scholarly journals Neural Network Based Kalman Filters for the Spatio-Temporal Interpolation of Satellite-Derived Sea Surface Temperature

2018 ◽  
Vol 10 (12) ◽  
pp. 1864 ◽  
Author(s):  
Said Ouala ◽  
Ronan Fablet ◽  
Cédric Herzet ◽  
Bertrand Chapron ◽  
Ananda Pascual ◽  
...  
Keyword(s):  
Neural Network ◽  
Data Assimilation ◽  
Kalman Filters ◽  
State Of The Art ◽  
Sea Surface ◽  
Data Driven ◽  
Optimal Interpolation ◽  
Computationally Efficient ◽  
Temporal Interpolation ◽  
Spatio Temporal

The forecasting and reconstruction of oceanic dynamics is a crucial challenge. While model driven strategies are still the state-of-the-art approaches in the reconstruction of spatio-temporal dynamics. The ever increasing availability of data collections in oceanography raised the relevance of data-driven approaches as computationally efficient representations of spatio-temporal fields reconstruction. This tools proved to outperform classical state-of-the-art interpolation techniques such as optimal interpolation and DINEOF in the retrievement of fine scale structures while still been computationally efficient comparing to model based data assimilation schemes. However, coupling this data-driven priors to classical filtering schemes limits their potential representativity. From this point of view, the recent advances in machine learning and especially neural networks and deep learning can provide a new infrastructure for dynamical modeling and interpolation within a data-driven framework. In this work we adress this challenge and develop a novel Neural-Network-based (NN-based) Kalman filter for spatio-temporal interpolation of sea surface dynamics. Based on a data-driven probabilistic representation of spatio-temporal fields, our approach can be regarded as an alternative to classical filtering schemes such as the ensemble Kalman filters (EnKF) in data assimilation. Overall, the key features of the proposed approach are two-fold: (i) we propose a novel architecture for the stochastic representation of two dimensional (2D) geophysical dynamics based on a neural networks, (ii) we derive the associated parametric Kalman-like filtering scheme for a computationally-efficient spatio-temporal interpolation of Sea Surface Temperature (SST) fields. We illustrate the relevance of our contribution for an OSSE (Observing System Simulation Experiment) in a case-study region off South Africa. Our numerical experiments report significant improvements in terms of reconstruction performance compared with operational and state-of-the-art schemes (e.g., optimal interpolation, Empirical Orthogonal Function (EOF) based interpolation and analog data assimilation).

scholarly journals Data-Driven Interpolation of Sea Surface Suspended Concentrations Derived from Ocean Colour Remote Sensing Data

2021 ◽  
Vol 13 (17) ◽  
pp. 3537
Author(s):  
Jean-Marie Vient ◽  
Frederic Jourdin ◽  
Ronan Fablet ◽  
Baptiste Mengual ◽  
Ludivine Lafosse ◽  
...  
Keyword(s):  
Neural Network ◽  
Data Assimilation ◽  
State Of The Art ◽  
Satellite Observations ◽  
Data Driven ◽  
Network Approach ◽  
Satellite Measurements ◽  
Neural Network Approach ◽  
Reconstruction Performance

Due to complex natural and anthropogenic interconnected forcings, the dynamics of suspended sediments within the ocean water column remains difficult to understand and monitor. Numerical models still lack capabilities to account for the variabilities depicted by in situ and satellite-derived datasets. Besides, the irregular space-time sampling associated with satellite sensors make crucial the development of efficient interpolation methods. Optimal Interpolation (OI) remains the state-of-the-art approach for most operational products. Due to the large increase of both in situ and satellite measurements more and more available information is coming from in situ and satellite measurements, as well as from simulation models. The emergence of data-driven schemes as possibly relevant alternatives with increased capabilities to recover finer-scale processes. In this study, we investigate and benchmark three state-of-the-art data-driven schemes, namely an EOF-based technique, an analog data assimilation scheme, and a neural network approach, with an OI scheme. We rely on an Observing System Simulation Experiment based on high-resolution numerical simulations and simulated satellite observations using real satellite sampling patterns. The neural network approach, which relies on variational data assimilation formulation for the interpolation problem, clearly outperforms both the OI and the other data-driven schemes, both in terms of reconstruction performance and of a greater ability to recover high-frequency events. We further discuss how these results could transfer to real data, as well as to other problems beyond interpolation issues, especially short-term forecasting problems from partial satellite observations.

