S. Liu, J. Chen, Y. Ai and S. Rahardja, "An Optimized Quantization Constraints Set for Image Restoration and its GPU Implementation," in IEEE Transactions on Image Processing, vol. 29, pp. 6043-6053, 2020, doi: 10.1109/TIP.2020.2988131.
Abstract:
This paper presents a novel optimized quantizationconstraint set, acting as an add-on to existing DCT-based imagerestoration algorithms. The constraint set is created based ongeneralized Gaussian distribution which is more accurate thanthe commonly used uniform, Gaussian or Laplacian distributionswhen modeling DCT coefficients. More importantly, the proposedconstraint set is optimized for individual input images and thus itis able to enhance image quality significantly in terms of signal-to-noise ratio. Experimental results indicate that the signal-to-noise ratio is improved by at least 6.78% on top of the existingstate-of-the-art methods, with a corresponding expense of only0.38% in processing time. The proposed algorithm has also beenimplemented in GPU, and the processing speed increases furtherby 20 times over that of CPU implementation. This makes thealgorithm well suited for fast image retrieval in security andquality monitoring system.
License type:
PublisherCopyrights
Funding Info:
There was no specific funding for the research done