摘要
Reconstructing an HR （high-resolution） image which preserves the image intrinsic structures from its LR （low-resolution） counterpart is highly challenging. This paper proposes a new surface reconstruction algorithm applied to image interpolation. The interpolation surface for the whole image is generated by putting all the quadratic polynomial patches together. In order to eliminate the jaggies of the edge, a new weight fimction containing edge information is incorporated into the patch reconstruction procedure as a constraint. Extensive experimental results demonstrate that our method produces better results across a wide range of scenes in terms of both quantitative evaluation and subjective visual quality.
Reconstructing an HR （high-resolution） image which preserves the image intrinsic structures from its LR （low-resolution） counterpart is highly challenging. This paper proposes a new surface reconstruction algorithm applied to image interpolation. The interpolation surface for the whole image is generated by putting all the quadratic polynomial patches together. In order to eliminate the jaggies of the edge, a new weight fimction containing edge information is incorporated into the patch reconstruction procedure as a constraint. Extensive experimental results demonstrate that our method produces better results across a wide range of scenes in terms of both quantitative evaluation and subjective visual quality.
基金
Supported by Shandong Province Higher Educational Science and Technology Program(No.J12LN34)
Shandong Ji'nan College and Institute Independent Innovation Project(No.201303011)
Shandong Ji'nan College and Institute Independent Innovation Project(No.201303021)
the Scientific Research Foundation of Shandong Province of Outstanding Young Scientist Award(No.BS2011DX025)
