Single image super-resolution reconstruction using multiple dictionaries and improved iterative back-projection

Abstract

In order to improve the super-resolution reconstruction effect of the single image, a novel multiple dictionaries learning via support vector regression (SVR) and improved iterative back-projection (IBP) are proposed. To characterize the image structure, the low-frequency dictionary is constructed from the normalized brightness of low-frequency image patches in a discrete-cosine-transform (DCT) domain. Pixels determined by Gaussian weighting are added to the input vector to restore more high-frequency information when training the high-frequency image patch dictionary in the space domain. During post-processing, the improved IBP is employed to reduce regression errors each time. Experiment results show that the peak signal-to-noise ratio (PSNR)and structural similarity (SSIM) of the proposed method are enhanced by 1.6%–5.5% and 1.5%–13.1% compared with those of bicubic interpolation, and the proposed method visually outperforms several algorithms.

References

[1] ChanziL, QingchunC, HengchaoL. Multimedia Tools and Applications, 2017, 76: 14759

[2] QiY, YanzhuZ, TiebiaoZ, YangquanC. ISA Transactions, 2017, 82: 163

[3] Xiang-junZ, Xiao-linW. IEEE Transactions on Image Processing, 2008, 17: 887

[4] Jafari-KhouzaniK. IEEE Transactions on Medical Imaging, 2014, 33: 1969

[5] Shao-shengD, Jin-songL, Hai-yanX, Zhi-huiD, QinL. Optoelectronics Letters, 2014, 10: 313

[6] YangJ, WangZ, LinZ, CohenS, HuangT. IEEE Transactions on Image Processing, 2012, 21: 3467

[7] Wang Zhang-yang, Yang Ying-zhen, Wang Zhao-wen, Chang Shi-yu, Han Wei, Yang Jian-chao and Thomas S. Huang, Self-Tuned Deep Super Resolution, IEEE Conference on Computer Vision and Pattern Recognition Workshops, 1 (2015).

[8] Yuan-feiH, JieL, Xin-boG, Li-huoH, WenL. IEEE Transactions on Image Processing, 2018, 27: 5904

[9] HuangD-T, HuangW-Q, HuangH, ZhengL-X. Optoelectronics Letters, 2017, 13: 439

[10] Radu Timofte, Vincent De Smet and Luc Van Gool, A+: Adjusted Anchored Neighborhood Regression for Fast Super-Resolution, Asian Conference on Computer Vision, 111 (2014).

[11] HongW, Fang-fangL, Jian-wuL. Journal of Image and Graphics, 2016, 21: 986(in Chinese)

[12] Qi-pingY, Hai-jieL, Zhi-hongC, Xiao-pingY. Optics and Precision Engineering, 2016, 24: 2302 in Chinese)

[13] NiK S, NguyenT Q. IEEE Transactions on Image Processing, 2007, 16: 1596

[14] Zhi-zhouL. Dictionary Learning Based Super-Resolution Image Reconstruction, Xian University of Electronic Technology, 2011, (in Chinese)

[15] Feng-lainL, Meng-yaoS, Wen-naC. Optoelectronics Letters, 2017, 13: 237

[16] Chih-chungC, Chih-jenL. ACM Transactions on Intelligent Systems and Technology, 2011, 2: 27

[17] DongC, LoyC C, HeK, TangX. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2016, 38: 295