DocumentCode
2684030
Title
A fusion method for mixed pixels based on prior type judgment
Author
Jing, Linhai ; Cheng, Qiuming
Author_Institution
York Univ., Toronto
fYear
2007
fDate
23-28 July 2007
Firstpage
5162
Lastpage
5165
Abstract
In reality, a low-spatial-resolution (LR) mixed pixel superimposes some high-spatial-resolution (HR) pixels, some of which are mixed and some of which are pure. In current fusion methods, mixed multispectral pixels are seldom discriminated from pure pixels and specifically processed. The resultant fused versions of the mixed pixels usually keep spectrally mixed and visually blurred. In this paper, for each spatially expanded mixed pixel, once the type of its corresponding HR pixel, soil or vegetation pixel, can be determined with reference to the relationship of panchromatic and multispectral images and with respect to the corresponding panchromatic pixel, its fused version may be spectrally un-mixed. In a test, the method proposed in this paper offered synthetic products with overall high spectral accuracy and had better sharpened visualization effects on some mixed boundary pixels than some current fusion methods.
Keywords
geophysical signal processing; remote sensing; sensor fusion; soil; vegetation; fusion method; mixed boundary pixels; mixed multispectral pixels; multispectral images; panchromatic images; prior type judgment; sharpened visualization effects; soil pixel; vegetation pixel; Geology; Geoscience; Image fusion; Image resolution; Mineral resources; Multispectral imaging; Pixel; Soil; Spatial resolution; Vegetation mapping; image enhancement; image fusion; mixed pixel; spatial resolution;
fLanguage
English
Publisher
ieee
Conference_Titel
Geoscience and Remote Sensing Symposium, 2007. IGARSS 2007. IEEE International
Conference_Location
Barcelona
Print_ISBN
978-1-4244-1211-2
Electronic_ISBN
978-1-4244-1212-9
Type
conf
DOI
10.1109/IGARSS.2007.4424024
Filename
4424024
Link To Document