Remote Sensing Images Fusion Based On Curvelet Transform And Shearlet Transform

With the continuous development of sensing technology, romte sensing technology provides a wide range of information to various research provides more space for development, but under the influence of various factors in the process of filming, obtained a single remote sensing image is very difficult to meet the needs of information extraction. Remote sensing image fusion is to keep the original spectral image spectral characteristics at the same time add panchromatic image more detailed spatial details, it can obtain more useful information than a single image. Curvelet transform not only inherit and develop the excellent airspace and better directional selectivity, escription of geometrical characteristics of the image is more suitable than the traditional wavelet transform, and more advantageous to extract the image detail information, so it has the better resolution. Shearlet transform has more direction, multi-scale temporal locality and width of each scale than change and compactly supported interval of anisotropy, to a better image sparse representation. This thesis mainly research of remote sensing image fusion based on curvelet transform and shearlet algorithm, the main work is as follows:1. Summarizes the research status of remote sensing image fusion and the significance of research, this thesis discusses the subjective and objective evaluation of remote sensing image fusion quality assessment standards, further introduces several traditional fusion method of remote sensing image fusion.2. The thesis introduces the related concepts of Curvelet transform and Shearlet transform, study the principle of decomposition and reconstruction, further analysis their application in remote sensing image.3. Aimed at the shortcomings in traditional wavelet in remote sensing image fusion, this thesis puts forward the remote sensing image fusion based on Curvelet transform and Shearlet transform algorithm. Its characteristic is make full use of the direction of the correlation between coefficient, high frequency coefficient of regional variance in weighted fusion algorithm amended, at the same time in the high frequency coefficients with the improved fusion energy of Laplace operator. The experimental results show that the proposed method in the image visual effect and objective evaluation index is superior to the traditional fusion method.