Computation Time Optimization in Super-Resolution Applications

摘要

The necessity to improve image resolution is of great concern in multiple diverse fields such as: medicine, communications, or satellite and underwater applications. A high variety of techniques for image enhancement has been proposed in the literature, being a trade-off the relation between the computation time and the quality of the obtained results. This work is focused on a test environment that permits to objectively compare the quality enhancement of images processed by two different improvement methods: bilinear interpolation and Super-Resolution (SR), presenting how these results relate to the computation time. The objective comparison is based on the PSNR (Peak Sig-nal-to-Noise Ratio) and the SSIM (Structural SIMilarity).