In this letter, a new image fusion methodology for integration of SAR and optical images using combined dictionary is proposed. The approach taken is based on sparse and redundant representations by employing a combined dictionary consisting of wavelets, shearlets and discrete cosine transform (DCT). Wavelets and shearlets provide pointlike and curvelike structures for the optical image, and DCT are taken as obtaining the best performance on SAR image. The experimental results demonstrate feasibility and effectiveness of the method.
2. Kong, W. W., "Multi-sensor image fusion based on NSST domain I2CM," Electron. Lett., Vol. 49, No. 13, 802-803, Jun. 2013.
doi:10.1049/el.2013.1192
3. Elad, M. and M. Aharon, "Image denoising via sparse and redundant representations over learned dictionaries," IEEE Trans. Image Processing, Vol. 15, No. 12, 3736-3745, Dec. 2006.
doi:10.1109/TIP.2006.881969
4. Kutyniok, G. and W. Q. Lim, "Image separation using wavelets and shearlets,", 2010, http://www.shearlab.org/index publications.html.
5. Guo, R., L. Zhang, M. Xing, and J. Li, "Polarimetric SAR image fusion using nonnegative matrix factorisation and improved-RGB model," Electron. Lett., Vol. 46, No. 20, 3736-3745, Sep. 2010.
doi:10.1049/el.2010.1476
Submit
