ICENET: A Semantic Segmentation Deep Network for River Ice by Fusing Positional and Channel-Wise Attentive Features

Xin Yu,Zeze Lan,Yanning Zhang,Jiaojiao Jin,Chunjiang Li,Minhao Fan,Yafei Wang,Xiuwei Zhang

doi:10.3390/rs12020221

Abstract

River ice monitoring is of great significance for river management, ship navigation and ice hazard forecasting in cold-regions. Accurate ice segmentation is one most important pieces of technology in ice monitoring research. It can provide the prerequisite information for the calculation of ice cover density, drift ice speed, ice cover distribution, change detection and so on. Unmanned aerial vehicle (UAV) aerial photography has the advantages of higher spatial and temporal resolution. As UAV technology has become more popular and cheaper, it has been widely used in ice monitoring. So, we focused on river ice segmentation based on UAV remote sensing images. In this study, the NWPU_YRCC dataset was built for river ice segmentation, in which all images were captured by different UAVs in the region of the Yellow River, the most difficult river to manage in the world. To the best of our knowledge, this is the first public UAV image dataset for river ice segmentation. Meanwhile, a semantic segmentation deep convolution neural network by fusing positional and channel-wise attentive features is proposed for river ice semantic segmentation, named ICENET. Experiments demonstrated that the proposed ICENET outperforms the state-of-the-art methods, achieving a superior result on the NWPU_YRCC dataset.

Highlights

Every winter and spring, river ice freeze-up and break-up are big events in cold regions
We focus on building a Unmanned aerial vehicle (UAV) image dataset for river ice segmentation and designing a novel network architecture which effectively exploits multilevel features for generating high-resolution predictions
Dabboor and Geldsetzer (2013) [19] applied a supervised maximum likelihood (ML) classification approach to classify the river covers as first-year ice (FYI), multiyear ice (MYI) and open water (OW) using synthetic aperture radar (SAR) imagery in the Canadian Arctic

Summary

Introduction

River ice freeze-up and break-up are big events in cold regions. In the period of river ice freeze-up and break-up, a large jumbled accumulation of drift river ice can form an ice jam, which partially blocks a river channel, raises water levels and potentially causes flooding. This kind of flood by large rivers in cold regions is a well-known serious hazard called an ice jam flood. It is potentially more destructive than open-water flooding and can produce much deeper and faster flooding. It damage an economy by causing river-side industrial facilities such as hydro-electric generating stations to shut down and to interfere with ship transport [2,3,4]

Methods

Results

Conclusion