2 March 2015 Quality assessment of ZiYuan-3 multispectral and panchromatic images fusion: applied in Jiangsu coastal wetland area, China
Ruijuan Wu, Xiufeng He, Jing Wang
Author Affiliations +
Abstract
The new launched ZiYuan-3 (ZY-3) satellite with multispectral (MS) bands and a panchromatic (PAN) band has presented a new opportunity to assess image fusion methods for coastal wetland mapping. This study focuses on image fusion quality assessment through both quantitative spectral and spatial quality analysis and object-oriented classification comparison. Various methods for pan-sharpening ZY-3 MS and PAN bands are tested, including generalized intensity-hue-saturation transform (GIHS), à trous wavelet transform (AWT), nonsubsampled contourlet transform (NSCT), and a combination of NSCT with GIHS (NSCT_GIHS). Spectral fidelity and spatial preservation of fused bands are compared with the original MS bands as reference, and spatial information injections of fused bands are compared with the resampled PAN band as reference. The fusion results demonstrate that, on average, the NSCT_GIHS method has the best performance on spectral fidelity and spatial preservation as well as spatial information injection. The near-infrared (NIR) band has the best spatial information injection in terms of entropy and cross-entropy indices, whereas the NIR band has the best spatial preservation in terms of entropy and structure similarity indices. The classification results show that NSCT_GIHS method also obtains the highest overall accuracy (96%) and Kappa coefficient (0.9425); this is in agreement with the quantitative analysis.
© 2015 Society of Photo-Optical Instrumentation Engineers (SPIE) 1931-3195/2015/$25.00 © 2015 SPIE
Ruijuan Wu, Xiufeng He, and Jing Wang "Quality assessment of ZiYuan-3 multispectral and panchromatic images fusion: applied in Jiangsu coastal wetland area, China," Journal of Applied Remote Sensing 9(1), 095089 (2 March 2015). https://doi.org/10.1117/1.JRS.9.095089
Published: 2 March 2015
Lens.org Logo
CITATIONS
Cited by 2 scholarly publications.
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Image fusion

Image classification

Image quality

Multispectral imaging

Near infrared

Image segmentation

Image processing

RELATED CONTENT

Monitoring of water bodies using remote sensing data
Proceedings of SPIE (December 18 2019)

Back to Top