Comparison of ice/water classification in Fram Strait from C- and L-band SAR imagery

Aldenhoff, Wiebke; Heuzé, Céline; Eriksson, Leif E. B.

doi:10.1017/aog.2018.7

Cited by 34 publications

(22 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although the European Space Agency (ESA) has provided calibrated noise vectors for the noise equivalent sigma nought (NESZ) subtraction, the residual noise contribution is still significant in the extra wide (EW) swath mode ground range detected medium (GRDM) resolution product considering the relatively narrow backscattering distribution of the cross-polarization channels. This calibration issue has been reported in [1]- [3].…”

mentioning

confidence: 75%

Textural Noise Correction for Sentinel-1 TOPSAR Cross-Polarization Channel Images

Park¹,

Won

Korosov³

et al. 2019

IEEE Trans. Geosci. Remote Sensing

View full text Add to dashboard Cite

The thermal noise-induced distortions in a Sentinel-1 terrain observation with progressive scans synthetic aperture radar image cannot be corrected by the noise equivalent sigma nought (NESZ) subtraction only. Since the thermal noise is scaled during synthetic aperture radar processing, it resides not only as an additive noise in each pixel but also as a multiplicative noise in the interpixel contrast. In this paper, we investigate the noise characteristics and propose an efficient method for the multiplicative textural noise correction. The core ideas are to find the optimal coefficient of the noise-induced standard deviation (SD) and model the noise contribution to the local SD as a function of the NESZ and the signal-to-noise ratio. Denoising is accomplished by a subwindow-wise adaptive rescaling of the pixel values. The improvements in the first-and second-order statistical textural features demonstrate the effectiveness of the proposed method.

show abstract

mentioning

confidence: 75%

Textural Noise Correction for Sentinel-1 TOPSAR Cross-Polarization Channel Images

Park¹,

Won

Korosov³

et al. 2019

IEEE Trans. Geosci. Remote Sensing

View full text Add to dashboard Cite

show abstract

“…The quasi-linear slopes in the plane of incidence angle versus sigma nought in decibel scale for typical first-year ice are reported as -0.24 and -0.16 dB/degree for HH-and HV-polarization, respectively (Mäkynen and Karvonen, 2017). To normalize the backscattering intensity for all swath range, these slopes are compensated for or used as input layer in serveral ice classification algorithms in literature (Liu et al, 2015;Zakhvatkina et al, 2013Zakhvatkina et al, , 2017Karvonen, 2014Karvonen, , 2017Aldenhoff et al, 2018). Although the angular dependency is not a system-dependent variable but is governed by physical characteristics of the backscattered surface, the numbers need to be reassessed because the estimations of Mäkynen and Karvonen (2017) might have been affected by the residual thermal noise which used to be very strong before the ESA has updated the noise removal scheme in 2018 (Miranda, 2018).…”

Section: Incidence Angle Correctionmentioning

confidence: 99%

Classification of Sea Ice Types in Sentinel-1 SAR images

Park¹,

Korosov²,

Babiker³

et al. 2019

Preprint

View full text Add to dashboard Cite

Abstract. A new Sentinel-1 image-based sea ice classification algorithm is proposed to support automated ice charting. Previous studies mostly rely on manual work in selecting training and validation data. We show that the use of readily available ice charts from an operational ice services allow to automate selection of large amount of training/testing data void of biased, subjective decisions. The proposed scheme has two phases: training and operational. Both phases start from removal of thermal, scalloping and textural noise from Sentinel-1 data and calculation of gray level co-occurrence matrix and Haralick texture features in a sliding window. In the training phase, the weekly ice charts are reprojected into the SAR image geometry. A random forest classifier is trained with the texture features on input and labels from the rasterized ice charts on output. Then the trained classifier is directly applied to the texture features from Sentinel-1 images in operational manner. Test results from two winter season dataset acquired over the Fram Strait and Barents Sea area showed that the classifier is capable of retrieving 3 generalized cover types (ice free, integrated first-year ice, old ice) with overall accuracies of 85 % and 5 cover types (ice free, new ice, young ice, first-year ice, old ice) with accuracy of 58 %. The errors are attributed both to incorrect manual classification on the ice charts and to the automated algorithm. We demonstrate the potential for near-real time service of ice type classification through an example of ice maps made from daily mosaiced images.

show abstract

“…In several recent studies automated algorithms have been developed for dual-polarization C-band SAR image segmentation and ice/water classification [4][5][6][7][8][9][10]; for retrieval of ice concentration [11,12]; for classification of several ice types [13][14][15]. It is emphasized in these studies that a threshold based classification of ice types on single frequency dual-polarization SAR data is not possible and more advanced methods employing either full polarization, or multi-frequency, or texture analysis are required.…”

Section: Introductionmentioning

confidence: 99%

Classification of Sea Ice Types in Sentinel-1 SAR Data Using Convolutional Neural Networks

Boulze¹,

Korosov²,

Brajard³

2020

Remote Sensing

100

View full text Add to dashboard Cite

A new algorithm for classification of sea ice types on Sentinel-1 Synthetic Aperture Radar (SAR) data using a convolutional neural network (CNN) is presented. The CNN is trained on reference ice charts produced by human experts and compared with an existing machine learning algorithm based on texture features and random forest classifier. The CNN is trained on two datasets in 2018 and 2020 for retrieval of four classes: ice free, young ice, first-year ice and old ice. The accuracy of our classification is 90.5% for the 2018-dataset and 91.6% for the 2020-dataset. The uncertainty is a bit higher for young ice (85%/76% accuracy in 2018/2020) and first-year ice (86%/84% accuracy in 2018/2020). Our algorithm outperforms the existing random forest product for each ice type. It has also proved to be more efficient in computing time and less sensitive to the noise in SAR data. The code is publicly available.

show abstract

Comparison of ice/water classification in Fram Strait from C- and L-band SAR imagery

Cited by 34 publications

References 46 publications

Textural Noise Correction for Sentinel-1 TOPSAR Cross-Polarization Channel Images

Textural Noise Correction for Sentinel-1 TOPSAR Cross-Polarization Channel Images

Classification of Sea Ice Types in Sentinel-1 SAR images

Classification of Sea Ice Types in Sentinel-1 SAR Data Using Convolutional Neural Networks

Contact Info

Product

Resources

About