Fast Super-Resolution Algorithm Based on Dictionary Size Reduction Using k-Means Clustering

ETRI Journal

关键词: learning;    image-based;    clustering;    dictionary;    training;    Super-resolution;   

Others  :  1185920
DOI  :  10.4218/etrij.10.0109.0637

【 摘 要 】

This paper proposes a computationally efficient learning-based super-resolution algorithm using k-means clustering. Conventional learning-based super-resolution requires a huge dictionary for reliable performance, which brings about a tremendous memory cost as well as a burdensome matching computation. In order to overcome this problem, the proposed algorithm significantly reduces the size of the trained dictionary by properly clustering similar patches at the learning phase. Experimental results show that the proposed algorithm provides superior visual quality to the conventional algorithms, while needing much less computational complexity.

   

【 参考文献 】

• [1]R.G. Keys, "Cubic Convolution Interpolation for Digital Image Processing," IEEE Trans. Acoustics, Speech, Signal Process., vol. 29, no. 6, Dec. 1981, pp. 1153-1160.
• [2]H.S. Hou and H.C. Andrews, "Cubic Splines for Image Interpolation and Digital Filtering," IEEE Trans. Acoustics, Speech, Signal Process., vol. 26, no. 6, Dec. 1978, pp. 508-517.
• [3]J. Allebach and P.W. Wong, "Edge-Directed Interpolation," Proc. Int. Conf. Image Process., vol. 3, Sept. 1996, pp. 707-710.
• [4]X. Li and M. Orchard, "New Edge-Directed Interpolation," IEEE Trans. Image Process., vol. 10, no. 10, Oct. 2001, pp. 1521-1527.
• [5]Q. Wang and R. Kreidieh, "A New Orientation-Adaptive Interpolation Method," IEEE Trans. Image Process., vol. 16, no. 4, Apr. 2007, pp. 889-900.
• [6]S.M. Kwak, J.H. Moon, and J.K. Han, "Modified Cubic Convolution Scaler for Edge-Directed Nonuniform Data," Optical Eng., vol. 46, no. 10, 107001, Oct. 2007, doi:10.1117/12.782389.
• [7]M. Irani and S. Peleg, "Motion Analysis for Image Enhancement: Resolution, Occlusion, and Transparency," J. Visual Commun. Image Representation, vol. 4, no. 4, 1993, pp. 324-335.
• [8]Z. Lin and H.Y. Shum, "Fundamental Limits of Reconstruction-Based Superresolution Algorithms Under Local Translation," IEEE Trans. Pattern Anal. Machine Intell., vol. 26, no. 1, Jan. 2004, pp. 83-97.
• [9]S.C. Park, M.K. Park, and M.G. Kang, "Super-Resolution Image Reconstruction: A Technical Overview," IEEE Signal Process. Magazine, vol. 20, no. 3, 2003, pp. 21-36.
• [10]W.T. Freeman, T.R. Jones, and E.C. Pasztor, "Example-Based Super-Resolution," IEEE Computer Graphics Appl., vol. 22, no. 2, Oct. 2002, pp. 56-65.
• [11]J. Sun et al., "Image Hallucination with Primal Sketch Priors," Proc. IEEE Comput. Soc. Conf. Comp. Vision Pattern Recog., vol. 2, 2003, pp. 729-736.
• [12]H. Chang and D. Yeung, and Y. Xiong, "Super-Resolution through Neighbor Embedding," Proc. IEEE Comput. Soc. Conf. Comp. Vision Pattern Recog., vol. 1, 2004, pp. 275-282.
• [13]W. Fan and D. Yeung, "Image Hallucination Using Neighbor Embedding over Visual Primitive Manifolds," Proc. IEEE Comput. Soc. Conf. Comp. Vision Pattern Recog., 2007, pp. 1-7.
• [14]M. Varma and A. Zisserman, "A Statiscal Approach to Texture Classification from Single Images," Int. J. Computer Vision, vol. 62, no. 1-2, 2005, pp. 61-81.
• [15]C. Kim, K. Choi, and J. Ra, "Improvement on Learning-Based Super-Resolution by Adopting Residual Information and Patch Reliability," IEEE Int. Conf. Image Process., 2009, pp. 1197-1200.