Classification of the global Sentinel-1 SAR vignettes for ocean surface process studies

Wang, Chen; Tandeo, Pierre; Mouche, Alexis; Stopa, Justin E.; Gressani, Victor; Longépé, Nicolas; Vandemark, D.; Foster, R. C.; Chapron, Bertrand

doi:10.1016/j.rse.2019.111457

Cited by 48 publications

(48 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…These features are distinct enough to be identified visually or automatically using pattern recognition image processing techniques. A geophysical classification model for the WV SAR data, called CMwv, has been successfully developed by fine-tuning the Inception-v3 deep convolutional neural network [11]. This tool is able to distinguish ten pre-defined geophysical phenomena.…”

Section: Roll Vortices On S-1 Wv Imagesmentioning

confidence: 99%

“…This study exploits the global Sentinel-1 (S-1) WaVe mode (WV) SAR data for investigation of ABL roll features. The geophysical classification tool created by a deep learning approach is applied to automatically identify WV SAR images of roll vortices [10,11]. Section 2 of this paper documents the S-1 WV classification tool and collocated meteorological variables.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Characteristics of Marine Atmospheric Boundary Layer Roll Vortices from Sentinel-1 Sar Wave Mode

Wang

Mouche

Foster

et al. 2019

IGARSS 2019 - 2019 IEEE International Geoscience and Remote Sensing Symposium

Self Cite

View full text Add to dashboard Cite

Millions of wave mode synthetic aperture radar (SAR) images are routinely acquired by the ESA's two sentinel-1 satellites every month over the open ocean. These SAR images capture clear imprints of atmospheric boundary layer (ABL) roll vortices. This provides a new and unique opportunity to investigate the characteristics of ABL rolls globally and statistically. In this study, we take advantage of the deep learning classification tool that has been successfully developed to automatically identify ABL rolls. For each SAR image classified with ABL rolls, roll wavelength and orientation are extracted through spectral analysis. Surface meteorological variables are also collocated with each SAR image to address the atmospheric conditions of roll occurrence. Results show that roll vortices are prevalent over the whole ocean and mainly occur in unstable to near-neutral stratification. Roll characteristics follow the theoretical expectation and are in good agreement with previous studies.

show abstract

Section: Roll Vortices On S-1 Wv Imagesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Characteristics of Marine Atmospheric Boundary Layer Roll Vortices from Sentinel-1 Sar Wave Mode

Wang

Mouche

Foster

et al. 2019

IGARSS 2019 - 2019 IEEE International Geoscience and Remote Sensing Symposium

Self Cite

View full text Add to dashboard Cite

show abstract

“…In airborne or spaceborne synthetic aperture radar (SAR) images, there are both large-scale wind streaks of 1-8 km and small-scale wind streaks of 100-500 m [35,36]. Research has revealed that the large-scale wind streaks are caused by the flow of air in the free atmosphere under the action of buoyancy [37], while the small-scale streaks are generated by wind shear stress. According to the imaging mechanism of X-band marine radar, the sea surface wind field appears in the form of small-scale wind streaks in X-band marine radar images.…”

Section: Characteristicsmentioning

confidence: 99%

An Energy Spectrum Algorithm for Wind Direction Retrieval From X-Band Marine Radar Image Sequences

Wang

Qiu

et al. 2021

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

In this study, an energy spectrum (ES) algorithm is proposed to retrieve wind direction from Xband marine radar image sequences. This algorithm is based on utilizing the occlusion area zero-pixel percentage (OZPP) to distinguish rain-free and rain-contaminated radar data. And then the rain-contaminated images are detected and discarded. The effect of radar radial attenuation in radar image sequences is modified by the piecewise fitting technique. Wind direction is determined from rain-free and radial correction data, based on the energy spectrum of small-scale wind streaks. The energy spectrum of small-scale wind streaks is obtained by establishing an energy spectrum scale separation filter. Based on the wind streak characteristics, a two-dimensional fast Fourier transform (FFT) is used to obtain the energy spectrum of radar images. The wind streak characteristics are derived from the distribution of the azimuth normalization radar cross-section (NRCS). The proposed algorithm is tested using data collected from X-band radar images and in-situ anemometer data from the coast of the East China Sea. Compared with the anemometer data, after using the proposed algorithm, the root-mean-square difference for wind direction is 12.13°, which is an acceptable result for engineering application.

show abstract

“…Zhang et al [39] used GoogleNet to classify Gaofen-3 PolSAR images and achieved good results on SAR target image tasks with different sampling intervals. Wang et al [40] developed an automatic classifier of S-1 WV marine SAR target images based on the Inception-v3 structure, showing great potential in marine SAR scene recognition.…”

Section: Cnnmentioning

confidence: 99%

TAI-SARNET: Deep Transferred Atrous-Inception CNN for Small Samples SAR ATR

Ying

Chen

Zhai

et al. 2020

Sensors

View full text Add to dashboard Cite

Since Synthetic Aperture Radar (SAR) targets are full of coherent speckle noise, the traditional deep learning models are difficult to effectively extract key features of the targets and share high computational complexity. To solve the problem, an effective lightweight Convolutional Neural Network (CNN) model incorporating transfer learning is proposed for better handling SAR targets recognition tasks. In this work, firstly we propose the Atrous-Inception module, which combines both atrous convolution and inception module to obtain rich global receptive fields, while strictly controlling the parameter amount and realizing lightweight network architecture. Secondly, the transfer learning strategy is used to effectively transfer the prior knowledge of the optical, non-optical, hybrid optical and non-optical domains to the SAR target recognition tasks, thereby improving the model's recognition performance on small sample SAR target datasets. Finally, the model constructed in this paper is verified to be 97.97% on ten types of MSTAR datasets under standard operating conditions, reaching a mainstream target recognition rate. Meanwhile, the method presented in this paper shows strong robustness and generalization performance on a small number of randomly sampled SAR target datasets.

show abstract

Classification of the global Sentinel-1 SAR vignettes for ocean surface process studies

Cited by 48 publications

References 49 publications

Characteristics of Marine Atmospheric Boundary Layer Roll Vortices from Sentinel-1 Sar Wave Mode

Characteristics of Marine Atmospheric Boundary Layer Roll Vortices from Sentinel-1 Sar Wave Mode

An Energy Spectrum Algorithm for Wind Direction Retrieval From X-Band Marine Radar Image Sequences

TAI-SARNET: Deep Transferred Atrous-Inception CNN for Small Samples SAR ATR

Contact Info

Product

Resources

About