A Clustering Based Automated Glacier Segmentation Scheme Using Digital Elevation Model

We present an automated scheme for segmentation of high mountain glaciers using Fast Adaptive Medoid Shift (FAMS) algorithm and Digital Elevation Model (DEM). FAMS is a non-parametric clustering technique that has been optimized and made data driven from its original Medoid Shift algorithm. 6 Band TM sensor satellite images are fed to FAMS as input along with height, slope and gradient information extracted from a DEM. Clean glacier and debris covered glacier are treated separately. Each glacier having its own regional minima and debris is delineated individually. A unique slope-gradient model is used to separate the debris covered portion from its surrounding and extension rocks as well as to exclude the lateral moraine. The proposed model is independent of the DN values of satellite image bands and therefore is able to perform well even in areas where debris covered glaciers exactly resemble the surrounding rocks. Experiments have been carried out on KaraKoram and Hindukush mountain ranges of Asia and validated against supervised manual segmentation results as well as Google EarthTM imagery. Results have shown our fully automated method to be time efficient, robust and accurate.

[1]  M. Hoelzle,et al.  On Rates and Acceleration Trends of Global Glacier Mass Changes , 1999 .

[2]  Syed Zulqarnain Gilani,et al.  Fast Block Clustering Based Optimized Adaptive Mediod Shift , 2009, CAIP.

[3]  Manfred F. Buchroithner,et al.  Automated delineation of debris-covered glaciers based on ASTER data , 2007 .

[4]  Frank Paul,et al.  EVALUATION OF DIFFERENT METHODS FOR GLACIER MAPPING USING LANDSAT TM , 2000 .

[5]  Frank Paul,et al.  Comparison of TM Derived Glacier Areas With Higher Resolution Data Sets , 2001 .

[6]  Oregon State.  Water Resources Research Institute , 1986 .

[7]  Ron Kikinis,et al.  Statistical validation of image segmentation quality based on a spatial overlap index. , 2004, Academic radiology.

[8]  M. Noguer,et al.  Climate change 2001: The scientific basis. Contribution of Working Group I to the Third Assessment Report of the Intergovernmental Panel on Climate Change , 2002 .

[9]  R. W. Sidjak.  Glacier mapping of the Illecillewaet icefield, British Columbia, Canada, using Landsat TM and digital elevation data , 1999 .

[10]  Naveed Iqbal Rao,et al.  Color Image Segmentation Using Multilevel Clustering Approach , 2008, 2008 Digital Image Computing: Techniques and Applications.

[11]  Luc Vincent,et al.  Watersheds in Digital Spaces: An Efficient Algorithm Based on Immersion Simulations , 1991, IEEE Trans. Pattern Anal. Mach. Intell..

[12]  J. Oerlemans.  Quantifying Global Warming from the Retreat of Glaciers , 1994, Science.

[13]  Adrian Luckman,et al.  Early recognition of glacial lake hazards in the Himalaya using remote sensing datasets , 2007 .

[14]  Danielle J. Marceau,et al.  Combining DEM Parameters With Landsat MSS And TM Imagery In A GIS For Mountain Glacier Characterization , 1990 .

[15]  John R. Jensen,et al.  Introductory Digital Image Processing: A Remote Sensing Perspective , 1986 .

[16]  Naveed Iqbal Rao,et al.  Semantics Sensitive Segmentation and Annotation of Natural Images , 2008, 2008 IEEE International Conference on Signal Image Technology and Internet Based Systems.

[17]  L. Breiman,et al.  Variable Kernel Estimates of Multivariate Densities , 1977 .

[18]  Ilan Shimshoni,et al.  Mean shift based clustering in high dimensions: a texture classification example , 2003, Proceedings Ninth IEEE International Conference on Computer Vision.

[19]  Andreas Kääb,et al.  Combining satellite multispectral image data and a digital elevation model for mapping debris-covered glaciers , 2004 .

[20]  M. Frezzotti,et al.  Snow and ice reflectance spectra of the Nansen Ice Sheet surfaces , 1996 .

[21]  Takeo Kanade,et al.  Mode-seeking by Medoidshifts , 2007, 2007 IEEE 11th International Conference on Computer Vision.

[22]  D. Easterbrook.  Surface processes and landforms , 1992 .