Reconstructing Sea Surface Dynamics Using a Linear Koopman Kalman Filter

2021 ◽  
Author(s):  
Said Ouala ◽  
Ronan Fablet ◽  
Ananda Pascual Pascual ◽  
Bertrand Chapron ◽  
Fabrice Collard ◽  
...  
Keyword(s):  
Kalman Filter ◽  
Data Assimilation ◽  
Linear Representation ◽  
Sea Surface ◽  
Dynamical Model ◽  
Optimal Interpolation ◽  
Model Based ◽  
Trade Offs ◽  
Non Linear ◽  
Spatio Temporal

<p>Spatio-temporal interpolation applications are important in the context of ocean surface modeling. Current state-of-the-art techniques typically rely either on optimal interpolation or on model-based approaches which explicitly exploit a dynamical model. While the optimal interpolation suffers from smoothing issues making it unreliable in retrieving fine-scale variability, the selection and parametrization of a dynamical model, when considering model-based data assimilation strategies, remains a complex issue since several trade-offs between the model's complexity and its applicability in sea surface data assimilation need to be carefully addressed. For these reasons, deriving new data assimilation architectures that can perfectly exploit the observations and the current advances in signal processing, modeling and artificial intelligence is crucial.</p><p>In this work, we explore new advances in data-driven data assimilation to exploit the classical Kalman filter in the interpolation of spatio-temporal fields. The proposed algorithm is written in an end-to-end differentiable setting in order to allow for the learning of the linear dynamical model from a data assimilation cost. Furthermore, the linear model is formulated on a space of observables, rather than the space of observations, which allows for perfect replication of non-linear dynamics when considering periodic and quasi-periodic limit sets and providing a decent (short-term) forecast of chaotic ones. One of the main advantages of the proposed architecture is its simplicity since it utilises a linear representation coupled with a Kalman filter. Interestingly, our experiments show that exploiting such a linear representation leads to better data assimilation when compared to non-linear filtering techniques, on numerous applications, including the sea level anomaly reconstruction from satellite remote sensing observations.</p>

An application of a data assimilation method based on the diffusion stochastic process theory using altimetry data in Atlantic

2016 ◽  
Vol 31 (3) ◽  
Author(s):  
Konstantin P. Belyaev ◽  
Andrey A. Kuleshov ◽  
Clemente A. S. Tanajura
Keyword(s):  
Stochastic Process ◽  
Data Assimilation ◽  
Interpolation Method ◽  
Planck Equation ◽  
Ocean Model ◽  
Sea Surface ◽  
Process Theory ◽  
Optimal Interpolation ◽  
Computationally Efficient ◽  
The Impact

AbstractA data assimilation (DA) method based on the application of the diffusion stochastic process theory, particularly, of the Fokker-Planck equation, is considered. The method was introduced in the previous works; however, it is substantially modified and extended to the multivariate case in the current study. For the first time, the method is here applied to the assimilation of sea surface height anomalies (SSHA) into the Hybrid Coordinate Ocean Model (HYCOM) over the Atlantic Ocean. The impact of assimilation of SSHA is investigated and compared with the assimilation by an Ensemble Optimal Interpolation method (EnOI). The time series of the analyses produced by both assimilation methods are evaluated against the results from a free model run without assimilation. This study shows that the proposed assimilation technique has some advantages in comparison with EnOI analysis. Particularly, it is shown that it provides slightly smaller error and is computationally efficient. The method may be applied to assimilate other data such as observed sea surface temperature and vertical profiles of temperature and salinity.

scholarly journals PedNet: A Spatio-Temporal Deep Convolutional Neural Network for Pedestrian Segmentation

2018 ◽  
Vol 4 (9) ◽  
pp. 107 ◽  
Author(s):  
Mohib Ullah ◽  
Ahmed Mohammed ◽  
Faouzi Alaya Cheikh
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Performance Metrics ◽  
State Of The Art ◽  
Temporal Information ◽  
Feature Maps ◽  
Current Frame ◽  
Low Level ◽  
Art Methods ◽  
Spatio Temporal

Articulation modeling, feature extraction, and classification are the important components of pedestrian segmentation. Usually, these components are modeled independently from each other and then combined in a sequential way. However, this approach is prone to poor segmentation if any individual component is weakly designed. To cope with this problem, we proposed a spatio-temporal convolutional neural network named PedNet which exploits temporal information for spatial segmentation. The backbone of the PedNet consists of an encoder–decoder network for downsampling and upsampling the feature maps, respectively. The input to the network is a set of three frames and the output is a binary mask of the segmented regions in the middle frame. Irrespective of classical deep models where the convolution layers are followed by a fully connected layer for classification, PedNet is a Fully Convolutional Network (FCN). It is trained end-to-end and the segmentation is achieved without the need of any pre- or post-processing. The main characteristic of PedNet is its unique design where it performs segmentation on a frame-by-frame basis but it uses the temporal information from the previous and the future frame for segmenting the pedestrian in the current frame. Moreover, to combine the low-level features with the high-level semantic information learned by the deeper layers, we used long-skip connections from the encoder to decoder network and concatenate the output of low-level layers with the higher level layers. This approach helps to get segmentation map with sharp boundaries. To show the potential benefits of temporal information, we also visualized different layers of the network. The visualization showed that the network learned different information from the consecutive frames and then combined the information optimally to segment the middle frame. We evaluated our approach on eight challenging datasets where humans are involved in different activities with severe articulation (football, road crossing, surveillance). The most common CamVid dataset which is used for calculating the performance of the segmentation algorithm is evaluated against seven state-of-the-art methods. The performance is shown on precision/recall, F 1 , F 2 , and mIoU. The qualitative and quantitative results show that PedNet achieves promising results against state-of-the-art methods with substantial improvement in terms of all the performance metrics.

