Deep Learning Based Sea Ice Classification with Gaofen-3 Fully Polarimetric SAR Data

Abstract

In this paper, the performance of C-band synthetic aperture radar (SAR) Gaofen-3 (GF-3) quad-polarization Stripmap (QPS) data is assessed for classifying late spring and summer sea ice types. The investigation is based on 18 scenes of GF-3 QPS data acquired in the Arctic Ocean in 2017. In this study, floe ice (FI), brash ice (BI) between floes and open water (OW, ice-free area) were classified based on a mini sea ice residual convolutional network, which we call MSI-ResNet. While investigating the optimal patch size for MSI-ResNet, we found that, as the patch size continues to grow, the classification accuracy first increases and then decreases. A patch size of 31 × 31 was found to achieve the best performance. The performance of classification using different polarization combinations from the QPS data was also assessed. The vertical-vertical (VV) polarization input overestimates the FI category while incorrectly identifying most of the BI as FI. The VH polarization produces a synchronous improvement in FI, BI, and OW discrimination, with a higher overall accuracy and kappa coefficient (91.09% and 0.85, respectively) than the VV polarization (83.37% and 0.70, respectively). The combination of VV and vertical-horizontal (VH) polarizations presents a modest precision improvement for BI and OW together with a slight overestimation for FI. With VV, VH, and horizontal-horizontal (HH) polarization data as the inputs, the user’s accuracy improves to 95.12%, 93.42%, and 95.17% for FI, BI, and OW, respectively. The accuracy was assessed against visual interpretation of the sea ice classes in the images using a stratified sampling method. The application of the MSI-ResNet method to data covering the Beaufort Sea and the north of the Severnaya Zemlya archipelago was found to achieve a high overall accuracy (kappa) of 94.62% (±0.92) and 94.23% (±0.90), respectively. This is similar to the classification accuracy obtained in the Fram Strait. From the results of this study, it is shown that the MSI-ResNet method performs better than the classical support vector machine (SVM) classifier for sea ice discrimination. The GF-3 QPS mode data also show more details in discriminating scattered sea ice floes than the coincident Sentinel-1A Extra Wide (EW) swath mode data.

An Improved Sea Ice Classification Algorithm with Gaofen-3 Dual-Polarization SAR Data Based on Deep Convolutional Neural Networks

“…For example, the classification accuracy of VV polarization data for NI reached 99.2%, which is higher than the classification accuracy of 1.6% using dual-polarization data. A similar situation is observed in [25]. The classification accuracy of FI using fully polarized data is lower than that when using dual-polarization data.…”

“…Tianyu Zhang et al simplified the ResNet network structure and used the full polarization strip mode data of the Gaofen-3 satellite for classification and obtained 94% accuracy. On this basis, it was verified that the network was still applicable to Sentinel-1 data [25].…”

“…Recently, there have been many studies using CNNs for sea ice classification, most of which achieved high accuracy [22][23][24][25]35,36]. However, the algorithms they used are not available, except for [22].…”

“…Current articles on sea ice classification algorithms based on deep learning study the effect of sample size on classification accuracy [22,25,35]. Their research approach was to select several sizes of samples to train the network to obtain the corresponding classification accuracy, using the size of the samples when the accuracy is at its highest for further study.…”

“…Their research approach was to select several sizes of samples to train the network to obtain the corresponding classification accuracy, using the size of the samples when the accuracy is at its highest for further study. Classification algorithms are mostly implemented as a superposition of convolutional and pooling layers [22][23][24][25]35,36]. However, we believe that it is difficult to verify the relationship between sample size and classification accuracy by only a few sets of experiments and the samples lose a large amount of spatial information after a large number of convolution and pooling operations.…”