Scaling is indispensable for multimedia display to convert image sources from input resolution to output resolution. Conventional scaling algorithms inevitably induce side effects such as aliasing effect; therefore, super-resolution (SR) scaling was raised to improve the scaling quality through complex computations. Even though the scaling quality will be greatly improved through complex SR scaling, the image details of some images with over-exposed and saturated scenes can not be recovered well. In this paper, we proposed a simple SR scaling algorithm to restore the unapparent details and edges of original over-exposed or saturated images by simple histogram analysis, remapping and blending process without destroying the tone of original images. Several over-exposed or under-exposed images are scaled through the proposed algorithm, and the results show great improvements over conventional scaling and SR scaling approaches.
[1]
Hsieh Hou,et al.
Cubic splines for image interpolation and digital filtering
,
1978
.
[2]
Thomas S. Huang,et al.
Image super-resolution as sparse representation of raw image patches
,
2008,
2008 IEEE Conference on Computer Vision and Pattern Recognition.
[3]
Moon Gi Kang,et al.
Super-resolution image reconstruction: a technical overview
,
2003,
IEEE Signal Process. Mag..
[4]
H. Demirel,et al.
Image Super Resolution Based on Interpolation of Wavelet Domain High Frequency Subbands and the Spatial Domain Input Image
,
2010
.
[5]
Thierry Blu,et al.
Linear interpolation revitalized
,
2004,
IEEE Transactions on Image Processing.