Joint structural similarity and entropy estimation for coded-exposure image restoration

摘要

We address the image deblurring using coded exposure which can keep image content that may be lost by a traditional shutter. In the restoration of a coded exposure image, the automatic estimation of smear length is the key problem. Because the coded exposure image does not lose high frequency information of the image, the structural similarity compared with the original image is retained. In this paper, we propose a joint coarse to fine estimation method. By comparing structural similarity between the coded-exposure image and its restored image, the smear length can be roughly estimated first. And then the entropy of the restored image is further computed within a small range of the previously estimated smear length. An image that is restored with the wrong smear length will be far from the structure of the coded image that will have high entropy and low structure similarity with the coded exposure image.