Statistical Model-based fusion of noisy multi-band images in the wavelet domain

Gradient entropy metric and p-Laplace diffusion constraint-based algorithm for noisy multispectral image fusion

“…In image processing, equation (20) should be utilized for discrete data set. In this research we use the finite difference method to obtain the numerical solution.…”

“…In this research we use the finite difference method to obtain the numerical solution. In order to solve equation (20), the semi-implicit method [31] is used to ensure the convergence. So, equation (20) is converted into the evolution with the step-size t liking equation (21).…”

“…It will reduce the noise of the fused image, but it cannot effectively extract the edge detail information of the input images. Nowadays, more robust algorithms are based on the wavelet transform [18][19][20]. Especially in [19], a noisy image fusion method based on non-Gaussian model in the wavelet domain was proposed to construct wavelet coefficients with the generalised Gaussian and alpha-stable distribution.…”

Fusion of noisy images based statistical model in shearlet domain

“…specifically, an image fusion prototype method, originally proposed in [4], has been modified by alpha-stable distribution. In [5] and [6], the generalized gaussian has been employed in a similar manner in [7]. In [8], a fusion scheme based on anisotropic bivariate shrinkage has been proposed.…”