Data fusion in satellite images and their quantitative analysis

Abstract

Data fusion may be considered a kind of complex image processing method which combines the different satellite images to bring complementary and useful information into one single composite image. The fused image will usually have better spectral and spatial resolutions to see more detailed and sometimes new features not seen in the original images. In the present work, Landsat 7 ETM, 1999 dated satellite data which is belong to Canakkale Province is used. The multispectral satellite image that is rich spectrally and panchromatic(PAN) image that is rich spatially but poor spectrally are fused by using two different methods based on Intensity, Hue, Saturation (IHS) Transformation and Principal Component Analysis. Both of the methods are assessed by looking through the relationship between the vegetation indexes of final images and multispectral satellite image.