Low-rankness transfer for denoising Sentinel-1 SAR images

Brodu, Nicolas

doi:10.1109/isivc.2018.8709210

Cited by 2 publications

(2 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is hard to balance performances between speckle noise reduction and detail preservation. In general, traditional despecking methods [277], [278], (such as multilook processing [262], filtering [263], [264], [282], [284], blocking matching 3D (BM3D) [265], [280], wavelet-based [266]- [269], [282], separated component-based [270]), transform the multiplicative noise into additive noise by logarithm operation of observed data. These methods, however, can introduce more or less bias into denoised image.…”

Section: A Sar Images Processingmentioning

confidence: 99%

A Comprehensive Survey of Machine Learning Applied to Radar Signal Processing

Lang,

Fu,

Martorella

et al. 2020

Preprint

View full text Add to dashboard Cite

Modern radar systems have high requirements in terms of accuracy, robustness and real-time capability when operating on increasingly complex electromagnetic environments. Traditional radar signal processing (RSP) methods have shown some limitations when meeting such requirements, particularly in matters of target classification. With the rapid development of machine learning (ML), especially deep learning, radar researchers have started integrating these new methods when solving RSP-related problems. This paper aims at helping researchers and practitioners to better understand the application of ML techniques to RSP-related problems by providing a comprehensive, structured and reasoned literature overview of ML-based RSP techniques. This work is amply introduced by providing general elements of ML-based RSP and by stating the motivations behind them. The main applications of ML-based RSP are then analysed and structured based on the application field. This paper then concludes with a series of open questions and proposed research directions, in order to indicate current gaps and potential future solutions and trends.

show abstract

Section: A Sar Images Processingmentioning

confidence: 99%

A Comprehensive Survey of Machine Learning Applied to Radar Signal Processing

Lang,

Fu,

Martorella

et al. 2020

Preprint

View full text Add to dashboard Cite

show abstract

“…The bands for the 1C TOA and 2A BOA tiles were resampled to 10 m resolution using the SNAP resample tool offered with the main software package designated "simple resolution". The 2A tiles were also resampled using the SNAP "super resolution plug-in "sen2res" (Brodu 2018;European Space Agency 2018). This plug-in produces a more sophisticated 10 m resolution product by using the existing 10 m bands to determine how best to unmix the 20 m and 60 m pixels.…”

Section: Satellite Model Imagery Processingmentioning

confidence: 99%

Comparing Two Species Distribution Models Using Satellite Only and Ready-Made Environmental Variables for the Dakota Skipper (Hesperia dacotae), Interlake Region of Manitoba, Canada

Murray

View full text Add to dashboard Cite

The Dakota skipper, Hesperia dacotae (Skinner, 1911) [Hesperiidae, Lepidopterida] is a rare prairie obligate butterfly with an affinity for anthropogenically undisturbed, grassland habitat with diverse native flora. Persistent threats include habitat fragmentation, destruction, and degradation. These and other threats have caused precipitous population declines and local extirpation across its range. Consequentially, the Dakota skipper is currently listed as Endangered in Canada and Threatened in the United States, and the province of Manitoba. Species distribution models (SDM) are a well-known technique which attempt to predict a species distribution on a landscape. These predictions can then be used to inform conservation actions such as guiding survey effort, land acquisitions, and reintroductions. The objectives of this project were to: 1) Compare Dakota skipper models using freely available high resolution remotely sensed products to those using more traditional environmental predictors. 2) Field validate both models to identify the most accurate model using efficient and economical methods. 3) Address issues of modelling rare species to produce a robust SDM for the Dakota skipper in Manitoba. I found that SDMs built from environmental variables generated from satellite imagery performed comparably to one produced from readily available geospatial information. I also found that field validation was more accurate for evaluating SDMs than purely statistical methods. I also produced usable SDMs for the Dakota skipper in the Interlake. Implications from this study are that the advantages of satellite imagery can be leveraged to create useable SDMs to guide conservation actions. This study also further supports the need to field validate an SDM over relying on model statistical output which can be misleading.

show abstract

Low-rankness transfer for denoising Sentinel-1 SAR images

Cited by 2 publications

References 12 publications

A Comprehensive Survey of Machine Learning Applied to Radar Signal Processing

A Comprehensive Survey of Machine Learning Applied to Radar Signal Processing

Comparing Two Species Distribution Models Using Satellite Only and Ready-Made Environmental Variables for the Dakota Skipper (Hesperia dacotae), Interlake Region of Manitoba, Canada

Contact Info

Product

Resources

About