[1]刘 哲,黄世奇,姜 杰.基于引导滤波和多尺度局部自相似单幅红外图像超分辨率方法[J].红外技术,2017,39(4):345-352.[doi:10.11846/j.issn.1001_8891.201704009]
 LIU Zhe,HUANG Shiqi,JIANG Jie.Single Image Super Resolution Method Based on Multi-scale Self-similarity and Non Local Means [J].Infrared Technology,2017,39(4):345-352.[doi:10.11846/j.issn.1001_8891.201704009]
点击复制

基于引导滤波和多尺度局部自相似单幅红外图像超分辨率方法
分享到:

《红外技术》[ISSN:1001-8891/CN:CN 53-1053/TN]

卷:
39卷
期数:
2017年第4期
页码:
345-352
栏目:
出版日期:
2017-04-20

文章信息/Info

Title:
Single Image Super Resolution Method Based on Multi-scale
Self-similarity and Non Local Means
文章编号:
1001-8891(2017)04-0345-08
作者:
刘 哲黄世奇姜 杰
 西京学院信息工程学院,陕西 西安 710123
Author(s):
 LIU ZheHUANG ShiqiJIANG Jie
 Department of Electronic and Information Engineering, Xijing University, Xi’an 710123, China
关键词:
 超分辨率多尺度局部自相似非局部均值
Keywords:
super resolutionmulti-scaleself-similaritynon-local mean
分类号:
O121.8,G558
DOI:
10.11846/j.issn.1001_8891.201704009
文献标志码:
A
摘要:
本文针对红外图像存在分辨率不高、对比度低的特点,提出了基于引导滤波和多尺度局部自相似性红外单幅图像超分辨率算法。首先,该方法引进了类高斯分布的“类高斯核”,在此基础上构建均值引导滤波器,该滤波器是一种线性边缘保持滤波器,可以得到图像的高频细节。再次,根据图像的自相似性,对初始高分辨率图像和原始低分辨率图像进行分块,得到待匹配窗和搜索窗,根据非局部均值(NLM),待匹配窗图像块的值利用搜索窗中相似块的加权平均计算得到。其次,利用图像自相似性,待匹配窗在搜索窗的邻域内进行匹配搜索,找到与待匹配窗最相似的匹配块,计算出最佳匹配块的高频细节图像块,与相似块的加权平均值相加,重构出高分辨率待匹配窗。最后,合并所有的超分辨率重构的待匹配窗,相邻图像块重叠区域的像素值使用平均融合得到,得到最终的超分辨率图像。实验结果表明,本文算法不仅能很好重构图像的高频细节,还能很好的恢复图像的纹理特征,得到的结果不仅边缘更清晰更真实,而且纹理更加丰富。
Abstract:
This paper is aimed at the characteristics of low resolution and low contrast of infrared image ,a new super resolution method from a single infrared image based on guided filtering and multi-s cale Local self-similarity is proposed.First of all, This method intr oduces a Gauss kernel, which is similar to the Gauss distribution. Based on this, we construct the mean direct filter,the filter is a linear edge preserving filter, which can get the high frequency details of the image. Again, according to the self similarity of the image, the initial high resolution image and low resolution of the orig inal image is divided into patchs, to obtain the matching window and the searching window, according to the non local mean (N LM), the value of matching window is the weighted mean value of searching window. Secondly, the use of image self similarity, matching window searching and matching in the neighborhood search window, and find the matching window matching patch is the most similar, calculate the b est matching image detail patchs, together with the weighted average value of similar blocks, to reconstruct high resolution matching window. Experimental results show that the proposed algorithm can not only reconstruct the high frequency details of the image, but also restore the texture features of the image. The results obtained are not only more clear and real, but also rich in texture

参考文献/References:

