基于L1范数和正交梯度算子的超分辨率重建

靳晓娟; 邓志良

基于L1范数和正交梯度算子的超分辨率重建

靳晓娟^1, ,,
邓志良²

1.
江苏科技大学计算机科学与工程学院,江苏镇江 212003;
2.
常州信息职业技术学院,江苏常州 213164

详细信息

通讯作者:
靳晓娟（1987-），女，山东济宁人，硕士研究生，主要从事光学超分辨率重建研究工作。

中图分类号: TN911.73;TP391
计量
- 文章访问数: 4327
- HTML全文浏览量: 94
- PDF下载量: 827
- 被引次数: 0
出版历程
- 刊出日期: 2012-03-15

Super resolution reconstruction based on L1-norm and orthogonal gradient operator

JIN Xiao-juan^{1
, ,},
DENG Zhi-liang²

1.
Department of Computer Science and Engineering,Jiangsu University of Science and Technology,Zhenjiang 212003,China;
2.
Changzhou College of Information Technology,Changzhou 213164,China

摘要

摘要: 针对超分辨率图像重建的病态问题，设计了一种新的自适应超分辨率图像序列重建算法。该算法在L1范数重建框架下，利用金字塔算法与Lucas-Kanade算法相结合的方法实现图像配准，获得亚像素的运动估计；通过引入移位算子给出了基于正交梯度算子的正则项的实现方法，并从自适应的角度选择正则化参数，最后通过最速下降法求解模型的目标泛函最小值。结果表明：对于模拟实验和真实序列实验，该方法相比于样条插值算法、Tikhonov正则化算法、双边全变差重建算法都有一定的优势，能够取得更好的复原效果，并且由于正则项较为简单，重建所需时间相对减少。
- 超分辨率重建 /
- 正交梯度算子 /
- 图像配准 /
- 自适应 /
- L1范数
Abstract: For the ill-posed problem of super resolution reconstruction, a new adaptive algorithm for image sequence was proposed. The new algorithm was based on the framework of L1-norm. In the new algorithm, the pyramidal algorithm coupled with Lucas-Kanade algorithm was used for images registration to obtain the sub-pixel motion estimation. Displacement operator was introduced to achieve the regular term based on the orthogonal gradient operator and the regularization parameter was determined adaptively. Finally, the steepest descent method was used to solve the minimum of the constraint equation. The simulation experiments and the true sequence experiments show that the method proposed has advantages over spline interpolation, Tikhonov reconstruction and bilateral total variation reconstruction. On the one hand it can provide better reconstructing results, on the other hand the reconstruction time is reduced at the same time since the regularization item is simple.
- super resolution reconstruction /
- orthogonal gradient operator /
- image registration /
- adaptive /
- L1 norm

HTML全文

参考文献(1)

［1］吴俊, 吴桢. 合成孔径光学系统的成像特性和图像复原［J］. 应用光学, 2010, 31(4): 567-573.
WU Jun, WU Zhen. Imaging characteristic and image reconstruction of synthetic aperture optical system［J］. Journal of Applied Optics, 2010, 31(4): 567-573. (in Chinese with an English abstract)
［2］王桥. 数字图像处理［M］. 北京: 科学出版社, 2009.
WANG Qiao. Digital image processing［M］. Beijing: Science Press, 2009. (in Chinese)
［3］张翼飞, 李良福, 王娇颖, 等. 基于超分辨率重建的图像增强算法研究［J］. 应用光学, 2011, 32(2): 250-255.
ZHANG Yi-fei, LI Liang-fu, WANG Jiao-ying, et al. Image enhancement algorithm based on super resolution reconstruction［J］. Journal of Applied Optics, 2011, 32(2): 250-255. (in Chinese with an English abstract)
［4］曾三友, 康立山, 丁立新. 正则图像恢复中正则化算子选取的定性分析［J］. 计算机工程与应用, 2003, 39(15): 42-44.
ZENG San-you, KANG Li-shan, DING Li-xin. A qualitative analysis of the selection of regularization operator in image resolution［J］. Computer Engineering and Applications, 2003, 39(15): 42-44. (in Chinese with an English abstract)
［5］LUCAS B, KANADA T. An iterative image registration technique with an application to stereo vision［C］. Vancouver: Proceedings of the International Joint Conference on Artificial Intelligence, 1981: 674-679.
［6］FARSIU S, ROUBINSON M D, ELAD M, et al. Fast and robust multi-frame super resolution［J］. IEEE Transaction on Image Processing, 2004, 13(10): 1327-1344.
［7］张正贤. 正则超分辨率图像复原算法研究［D］. 西安: 西北工业大学, 2006.
ZHANG Zheng-xian. Research on reconstruction technology of super resolution based on regularization［D］. Xi-an: Northwestern Polytechnical University, 2006. (in Chinese)
［8］路庆春, 胡访宇. L1范数的图像超分辨率重建改进算法［J］. 信号与信息处理, 2009, 9(39): 13-15.
LU Qing-chun, HU Fang-yu. Improved image super-resolution resolution algorithm based on L1-norm［J］. Signal and Information Processing, 2009, 9(39): 13-15. (in Chinese with an English abstract)
［9］倚海伦, 王庆. 基于L1范数的图像超分辨率及差分统计模型［J］. 计算机工程, 2007, 33(24): 203-205.
YI Hai-lun, WANG Qing. L1-norm-based image super-resolution and its differential statistical model［J］. Computer Engineering, 2007, 33(24): 203-205. (in Chinese with an English abstract)
［10］沈焕锋, 李平湘, 张良培. 一种自适应正则MAP超分辨率重建方法［J］. 武汉大学学报:信息科学版, 2006, 31(11): 949-952.
SHEN Huan-feng, LI Ping-xiang, ZHANG Liang-pei. Adaptive regularized MAP super-resolution reconstruction method［J］. Geomatics and Information Science of Wuhan University, 2006, 31(11): 949-952. (in Chinese with an English abstract)
［11］狄红卫, 刘显峰. 基于结构相似度的图像融合质量评价［J］. 光子学报, 2006, 35(5): 766-770.
DI Hong-wei, LIU Xian-feng. Image fusion quality assessment based on structural similarity［J］. Acta Photonica Sinica, 2006, 35(5): 766-770. (in Chinese with an English abstract)
［12］MILANFAR P, FARSIU S, ROBINSON D. MDSP super resolution and demosaicing datasets［DB/OL］. ［2011-04-15］. http://www.soe.ucsc.edu/~milanfar/software/srdatasets.html.
［13］熊兴华, 张丽. 一种基于灰度预测误差统计的影像质量评价方法［J］. 中国图像图形学报, 2004, 9(3): 302-307.
XIONG Xing-hua, ZHANG Li. An image quality evaluation method based on gray prediction error［J］. Journal of Image and Graphics, 2004, 9(3): 302-307. (in Chinese with an English abstract)

施引文献

资源附件(0)

访问统计