Sharpened Images Database (SID) contains images which are blurred, sharpened, and over-sharpened using several different algorithms. The purpose of the database is to train and/or test objective quality metrics in the enhancement scenario where the quality of the reference can be both increased and decreased.
MATLAB scripts for calculating scores of new metrics and analyzing their performance are included as well to enable reproducible research results on a common test bed.
35 versions of each reference, i.e. 144 images (including originals).
Gaussian blur – 3 different levels
Unsharp Mask – 4 sharpening levels for two different sizes of filtering mask – 8 images in total
Augmented Unsharp Mask – 4 sharpening levels for two different sizes of filtering mask – 8 images in total
GSEA – 4 sharpening levels for two different sizes of filtering mask – 8 images in total
SDME – 4 sharpening levels for two different sizes of filtering mask – 8 images in total
The subjective data has been collected from 38 observers using Adaptive Square Design Pair Comparison methodology. For more information about the procedure selection, data preparation, experimental design, and results processing, refer to the paper.
If you use any part of the provided dataset in your research, we kindly ask you to cite the paper:
L. Krasula, P. Le Callet, K. Fliegel, and M. Klíma, Quality Assessment of Sharpened Images: Challenges, Methodology, and Objective Metrics, IEEE Transactions on Image Processing, 2017.
If you have any questions regarding the database, code, the paper, or you want to report a malfunction of the software, feel free to contact us by email to L.Krasula@gmail.com.