GlacierNet: A Deep Learning Architecture for Debris-Covered Glacier Mapping

Title

GlacierNet: A Deep Learning Architecture for Debris-Covered Glacier Mapping

Authors

Files

Description

The global temperature has been continuously increasing over the past decades. The effect of temperature increase can directly affect the health, dynamics, and processes of alpine glaciers. In this research, the convolutional neural network (CNN), which is a deep learning, feed-forward neural network, is applied to the Landsat era satellite images for automated mapping of debris-covered glaciers. Our preliminary results indicate high accuracy in glacier mapping, a major step in developing a fully automated methodology for glacier mapping.

Publication Date

4-24-2019

Project Designation

Independent Research

Primary Advisor

Umesh K Haritashya, K. Asari Vijayan

Primary Advisor's Department

Electrical and Computer Engineering

Keywords

Stander Symposium poster

Comments

Presenter: Zhiyuan Xie

GlacierNet: A Deep Learning Architecture for Debris-Covered Glacier Mapping

Share

COinS