Relative value of radar and optical data for land cover/use mapping: Peru example. (2nd January 2018)

• Record Type: Journal Article
• Authors: Haack, Barry; Mahabir, Ron
• Is Part Of: International journal of image and data fusion. Volume 9:Number 1(2018)
• Page Start: 1
• Page End: 20
• Publication Date: 2018-01-02
• Subjects: Separability -- divergence -- PALSAR -- Landsat -- dual-polarisation -- sensor integration; Image processing -- Periodicals; Multisensor data fusion -- Periodicals; Multisensor data fusion; Periodicals; 621.36705
• DOI: 10.1080/19479832.2017.1398188
• Languages: English
• ISSNs: 1947-9832

Abstract

ABSTRACT: This study determined using divergence measures the best individual and combinations of various numbers of bands for six land cover/use classes around the city of Arequipa, Peru. A 15 band data stack consisting of PALSAR L-band dual-polarised radar, Landsat optical data, as well as six variance texture measures extracted from the PALSAR images, was used in this study. Spectral signatures were obtained for each class for the divergence examination. The band having the highest separability was the Landsat visible red band followed by the two largest window PALSAR texture measures. The best three band combination included three very different data types, Landsat visible red, near infrared and the PALSAR HH variance texture from a 17 × 17 pixel window. There was no need based upon the divergence values to use more than five bands for classification.