scholarly journals Regularized Training and Tight Certification for Randomized Smoothed Classifier with Provable Robustness

2020 ◽  
Vol 34 (04) ◽  
pp. 3858-3865
Author(s):  
Huijie Feng ◽  
Chunpeng Wu ◽  
Guoyang Chen ◽  
Weifeng Zhang ◽  
Yang Ning
Keyword(s):  
Neural Network ◽  
High Probability ◽  
Deep Neural Network ◽  
State Of The Art ◽  
Computationally Efficient ◽  
Base Classifier ◽  
Training Scheme ◽  
Adversarial Training ◽  
Gaussian Perturbation ◽  
Probabilistic Robustness

Recently smoothing deep neural network based classifiers via isotropic Gaussian perturbation is shown to be an effective and scalable way to provide state-of-the-art probabilistic robustness guarantee against ℓ2 norm bounded adversarial perturbations. However, how to train a good base classifier that is accurate and robust when smoothed has not been fully investigated. In this work, we derive a new regularized risk, in which the regularizer can adaptively encourage the accuracy and robustness of the smoothed counterpart when training the base classifier. It is computationally efficient and can be implemented in parallel with other empirical defense methods. We discuss how to implement it under both standard (non-adversarial) and adversarial training scheme. At the same time, we also design a new certification algorithm, which can leverage the regularization effect to provide tighter robustness lower bound that holds with high probability. Our extensive experimentation demonstrates the effectiveness of the proposed training and certification approaches on CIFAR-10 and ImageNet datasets.

scholarly journals Multi-Term Attention Networks for Skeleton-Based Action Recognition

2020 ◽  
Vol 10 (15) ◽  
pp. 5326
Author(s):  
Xiaolei Diao ◽  
Xiaoqiang Li ◽  
Chen Huang
Keyword(s):  
Neural Network ◽  
Time Scales ◽  
Action Recognition ◽  
State Of The Art ◽  
Attention Networks ◽  
Weighted Fusion ◽  
Temporal Features ◽  
Benchmark Datasets ◽  
Spatio Temporal ◽  
Different Time Scales

The same action takes different time in different cases. This difference will affect the accuracy of action recognition to a certain extent. We propose an end-to-end deep neural network called “Multi-Term Attention Networks” (MTANs), which solves the above problem by extracting temporal features with different time scales. The network consists of a Multi-Term Attention Recurrent Neural Network (MTA-RNN) and a Spatio-Temporal Convolutional Neural Network (ST-CNN). In MTA-RNN, a method for fusing multi-term temporal features are proposed to extract the temporal dependence of different time scales, and the weighted fusion temporal feature is recalibrated by the attention mechanism. Ablation research proves that this network has powerful spatio-temporal dynamic modeling capabilities for actions with different time scales. We perform extensive experiments on four challenging benchmark datasets, including the NTU RGB+D dataset, UT-Kinect dataset, Northwestern-UCLA dataset, and UWA3DII dataset. Our method achieves better results than the state-of-the-art benchmarks, which demonstrates the effectiveness of MTANs.

scholarly journals Data-Driven Interpolation of Sea Level Anomalies Using Analog Data Assimilation

2019 ◽  
Vol 11 (7) ◽  
pp. 858 ◽  
Author(s):  
Redouane Lguensat ◽  
Phi Huynh Viet ◽  
Miao Sun ◽  
Ge Chen ◽  
Tian Fenglin ◽  
...  
Keyword(s):  
Data Assimilation ◽  
Sea Level ◽  
Large Scale ◽  
Empirical Orthogonal Functions ◽  
Data Driven ◽  
Optimal Interpolation ◽  
Observation System ◽  
Orthogonal Functions ◽  
Sea Level Anomalies ◽  
Recent Developments

