Multisource classification and pattern recognition methods for polar geospatial information mining using WorldView-2 data

Parag S. Khopkar; Shridhar D. Jawak; Alvarinho J. Luis

doi:10.1117/12.2222984

30 April 2016 Multisource classification and pattern recognition methods for polar geospatial information mining using WorldView-2 data

Parag S. Khopkar, Shridhar D. Jawak, Alvarinho J. Luis

Author Affiliations +

Proceedings Volume 9880, Multispectral, Hyperspectral, and Ultraspectral Remote Sensing Technology, Techniques and Applications VI; 98801R (2016) https://doi.org/10.1117/12.2222984
Event: SPIE Asia-Pacific Remote Sensing, 2016, New Delhi, India

Abstract

Current research study emphasizes the importance of advanced digital image processing methods in order to delineate between various LULC features. In the case of the Antarctica, the present LC (snow/ice, landmass, water, vegetation etc.) and the present LU (research stations of various nations) needs to be mapped accurately for the hassle free routine activities. Geo-location has become the most important part of geosciences studies. In this paper we have tried to locate three most important features (snow/ice, landmass, and water) and also have extracted the extent of the same using the multisource classification (image fusion/pansharpening) and pattern recognition (supervised/unsupervised methods, index ratio methods). Innovation in developing spectral index ratios has led us to come up with an unique ratio named Normalized Difference Landmass Index (NDLI) which performed better (Avg. Bias: 51.99m) than other ratios such as Normalized Difference Snow/Ice Index (NDSII) (Avg. Bias: -1572.11m) and Normalized Difference Water Index (NDWI) (Avg. Bias: 1886.60m). The practiced trial and error methodology quantifies the productivity of not only the classification methods over one other but also that of the fusion methods. In present study, classifiers used (Mahalanobis and Winner Takes All) performed better (Avg. Bias: 122.16 m) than spectral index ratios (Avg. Bias: 620.16 m). The study also revealed that newly introduced bands in WorldView-2, band 1 (Coastal Blue), 4 (Yellow), 6 (Red-edge) and 8 (Near Infrared-2) along with traditional bands have the capacity to mine the polar geospatial information with utmost accuracy and efficiency.

Citation Download Citation

Parag S. Khopkar, Shridhar D. Jawak, and Alvarinho J. Luis "Multisource classification and pattern recognition methods for polar geospatial information mining using WorldView-2 data", Proc. SPIE 9880, Multispectral, Hyperspectral, and Ultraspectral Remote Sensing Technology, Techniques and Applications VI, 98801R (30 April 2016); https://doi.org/10.1117/12.2222984

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
7 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Image classification

Feature extraction

Pattern recognition

Satellites

Remote sensing

Earth observing sensors

Mining

Show All Keywords

Keywords/Phrases

Search In:

Publication Years