Abstract | ||
---|---|---|
This paper presents a neural network-based method for image super-resolution. In this technique, the super-resolution is considered as an ill-posed inverse problem which is solved by minimizing an evaluation function established based on an observation model that closely follows the physical image acquisition process. A Hopfield neural network is created to obtain an optimal solution to the problem. Not like some other single-frame super-resolution techniques, this technique takes into consideration PSF (point spread function) blurring as well as additive noise and generates high-resolution images with more preserved or restored image details. Experimental results demonstrate that the high-resolution images obtained by this technique have a very high quality in terms of PSNR (peak signal-to-noise ratio) and visually look more pleasant |
Year | DOI | Venue |
---|---|---|
2007 | 10.1109/ITNG.2007.105 | ITNG |
Keywords | Field | DocType |
hopfield neural network,image interpolation,point spread function,hopfield neural nets,noise,interpolation,image resolution,image resolution enhancement,image restoration,inverse problems,image acquisition,image detail,peak signal-to-noise ratio,image superresolution,high-resolution image,image enhancement,ill-posed inverse problem,single-frame super-resolution technique,image super-resolution,neural network-based method,point spread function blurring,physical image acquisition process,additive noise,super resolution,neural network,evaluation function,peak signal to noise ratio,neural networks,psnr | Computer vision,Peak signal-to-noise ratio,Computer science,Interpolation,Artificial intelligence,Inverse problem,Image restoration,Artificial neural network,Point spread function,Image resolution,Image scaling | Conference |
ISBN | Citations | PageRank |
0-7695-2776-0 | 3 | 0.40 |
References | Authors | |
6 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Shuangteng Zhang | 1 | 18 | 2.85 |
Yihong Lu | 2 | 3 | 0.74 |