Dissertation > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer applications > Information processing (information processing) > Pattern Recognition and devices > Image recognition device

The Research of Application of Ridgelet Transform in the Fusion of Multispectral and Panchromatic Images

Author ZhaoZiYi
Tutor ZhengYongGuo
School Shandong University of Science and Technology
Course Computer Software and Theory
Keywords Multispectral image Panchromatic image Ridgelet Radon transform Slant Stack transform Image Fusion
CLC TP391.41
Type Master's thesis
Year 2011
Downloads 27
Quotes 0
Download Dissertation

Image fusion is a fusion of an image for the study of the branch , that is a plurality of image fusion of different modes of image sensors in the multiple images of the same scene , or in the same sensor at different times or under different ways to obtain the image synthesis an input image containing information related to the image of the process. studied ridgelet transform in image fusion . Paper discussed the characteristics of the ridge wavelet transform and implementation process , the first two-dimensional Fourier transform of the image on the Radon transform domain Radon transform domain, and then on the Radon domain on each pole diameter one-dimensional wavelet transform . Radon transform is used in the realization of bilinear interpolation , although part of the expense of accuracy, but improves the efficiency of the algorithm . Slant Stack transform paper -based , using the low-frequency part of the weighted average of the absolute value of the largest high-frequency part of the fusion rules , the ridge is given based on wavelet transform multi -spectral image and panchromatic image fusion algorithm , using grayscale variance , clear degree , information entropy evaluation parameters based on ridgelet fusion and fusion based on wavelet results were compared , the results show that the effect of the former edge details of the image is better than the representation of the latter.

Related Dissertations
More Dissertations