 [1] Nasrollahi K, Moeslund T B. Super-resolution: a comprehensive survey[J]. Machine Vision and Applications, 2014, 25(6): 1423-1468. doi:10.1007/ s00138-014-0623-4.
[2] Van Ouwerkerk J D. Image super resolution survey[J]. Image and Vision Computing, 2006, 24(10):1039-1052.doi:10.1016/j.imavis.2006.02.026.
[3] Protter M, Elad M, Takeda H, et al. Generalizing the nonlocal-means to super-resolution reconstruction[J]. IEEE Transactions on Image Processing, 2009, 18(1): 36-51.doi:10.1109/TIP.2008.2008067.
[4] Mairal J, Bach F, Ponce J, et al. Non-local sparse models for image restoration[C]//The 12th IEEE International Conference on Computer Vision,Kyoto, Japan, 2009: 2272-2279.
[5] Suetake N, Sakano M, Uchino E. Image super-resolution based on localself similarity[J]. Optical Review, 2013, 15(1): 26 -30 .doi:10.1007/ s10043-008-0005-0.
[6] Glasner D, Bagon S, Irani M. Super-resolution from a single image[C] //The 12th IEEE International Conference on Computer Vision, Kyoto, Japan, 2009: 349-356 . doi:10.1109/ICCV.2009.5459271.
[7] Freedman G, Fattal R. Image and video up scaling from localself -examples[J]. ACM Transactions on Graphics, 2011, 30(2):1-11. doi: 10.1145/1944846.1944852.
[8] Kawano H, Suetake N, Cha B, et al. Sharpness preservingimage enlargement by using self-decomposed codebook and Mahalanob is distance[J]. Image and Vision Computing, 2014, 27(6): 684-693. doi:10.1016/j.imavis.2008.07.013
[9] ZHANG Y Q, LIU J Y, YANG W H, et al. Image super-resolution based on structure-modulated sparse representation[J]. IEEE Transactions on Image Processing, 2015, 24(9): 2797-2810. doi:10.1017/ s11444-014 -5768-4.
[10] ZHANG Y Q, XIAO J S, LI S H, et al. Learning block-structured incoherent dictionaries for sparse representation[J]. Science China Information Sciences, 2015, 58(10): 1-15. doi:10.1007/s 11432-014-5258-6.
[11] YANG C Y, HUANG J B, YANG M H. Exploiting self-similarities for single frame super-resolution[C]//Proceedings of the 10th Asian Conference on Computer Vision (ACCV), Queen-stown, New Zealand: Springer, 2016: 497-510.
[12] Glasner D, Bagon S, Irani M. Super-resolution from a single image[C]//The 12th IEEE International Conference on Computer Vision (ICCV). Kyoto, Japan, 2009: 349-356. doi:10.1109/ ICCV.2009. 5459271.
[13] Schulter S, Leistner C, Bischof H. Fast and accurate image up scaling with super-resolution forests[C]//Proceedings of the 2015 IEEE Conference on Computer Vision and Pattern Recognition, Boston, MA, 2015: 3791-3799.
[14] Sajjad M, Ejaz N, Baik S W. Multi-kernel based adaptiveinterpolation for image super-resolution[J]. MultimediaToolsand Applications, 2015, 72(3): 2063-2085. doi:10.1109/CVPR.2015.7299003.
[15] Catalin Ionescu, Orestis Vantzos, Cristian Sminchisescu. Matrix back propagation for deep networks with structured layers[C]//ICCV, 2015: doi:10.1101/ ICCV.2015.65789987.
[16] Getreuer P. Contour stencils: total variation along curvesfor adaptive image interpolation[J]. SIAM Journal on Imaging Sciences, 2011, 4(3): 954-979. doi:10.1137/100802785.
[17] XIAO Jinsheng, DU Kanghua, TU Chaoping, et al. Bokeh display based on depth information extraction of multi-focus images[J]. Acta Automatica Sinica, 2015,41(2): 304-311.
[18] XIAO J S, LI W H, LIU G X, et al. Hierarchical tone mapping based on image colour appearance model[J]. IET ComputerVision, 2015, 8(4): 358-364.

