By Topic

A Comparison of Pixel- and Object-Based Glacier Classification With Optical Satellite Images

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$33 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

4 Author(s)
Rastner, P. ; Dept. of Geogr., Univ. of Zurich, Zurich, Switzerland ; Bolch, T. ; Notarnicola, C. ; Paul, F.

Precise information about the size and spatial distribution of glaciers is needed for many research applications, for example water resources evaluation, determination of glacier specific changes in area and volume, and for calculation of the past and future contribution of glaciers to sea-level change. However, mapping glacier outlines is challenging even under optimal conditions due to time consuming manual corrections of wrongly classified pixels. In the last decades, advantages in computer technologies have led to the development of object-based-image analysis (OBIA), an image classification technique that can be seen as an alternative to the common pixel-based image analysis (PBIA). In this study we compare the performance of OBIA with PBIA for glacier mapping in three test regions with challenging mapping conditions. In both approaches, a ratio image was created to map clean snow and ice while thermal and slope information was used to assist in the identification of debris-covered ice. The mapping results of OBIA have overall a ~ 3% higher quality than PBIA, in particular in the processing of debris-covered glaciers where OBIA has a 12% higher accuracy. The post-processing possibilities in OBIA (e.g., the application of a processing loop and neighborhood analysis) are especially powerful to improve the final classification. This leads also to a reduction of the workload for the manual corrections, which are still required to achieve a sufficient accuracy.

Published in:

Selected Topics in Applied Earth Observations and Remote Sensing, IEEE Journal of  (Volume:7 ,  Issue: 3 )