Image evaluation based on the mean structural similarity for wavefront coding
Tsubasa Fukuda, Shinichi Komatsu · 2017
We would like a numerical value that indicates the extent to which restoration is achieved compared with the original image in wavefront coding. Using the mean structural similarity (MSSIM) as the image evaluation value, we evaluated restored images and compared phase masks. In this study, we confirmed that it is desirable to optimize the coefficients of the phase mask so that the MSSIM is highest to obtain high-quality restored images.