Variations in organic carbon content and sediment grain size are linked with hydrodynamic sorting of particles
Lignin biomarker indices support local source inputs of OC on the fjord‐head delta
Biomarker and organic carbon accumulation rates suggest that fjord‐head deltas are likely hot spots for carbon oxidation
Under the artificial intelligence adversarial environment, deep neural networks have an obvious vulnerability to adversarial samples. To improve the robustness of the model in the adversarial ...environment, a deep neural network model robustness optimization method AdvRob is proposed. Firstly, the target model is transformed into a feature pyramid structure, and then the prior knowledge of latent features is used to generate more aggressive adversarial samples for adversarial training. Experiments on the MNIST and CIFAR-10 datasets show that the adversarial samples generated by using latent features have a higher attack success rate, more diversity and stronger transferability than the AdvGAN method. Under high disturbances, on the MNIST dataset, compared with original model, the defensive ability of the AdvRob method against FGSM and JSMA attacks has been improved by at least 4 times, and the defensive ability against PGD, BIM, and C&W attacks has been improved by at least 10 times. Compared with original mode
Category-level 6D pose estimation, aiming to predict the location and orientation of unseen object instances, is fundamental to many scenarios such as robotic manipulation and augmented reality, yet ...still remains unsolved. Precisely recovering instance 3D model in the canonical space and accurately matching it with the observation is an essential point when estimating 6D pose for unseen objects. In this paper, we achieve accurate category-level 6D pose estimation via cascaded relation and recurrent reconstruction networks. Specifically, a novel cascaded relation network is dedicated for advanced representation learning to explore the complex and informative relations among instance RGB image, instance point cloud and category shape prior. Furthermore, we design a recurrent reconstruction network for iterative residual refinement to progressively improve the reconstruction and correspondence estimations from coarse to fine. Finally, the instance 6D pose is obtained leveraging the estimated dense correspondences between the instance point cloud and the reconstructed 3D model in the canonical space. We have conducted extensive experiments on two well-acknowledged benchmarks of category-level 6D pose estimation, with significant performance improvement over existing approaches. On the representatively strict evaluation metrics of \(3D_{75}\) and \(5^{\circ}2 cm\), our method exceeds the latest state-of-the-art SPD by \(4.9\%\) and \(17.7\%\) on the CAMERA25 dataset, and by \(2.7\%\) and \(8.5\%\) on the REAL275 dataset. Codes are available at https://wangjiaze.cn/projects/6DPoseEstimation.html.
Coning algorithm is the kernel during attitude algorithm design in strapdown inertial navigation. A new approach is presented for strapdown coning algorithm design based on batch-processing. Unlike ...previous time or frequency Taylor series expansion techniques, the new method achieves the rotation vector through discrete Fourier transforms(DFT) and inverse Fourier transforms(IFT) of the inputs. This methodology allows the coning algorithm to be design in frequency, which might be a novel aspect to the coning algorithm. Performance of the new algorithm design technique is evaluated by comparison with previous time approaches, typically four-sample algorithm, in pure coning environments. Simulation results indicate that the new method reduces the coning error when the coning frequency is under 100Hz. The new method might be an alternate method during ship navigation where the coning frequency is not very high.
Data augmentation is an effective regularization strategy for mitigating overfitting in deep neural networks, and it plays a crucial role in 3D vision tasks, where the point cloud data is relatively ...limited. While mixing-based augmentation has shown promise for point clouds, previous methods mix point clouds either on block level or point level, which has constrained their ability to strike a balance between generating diverse training samples and preserving the local characteristics of point clouds. The significance of each part component of the point clouds has not been fully considered, as not all parts contribute equally to the classification task, and some parts may contain unimportant or redundant information. To overcome these challenges, we propose PointPatchMix, a novel approach that mixes point clouds at the patch level and integrates a patch scoring module to generate content-based targets for mixed point clouds. Our approach preserves local features at the patch level, while the patch scoring module assigns targets based on the content-based significance score from a pre-trained teacher model. We evaluate PointPatchMix on two benchmark datasets including ModelNet40 and ScanObjectNN, and demonstrate significant improvements over various baselines in both synthetic and real-world datasets, as well as few-shot settings. With Point-MAE as our baseline, our model surpasses previous methods by a significant margin. Furthermore, our approach shows strong generalization across various point cloud methods and enhances the robustness of the baseline model. Code is available at https://jiazewang.com/projects/pointpatchmix.html.
