Adaptive image processing a computational intelligence perspective - Spatially Adaptive Image Restoration

In the previous chapter, it was seen that as long as the weights of the network remain constant from iteration to iteration, Algorithm 2.3 will converge. This result holds regardless of whether the weights of the network vary from pixel to pixel across the image to be restored. We will refer to this type of variation of the weights from region to region in the image as spatial variation. The analysis in the previous chapter focused solely on the case when the weights are not spatially variant. But there are situations when we may want to solve problems that involve spatially variant weights. The first case occurs when the type and/or strength of the degrading function varies from region to region in the image. For example, imagine taking a picture of a distant mountain when a nearby tree is in the picture. If the image of the mountain is in focus, then the image of the tree will most likely be out of focus.

Từ khóa: Spatially Adaptive Image Restoration, Adaptive image processing a computational intelligence perspective, image processing, network techniques, kỹ thuật xử lý, công nghệ mạng