Arctic Sea Ice Image Super-Resolution Based on Multi-Scale Convolution and Dual-Gating Mechanism

TL;DR

This paper introduces MFM-Net, a novel super-resolution method for Arctic Sea Ice Concentration images that effectively integrates multi-scale, spatial, and channel features to improve resolution and detail recovery.

Contribution

The paper presents MFM-Net, a new deep learning architecture that combines multi-scale feature aggregation with spatial and channel feature integration for Arctic SIC super-resolution.

Findings

MFM-Net outperforms existing methods on Arctic SIC datasets.

The model effectively captures multi-scale features for better resolution.

Experimental results validate the proposed approach's effectiveness.

Abstract

Arctic Sea Ice Concentration (SIC) is the ratio of ice-covered area to the total sea area of the Arctic Ocean, which is a key indicator for maritime activities. Nowadays, we often use passive microwave images to display SIC, but it has low spatial resolution, and most of the existing super-resolution methods of Arctic SIC don't take the integration of spatial and channel features into account and can't effectively integrate the multi-scale feature. To overcome the aforementioned issues, we propose MFM-Net for Arctic SIC super-resolution, which concurrently aggregates multi-scale information while integrating spatial and channel features. Extensive experiments on Arctic SIC dataset from the AMSR-E/AMSR-2 SIC DT-ASI products from Ocean University of China validate the effectiveness of porposed MFM-Net.