Fully polarimetric SAR mosaicing and land cover classification in the northern taiga region
The paper describes an algorithm development for production of large-scale fully polarimetric mosaic using multitemporal ALOS PALSAR acquisitions during snow melting season in Lapland. Several popular supervised and unsupervised polarimetric classification methods are benchmarked in order to identif...
Main Authors: | , , , |
---|---|
Format: | Other Non-Article Part of Journal/Newspaper |
Language: | English |
Published: |
European Space Agency (ESA)
2011
|
Subjects: | |
Online Access: | https://cris.vtt.fi/en/publications/c8d0b799-2c95-40cc-8076-39c7321174a1 http://articles.adsabs.harvard.edu/cgi-bin/nph-iarticle_query?2011ESASP.695E.45A&data_type=PDF_HIGH&whole_paper=YES&type=PRINTER&filetype=.pdf |
Summary: | The paper describes an algorithm development for production of large-scale fully polarimetric mosaic using multitemporal ALOS PALSAR acquisitions during snow melting season in Lapland. Several popular supervised and unsupervised polarimetric classification methods are benchmarked in order to identify more suitable and robust approach for multitemporal mosaic processing. Different variants of polarimetric seamhiding between original images are evaluated in order to effectively eliminate stripes in the mosaic. Also an impact of such seam-hiding on classification performance is studied. Obtained results indicate the necessity of seam-hiding procedure for producing homogeneous mosaics and obtaining consistent classification results in a single classification step |
---|