SAR Speckle Removal Using Hybrid Frequency Modulations

Liu, Shuaiqi; Gao, Lele; Lei, Yu; Wang, Miaohui; Hu, Qi; Ma, Xiaole; Zhang, Yudong

doi:10.1109/tgrs.2020.3014130

Cited by 59 publications

(23 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, in [74] and [75], multiresolution processing is considered. In MSR-Net [75], three dyadic resolution levels are considered, and a multiresolution denoiser with CNN engine is used at each level.…”

Section: B Model-based Despeckling Methods Exploiting DL Toolsmentioning

confidence: 99%

“…The denoiser has a simple encoder-decoder architecture, but the parameters of the bottleneck layer flow to the upper resolution levels by means of a long short-term memory network, so as to enable fast convergence. In [74], instead, the input image is subject to a nonsubsampled shearlet transform with two levels of decomposition. High-frequency coefficients are then denoised by means of a transform-domain method, while only low-frequency coefficients are denoised by using FFDNet in the log-domain.…”

Section: B Model-based Despeckling Methods Exploiting DL Toolsmentioning

confidence: 99%

See 1 more Smart Citation

Deep Learning Methods For Synthetic Aperture Radar Image Despeckling: An Overview Of Trends And Perspectives

Fracastoro,

Magli,

Poggi

et al. 2020

Preprint

View full text Add to dashboard Cite

Synthetic aperture radar (SAR) images are affected by a spatially-correlated and signal-dependent noise called "speckle", which is very severe and may hinder image exploitation. Despeckling is an important task that aims at removing such noise, so as to improve the accuracy of all downstream image processing tasks. The first despeckling methods date back to the 1970's, and several model-based algorithms have been developed in the subsequent years. The field has received growing attention, sparkled by the availability of powerful deep learning models that have yielded excellent performance for inverse problems in image processing. This paper surveys the literature on deep learning methods applied to SAR despeckling, covering both the supervised and the more recent self-supervised approaches. We provide a critical analysis of existing methods with the objective to recognize the most promising research lines, to identify the factors that have limited the success of deep models, and to propose ways forward in an attempt to fully exploit the potential of deep learning for SAR despeckling.

show abstract

Section: B Model-based Despeckling Methods Exploiting DL Toolsmentioning

confidence: 99%

Section: B Model-based Despeckling Methods Exploiting DL Toolsmentioning

confidence: 99%

Deep Learning Methods For Synthetic Aperture Radar Image Despeckling: An Overview Of Trends And Perspectives

Fracastoro,

Magli,

Poggi

et al. 2020

Preprint

View full text Add to dashboard Cite

show abstract

“…As a machine learning algorithm, deep learning has been applied in various fields in recent years, such as image denoising, image fusion, target detection, etc. [16] [17] [18]. Since it uses a multi-layer nonlinear processing unit to process different data, which can better extract the relevant features of EEG, it is more and more used in EEG processing.…”

Section: Introductionmentioning

confidence: 99%

An Attention-Based Wavelet Convolution Neural Network for Epilepsy EEG Classification

Liu

et al. 2022

IEEE Trans. Neural Syst. Rehabil. Eng.

Self Cite

View full text Add to dashboard Cite

As a kind of non-invasive, low-cost, and readily available brain examination, EEG has attached significance to the means of clinical diagnosis of epilepsy. However, the reading of long-term EEG records has brought a heavy burden to neurologists and experts. Therefore, automatic EEG classification for epileptic patients plays an essential role in epilepsy diagnosis and treatment. This paper proposes an Attention Mechanism-based Wavelet Convolution Neural Network for epilepsy EEG classification. Attention Mechanism-based Wavelet Convolution Neural Network firstly uses multi-scale wavelet analysis to decompose the input EEGs to obtain their components in different frequency bands. Then, these decomposed multi-scale EEGs are input into the Convolution Neural Network with an attention mechanism for further feature extraction and classification. The proposed algorithm achieves 98.89% triple classification accuracy on the Bonn EEG database and 99.70% binary classification accuracy on the Bern-Barcelona EEG database. Our experiments prove that the proposed algorithm achieves a state-of-the-art classification effect on epilepsy EEG.

show abstract

“…Their principal advantages are the capacity to obtain information independent of the weather conditions, minimising the atmospheric effects, and the capacity to penetrate through the soil. However, they also present disadvantages compared to optical remote sensing techniques, such as possible speckle effects [ 15 ], that could reduce the quality of SAR imagery and the disturbance of topography [ 16 ] in hilly regions.…”

Section: Introductionmentioning

confidence: 99%

The Vegetation–Climate System Complexity through Recurrence Analysis

Almeida-Ñauñay

Benito

Quemada

et al. 2021

Entropy

View full text Add to dashboard Cite

Multiple studies revealed that pasture grasslands are a time-varying complex ecological system. Climate variables regulate vegetation growing, being precipitation and temperature the most critical driver factors. This work aims to assess the response of two different Vegetation Indices (VIs) to the temporal dynamics of temperature and precipitation in a semiarid area. Two Mediterranean grasslands zones situated in the center of Spain were selected to accomplish this goal. Correlations and cross-correlations between VI and each climatic variable were computed. Different lagged responses of each VIs series were detected, varying in zones, the year’s season, and the climatic variable. Recurrence Plots (RPs) and Cross Recurrence Plots (CRPs) analyses were applied to characterise and quantify the system’s complexity showed in the cross-correlation analysis. RPs pointed out that short-term predictability and high dimensionality of VIs series, as well as precipitation, characterised this dynamic. Meanwhile, temperature showed a more regular pattern and lower dimensionality. CRPs revealed that precipitation was a critical variable to distinguish between zones due to their complex pattern and influence on the soil’s water balance that the VI reflects. Overall, we prove RP and CRP’s potential as adequate tools for analysing vegetation dynamics characterised by complexity.

show abstract

SAR Speckle Removal Using Hybrid Frequency Modulations

Cited by 59 publications

References 44 publications

Deep Learning Methods For Synthetic Aperture Radar Image Despeckling: An Overview Of Trends And Perspectives

Deep Learning Methods For Synthetic Aperture Radar Image Despeckling: An Overview Of Trends And Perspectives

An Attention-Based Wavelet Convolution Neural Network for Epilepsy EEG Classification

The Vegetation–Climate System Complexity through Recurrence Analysis

Contact Info

Product

Resources

About