Abstract | ||
---|---|---|
Subpixel-based downsampling is a new downsampling technique which utilizes the fact that each pixel in LCD is composed of three individually addressable subpixels. Subpixel-based downsampling can provide higher apparent resolution than pixel-based downsampling. In this paper we study the inverse problem of subpixel-based downsampling. We found that conventional pixel-based super resolution algorithms are not suitable for subpixel-based downsampled images due to the special downsampling pattern. In this paper we propose a super resolution algorithm specially for subpixel-based downsampled images, which use piecewise autoregressive model to model spatial correlation of neighboring pixels, and incorporate the special data degradation term corresponding to the subpixel downsampling pattern. We formulate the super resolution problem as a constrained least square problem and solve it using Gauss-Seidel iteration. Experiment results demonstrate the effectiveness of the proposed algorithm. |
Year | Venue | Keywords |
---|---|---|
2012 | Signal & Information Processing Association Annual Summit and Conference | autoregressive processes,correlation methods,image resolution,image sampling,iterative methods,least squares approximations,Gauss-Seidel iteration,LCD,constrained least square problem,data degradation term,piecewise autoregressive model,spatial correlation,subpixel downsampling pattern,subpixel-based downsampling,super resolution algorithm |
Field | DocType | ISSN |
Autoregressive model,Computer vision,Spatial correlation,Artificial intelligence,Inverse problem,Pixel,Subpixel rendering,Upsampling,Image resolution,Piecewise,Mathematics | Conference | 2309-9402 |
ISBN | Citations | PageRank |
978-1-4673-4863-8 | 0 | 0.34 |
References | Authors | |
0 | 6 |
Name | Order | Citations | PageRank |
---|---|---|---|
Ketan Tang | 1 | 106 | 12.98 |
Oscar C. Au | 2 | 1592 | 176.54 |
Lu Fang | 3 | 343 | 55.27 |
Yuanfang Guo | 4 | 95 | 18.21 |
Pengfei Wan | 5 | 45 | 9.40 |
Lingfeng Xu | 6 | 53 | 9.81 |