Data augmentation has proven to be a vital tool for enhancing the
generalization capabilities of deep learning models, especially in the context
of 3D vision where traditional datasets are often ...limited. Despite previous
advancements, existing methods primarily cater to unimodal data scenarios,
leaving a gap in the augmentation of multimodal triplet data, which integrates
text, images, and point clouds. Simultaneously augmenting all three modalities
enhances diversity and improves alignment across modalities, resulting in more
comprehensive and robust 3D representations. To address this gap, we propose
TripletMix, a novel approach to address the previously unexplored issue of
multimodal data augmentation in 3D understanding. TripletMix innovatively
applies the principles of mixed-based augmentation to multimodal triplet data,
allowing for the preservation and optimization of cross-modal connections. Our
proposed TripletMix combines feature-level and input-level augmentations to
achieve dual enhancement between raw data and latent features, significantly
improving the model's cross-modal understanding and generalization capabilities
by ensuring feature consistency and providing diverse and realistic training
samples. We demonstrate that TripletMix not only improves the baseline
performance of models in various learning scenarios including zero-shot and
linear probing classification but also significantly enhances model
generalizability. Notably, we improved the zero-shot classification accuracy on
ScanObjectNN from 51.3 percent to 61.9 percent, and on Objaverse-LVIS from 46.8
percent to 51.4 percent. Our findings highlight the potential of multimodal
data augmentation to significantly advance 3D object recognition and
understanding.
The outbreak of the COVID-19 pneumonia in 2019 has caused great damage to the world economy. With the continuous growth of the amount of data, using machine learning algorithm to analyze and predict ...the economic development of different countries and regions is a hot topic in recent years. In this paper, three machine learning algorithms (XGBoost, AdaBoost and random forest algorithms) are coupled together, and a new algorithm is proposed. Combined with data preprocessing and fine feature engineering processing, GDP values of different countries and regions are predicted. Experimental results show that our coupled method has better performance than each single machine learning algorithm used in this paper. Specifically, the MSE metrics of proposed model is 1.64%, 3.69% and 8.95% lower than XGBoost, AdaBoost and Random Forest algorithm, respectively. In addition, we also study the correlation coefficient between features and get some constructive guidance to improve the accuracy of the algorithm and restrain the further development of the epidemic situation.
Weather forecasting plays a critical role in various sectors, driving
decision-making and risk management. However, traditional methods often
struggle to capture the complex dynamics of ...meteorological systems,
particularly in the presence of high-resolution data. In this paper, we propose
the Spatial-Frequency Attention Network (SFANet), a novel deep learning
framework designed to address these challenges and enhance the accuracy of
spatiotemporal weather prediction. Drawing inspiration from the limitations of
existing methodologies, we present an innovative approach that seamlessly
integrates advanced token mixing and attention mechanisms. By leveraging both
pooling and spatial mixing strategies, SFANet optimizes the processing of
high-dimensional spatiotemporal sequences, preserving inter-component
relational information and modeling extensive long-range relationships. To
further enhance feature integration, we introduce a novel spatial-frequency
attention module, enabling the model to capture intricate cross-modal
correlations. Our extensive experimental evaluation on two distinct datasets,
the Storm EVent ImageRy (SEVIR) and the Institute for Climate and Application
Research (ICAR) - El Ni\~{n}o Southern Oscillation (ENSO) dataset, demonstrates
the remarkable performance of SFANet. Notably, SFANet achieves substantial
advancements over state-of-the-art methods, showcasing its proficiency in
forecasting precipitation patterns and predicting El Ni\~{n}o events.