相似文献/References:

[1]陈海欣,顾国华,钱惟贤,等. 基于TMS320C6678的背景更新算法的改进与实现[J].红外技术,2013,35(01):021.
 CHEN Hai-xin,GU Guo-hua,QIAN Wei-xian,et al. Improvement and Implementation of Background Updating Algorithm Based on TMS320C6678[J].Infrared Technology,2013,35(4):021.
[2]陆凯,李成金,赵勋杰,等. 一种快速的亚像素图像配准算法[J].红外技术,2013,35(01):027.
 LU Kai,LI Cheng-jin,ZHAO Xun-jie,et al.A Fast Sub-pixel Image Registration Algorithm[J].Infrared Technology,2013,35(4):027.
[3]余慧娟,黎兰,刘相新,等. 地面目标红外成像仿真系统的设计与实现[J].红外技术,2013,35(01):031.
 YU Hui-juan,LI Lan,LIU Xiang-xin,et al. Design and Realization of Ground Objects Infrared Scene Simulator[J].Infrared Technology,2013,35(4):031.
[4]郭水旺,王宝红,季钢,等.基于基因表达式编码算法的红外图像轮廓提取[J].红外技术,2013,35(01):038.
 GUO Shui-wang,WANG Bao-hong,JI Gang,et al. Infrared Image Contour Extraction Based on the Gene Expression Coding Algorithm[J].Infrared Technology,2013,35(4):038.
[5]王兵学,雍杨,霍义华,等. 基于纹理特征分析的海天线检测方法[J].红外技术,2013,35(01):042.
 WANG Bing-xue,YONG Yang,HUO Yi-hua,et al. Sea-sky-line Detection Algorithm Based on Texture Analysis[J].Infrared Technology,2013,35(4):042.
[6]徐航,孟卫华,周珩,等. 半透明平板对辐射亮度的衰减影响分析[J].红外技术,2013,35(01):047.
 XU Hang,MENG Wei-hua,ZHOU Heng,et al.An Analysis on the Influence of Radiance Attenuation from a Semi-transparent Slab[J].Infrared Technology,2013,35(4):047.
[7]孙爱平,皮冬明,安长亮,等. 光机装校阶段红外与可见光图像配准技术研究[J].红外技术,2013,35(01):050.
 SUN Ai-ping,PI Dong-ming,AN Chang-liang,et al. Study on IR/Visible Image Registration for Lens Assembly[J].Infrared Technology,2013,35(4):050.
[8]王宇,汤心溢,刘鹏,等. 拼接型长线列红外双波段图像实时传输系统[J].红外技术,2013,35(02):069.
 WANG Yu,TANG Xin-Yi,LIU Peng,et al. A Real-Time Transmission System of Splicing Long Linear Dual-Band IR Image[J].Infrared Technology,2013,35(4):069.
[9]潘胜达,林长青. 512×8中波TDI红外探测器成像系统设计[J].红外技术,2013,35(02):074.
 PAN Sheng-da,LIN Chang-qing.Design of Scanning Imaging System Based on 512×8 MW TDI IRFPA[J].Infrared Technology,2013,35(4):074.
[10]秦金明,陈宝国,李丽娟,等. 4×128双色线列红外探测器成像电路设计[J].红外技术,2013,35(02):078.
 QIN Jin-ming,CHEN Bao-guo,LI Li-juan,et al. Imaging Circuit Design of 4×128 Two-color Linear Array Infrared Detector[J].Infrared Technology,2013,35(4):078.

备注/Memo

备注/Memo:
收稿日期:2016-11-02;修订日期:2016-12-28.
作者简介:刘哲(1972-),男,博士,教授级高工,研究方向为机器视觉、人工智能及模式识别。
基金项目:国家自然科学基金(61473237)资助。
更新日期/Last Update: 2017-04-18