|
Descripción
|
|
|---|---|
| It has been proposed a search algorithm, based on the Simulating Annealing, which improves the global quality of satellite images fused through a fusion algorithm based on a joint MDMR representation. In a first phase, the search algorithm has been implemented for carrying out an exhaustive exploration of the space defined by two parameters (elongation and scale), involved in design of the filter bank used by the fusion algorithm. The analysis of the influence of these parameters on the fused images quality has allowed establishing an oriented search criterion, which reduces significantly the number of iterations requiered for converging. This search algorithm can be applied to different functions. But, since the aim of this work is to get fused images with the spatial and spectral quality balanced, here OF has been defined as the difference of two quality indices: one spatial and one spectral. Experimental results have proved that the convergence of the algorithm is independent on initial conditions and that the number of iterations is significatively reduced when the oriented criterion is applied. Fromthe qualitative and quantitative analysis of the fused images quality, it can be concluded that the MDMR fusion methodology complemented with the oriented search algorithm, which is proposed in this paper, provides fused images with a higher spectral quality than the other algorithms evaluated; and spatial quality comparable to that offered by the WAT method. Still, the most notable feature of the proposed methodology is that it provides a total balance between the two qualities, spatial and spectral, for the two kind of images used. | |
|
Internacional
|
Si |
|
DOI
|
|
|
Edición del Libro
|
|
|
Editorial del Libro
|
|
|
ISBN
|
978-953-307-483-2 |
|
Serie
|
|
|
Título del Libro
|
Search Algorithms |
|
Desde página
|
117 |
|
Hasta página
|
132 |