Blurness Removal via Weighted Fourier Burst Accumulation with Sift and Smoothing

Camera shakes during exposure leads to objectionable image blur and ruins many photographs. If the photographer takes a burst of images, a modality available in virtually all modern digital cameras, we show that it is possible to combine them to get a clean sharp version. This is done without explicitly solving any blur estimation and subsequent inverse problem. In previous method they are used blind deconvolution algorithm. Most blind deconvolution algorithms try to estimate the latent image without any other input than the noisy blurred image itself. In our proposed method we implement the new method called Fourier Burst Accumulation. It performs a weighted average in the Fourier domain, with weights depending on the Fourier spectrum magnitude. We directly compute the corresponding Fourier transforms. Since camera shake motion kernels have a small spatial support, their Fourier spectrum magnitudes vary very smoothly. Thus, can apply low pass filter before computing the weights, with filter of standard deviation �ƒ. The strength of the low pass filter (controlled by the parameter �ƒ) should depend on the assumed kernel size (the smaller the kernel the more regular its Fourier spectrum magnitude). Although this low pass filter is important, the results are not too sensitive to the value of �ƒ. The final Fourier burst aggregation is (note that the smoothing is only applied to the weights calculation). The extension to color images is straightforward. The accumulation is done channel by channel using the same Fourier weights for all channels. The weights are computed by arithmetically averaging the Fourier magnitude of the channels before the low pass filtering.