一种自适应正则化的图像超分辨率算法

安耀祖; 陆耀; 赵红

doi:10.3724/SP.J.1004.2012.00601

一种自适应正则化的图像超分辨率算法

doi: 10.3724/SP.J.1004.2012.00601

安耀祖^1, ,,
陆耀¹,
赵红^1,2

1.
北京理工大学计算机学院智能信息技术北京市重点实验室北京 100081;
2.
河北大学数学与计算机学院保定 071002

详细信息

通讯作者:
安耀祖北京理工大学计算机学院博士研究生. 主要研究方向为图像处理,图像及视频超分辨率. E-mail: bitanyz@bit.edu.cn

计量
- 文章访问数: 2451
- HTML全文浏览量: 99
- PDF下载量: 1115
- 被引次数: 0
出版历程
- 收稿日期: 2011-03-16
- 修回日期: 2011-10-18
- 刊出日期: 2012-04-20

An Adaptive-regularized Image Super-resolution

AN Yao-Zu^{1
, ,},
LU Yao¹,
ZHAO Hong^1,2

1.
Beijing Key Laboratory of Intelligent Information Technology, School of Computer Science and Technology, Beijing Institute of Technology, Beijing 100081;
2.
College of Mathematics and Computer Science, Hebei University, Baoding 071002

摘要

摘要: 提出一种自适应正则化的图像超分辨率重建算法. 首先, 利用局部残差均值自适应地计算各低分辨率图像通道的权值参数矩阵, 可有效地利用各通道对应区域间的交叉信息; 其次, 利用正则项局部误差均值自适应地计算平衡正则项和保真项的正则化参数矩阵, 能较好地保持图像边缘纹理等信息.实验结果表明本文算法不但具有较高峰值信噪比(Peak signal to noise ratio, PSNR) 和结构相似度(Structural similarity, SSIM), 而且在边缘、纹理等细节区域具有更好的重建效果.
- 超分辨率 /
- 最大后验估计 /
- 自适应正则化 /
- 邻域约束
Abstract: This paper presents an adaptive-regularized super-resolution method for image sequence. Firstly, an adaptive weight parameter matrix calculated by local residual mean is used to weight each low-resolution channel, which can utilize the information between channels sufficiently. Secondly, a new adaptive regularization parameter matrix calculated by the neighborhood mean of prior term is determined to balance prior term and fidelity term at each iteration, which can preserve edge and texture well. Experimental results indicate that the proposed method is of higher peak signal to noise ratio (PSNR) and structural similarity (SSIM) and has better reconstruction effect in edge and texture part.
- Super resolution /
- maximum a posteriori (MAP) /
- adaptive regularization /
- neighborhood constrains

HTML全文

参考文献(1)

[1]

Huang T S, Tsai R Y. Multiple frame image restoration and registration. Advances in Computer Vision and Image Processing. Greenwich: JAI Press, 1984. 317-339[2] Irani M, Peleg S. Improving resolution by image registration. CVGIP: Graphical Models and Image Processing, 1991, 53(3): 231-239[3] Yu Jing, Su Kai-Na, Xiao Chuang-Bai. Edge artifact reduction for super-resolution image reconstruction. Acta Automatica Sinica, 2007, 33(6): 577-582(禹晶, 苏开娜, 肖创柏. 一种改善超分辨率图像重建中边缘质量的方法. 自动化学报, 2007, 33(6): 577-582)[4] Zhang Dong-Ming, Pan Wei, Chen Huai-Xin. Spatio-temporal adaptive super-resolution reconstruction of video sequence based on MAP frame. Acta Automatica Sinica, 2009, 35(5): 484-490(张冬明, 潘炜, 陈怀新. 基于MAP框架的时空联合自适应视频序列超分辨率重建. 自动化学报, 2009, 35(5): 484-490)[5] Shao Wen-Ze, Wei Zhi-Hui. Super-resolution reconstruction based on generalized Huber-MRF image modeling. Journal of Software, 2007, 18(10): 2434-2444(邵文泽, 韦志辉. 基于广义Huber-MRF图像建模的超分辨率复原算法. 软件学报, 2007, 18(10): 2434-2444)[6] Yan Hua, Liu Ju. Super-resolution image restoration considering sub-pixel registration error. Acta Electronica Sinica, 2007, 35(7): 1409-1413(闫华, 刘琚. 考虑亚像素配准误差的超分辨率图像复原. 电子学报 2007, 35(7): 1409-1413)[7] Lee E S, Kang M G. Regularized adaptive high-resolution image reconstruction considering inaccurate subpixel registration. IEEE Transactions on Image Processing, 2003, 12(7): 826-837[8] He H, Kondi L P. Resolution enhancement of video sequences with adaptively weighted low-resolution images and simultaneous estimation of the regularization parameter. In: Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing. Montreal, Canada: IEEE, 2004. 213-216[9] Marquina A, Osher S. Image super-resolution by TV-regularization and Bregman iteration. Journal of Scientific Computing, 2008, 37(3): 367-382[10] Farsiu S, Robinson M D, Elad M, Milanfar P. Fast and robust multiframe super resolution. IEEE Transactions on Image Processing, 2004, 13(10): 1327-1344[11] Omer O A, Tanaka T. Region-based weighted-norm approach to video super-resolution with adaptive regularization. In: Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing. Taipei, China: IEEE, 2009. 833-836[12] Liu C, Shum H, Freeman W. Face hallucination: theory and practice. International Journal of Computer Vision, 2007, 75(1): 115-134[13] Gajjar P P, Joshi M V. New learning based super-resolution: use of DWT and IGMRF prior. IEEE Transactions on Image Processing, 2010, 19(5): 1201-1213[14] Wang J, Zhu S, Gong Y. Resolution enhancement based on learning the sparse association of image patches. Pattern Recognition Letters, 2010, 31(1): 1-10[15] Kim K, Kwon Y. Single-image super-resolution using sparse regression and natural image prior. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010, 32(6): 1127-1133[16] Yang J C, Wright J, Huang T S, Ma Y. Image super-resolution via sparse representation. IEEE Transactions on Image Processing, 2010, 19(11): 2861-2873[17] Glasner D, Bagon S, Irani M. Super-resolution from a single image. In: Proceedings of the 12th IEEE International Conference on Computer Vision. Kyoto, Japan: IEEE, 2009. 349 -356[18] Freedman G, Fattal R. Image and video upscaling from local self-examples. ACM Transactions on Graphics, 2011, 30(2): 12 1-11[19] Li X L, Hu Y T, Gao X B, Tao D C, Ning B J. A multi-frame image super-resolution method. Signal Processing, 2010, 90(2): 405-414[20] Katsaggelos A K, Biemond J, Schafer R W, Mersereau R M. A regularized iterative image restoration algorithm. IEEE Transactions on Signal Processing, 1991, 39(4): 914-929[21] Galatsanos N P, Katsaggelos A K. Cross-validation and other criteria for estimating the regularization parameter. In: Proceedings of the International Conference on Acoustics, Speech and Signal Processing. Toronto, Canada: IEEE, 1991. 3021-3024[22] Bose N K, Lertrattanapanich S, Koo J. Advances in super resolution using L-curve. In: Proceedings of the IEEE International Symposium on Circuits and Systems. Sydney, Australia: IEEE, 2001. 433-436

施引文献

资源附件(0)

访问统计