Aggregation of blurred images through weighted FBA to remove camera shake
K. Gayathri, P. Marikkannu · 2016
Many algorithms were designed to remove image blur due to camera shake, either with one or multiple input images, by using the deconvolution problem. If the photographer takes a burst of images, a modality available in virtually all modern digital cameras, it is possible to combine all the images to get a clean sharp version. This algorithm does not use blur estimation or its inverse problem. The proposed algorithm is strikingly simple where the average weight is calculated using Fourier domain which depends on the Fourier spectrum magnitude. Here, burst of images are taken into an account and each image in the burst is blurred differently. The proposed Fourier burst accumulation algorithm shows that one can obtain a sharp image by combining all the images together which is blurred differently. This can be implemented in modern smart phones.