Super-Resolution Image with Estimated High Frequency Compensated algorithm
Jong-Tzy Wang, Kai-Wen Liang, Shu-Fan Chang, Pao‐Chi Chang · 2009
In this paper, we propose an estimated high frequency compensated (EHFC) algorithm for super resolution images. It is based on iterative back projection (IBP) method combined with compensated high frequency models according to different applications. The proposed algorithm not only improves the quality of enlarged images produced by zero-order, bilinear, or bicubic interpolation methods, but also accelerates the convergence speed of IBP. In experiments with general tested images, EHFC method can increase the speed by 1 ~ 6.5 times and gets 0.4 ~ 0.7 dB PSNR gain. In text image tests, EHFC method can increase 1.5 ~ 6.5 times in speed and 1.2 ~ 8.3 dB improvement in PSNR.