From the recent developments of data-driven methods as a means to better exploit large-scale observation, simulation and reanalysis datasets for solving inverse problems, this study addresses the improvement of the reconstruction of higher-resolution Sea Level Anomaly (SLA) fields using analog strategies. This reconstruction is stated as an analog data assimilation issue, where the analog models rely on patch-based and Empirical Orthogonal Functions (EOF)-based representations to circumvent the curse of dimensionality. We implement an Observation System Simulation Experiment (OSSE) in the South China Sea. The reported results show the relevance of the proposed framework with a significant gain in terms of Root Mean Square Error (RMSE) for scales below 100 km. We further discuss the usefulness of the proposed analog model as a means to exploit high-resolution model simulations for the processing and analysis of current and future satellite-derived altimetric data with regard to conventional interpolation schemes, especially optimal interpolation.

scholarly journals A STUDY OF THE IMPACT OF ALTIMETRY DATA ASSIMILATION ON SHORT-TERM PREDICTABILITY OF THE HYCOM OCEAN MODEL IN REGIONS OF THE TROPICAL AND SOUTH ATLANTIC OCEAN

2013 ◽  
Vol 31 (2) ◽  
pp. 271 ◽  
Author(s):  
Leonardo Nascimento Lima ◽  
Clemente Augusto Souza Tanajura
Keyword(s):  
Data Assimilation ◽  
Sea Surface Height ◽  
Ocean Model ◽  
South Atlantic ◽  
Sea Surface ◽  
Optimal Interpolation ◽  
Altimetry Data ◽  
Hybrid Coordinate Ocean Model ◽  
Along Track ◽  
The Impact

ABSTRACT. In this study, assimilation of Jason-1 and Jason-2 along-track sea level anomaly (SLA) data was conducted in a region of the tropical and South Atlantic (7◦N-36◦S, 20◦W up to the Brazilian coast) using an optimal interpolation method and the HYCOM (Hybrid Coordinate Ocean Model). Four 24 h-forecast experiments were performed daily from January 1 until March 31, 2011 considering different SLA assimilation data windows (1 day and 2 days) and different coefficients in the parameterization of the SLA covariance matrix model. The model horizontal resolution was 1/12◦ and the number of vertical layers was 21. The SLA analyses added to the mean sea surface height were projected to the subsurface with the Cooper & Haines (1996) scheme. The results showed that the experiment with 2-day window of along-track data and with specific parameterizations of the model SLA covariance error for sub-regions of the METAREA V was the most accurate. It completely reconstructed the model sea surface height and important improvements in the circulation were produced. For instance, there was a substantial improvement in the representation of the Brazil Current and North Brazil Undercurrent. However, since no assimilation of vertical profiles of temperature and salinity and of sea surface temperature was performed, the methodology employed here should be considered only as a step towards a high quality analysis for operational forecasting systems.   Keywords: data assimilation, optimal interpolation, Cooper & Haines scheme, altimetry data.   RESUMO. Neste estudo, a assimilação de dados de anomalia da altura da superfície do mar (AASM) ao longo da trilha dos satélites Jason-1 e Jason-2 foi conduzida em uma região do Atlântico tropical e Sul (7◦N-36◦S, 20◦W até a costa do Brasil) com o método de interpolação ótima e o modelo oceânico HYCOM (Hybrid Coordinate Ocean Model). Foram realizados quatro experimentos de previsão de 24 h entre 1 de janeiro e 31 de março de 2011, considerando diferentes janelas de assimilação de AASM (1 dia e 2 dias) e diferentes coeficientes na parametrização da matriz de covariância dos erros de AASM do modelo. A resolução horizontal empregada no HYCOM foi 1/12◦ para 21 camadas verticais. As correções de altura da superfície do mar devido à assimilação de AASM foram projetadas abaixo da camada de mistura através da técnica de Cooper & Haines (1996). Os resultados mostraram que o experimento com assimilação de dados ao longo da trilha dos satélites com a janela de 2 dias e com parametrizações da matriz de covariância específicas para sub-regiões da METAREA V foi o mais acurado. Ele reconstruiu completamente a altura da superfície do mar e também proporcionou melhorias na circulação oceânica reproduzida pelo modelo. Por exemplo, houve substancial melhoria da representação nos campos da Corrente do Brasil e Subcorrente Norte do Brasil. Entretanto, tendo em vista que não foi realizada a assimilação de perfis verticais de temperatura e de salinidade e da temperatura da superfície do mar, a metodologia apresentada deve ser considerada apenas como um passo na conquista de uma análise oceânica e de um sistema previsor de qualidade para fins operacionais.   Palavras-chave: assimilação de dados, interpolação ótima, técnica de Cooper & Haines, dados de altimetria.

Sign in / Sign up

Export Citation Format

Share Document