|Table of Contents|

Super-resolution Restoration Algorithm Based onSVM Multi-figure Classification Learning(PDF)

《南京师大学报（自然科学版）》[ISSN:1001-4616/CN:32-1239/N]

Issue:
2018年03期
Page:
28-
Research Field:
·人工智能算法与应用专栏·
Publishing date:

Info

Title:
Super-resolution Restoration Algorithm Based onSVM Multi-figure Classification Learning
Author(s):
Tang Jiali1; Zhu Guangping1; Du Zhuoming1; 2
(1.College of Computer Engineering,Jiangsu University of Technology,Changzhou 213001,China)(2.School of Mathematical Sciences,Nanjing Normal University,Nanjing 210023,China)
Keywords:
super-resolution restoration; support vector machine; multi-figure classification; sample learning
PACS:
TP391
DOI:
10.3969/j.issn.1001-4616.2018.03.005
Abstract:
The SVM pre-classified super-resolution algorithm is based on single image feature and builds off-line disaggregated models. It reduces the mis-matching of tranditional example-based restoration algorithms,improves the image quality and running speed. However,the SVM-based algorithm easily leads to unstable recovered results because of the diversity of image features. For such problems,we propose a super-resolution restoration algorithm based on multi-figure classification learning. The algorithm saves the corresponding color-texture information into the sample set and selects object subset by SVM pre-classified learning. Then in the high frequency prediction process it makes precise matching search from the subset of sample database which has similar color and texture features with the object image. Experimental results show that compared with traditional algorithms,PSNR and SSIM are improved respectively. In addition,the proposed algorithm further reduces the matching range of low resolution image blocks and promotes the restoration effectiveness.

References:

[1] WANG J J,ZHU S,GONG Y. Resolution enhancement based on learning the sparse association of image patches[J]. Pattern recognition letters,2010,31(1):1-10.
[2]KURSUN O,FAVOROV O. Single-frame super resolution by inference from learned features[J]. Istanbul university journal of electrical and electronics engineering,2003,3(1):673-681.
[3]KATSUKI T,TORII A,INOUE M. Posterior-mean super-resolution with a causal Gaussian Markov random field prior[J]. IEEE transactions on image processing,2012,21(7):3182-3193.
[4]TANG Y,YAN P K,YUAN Y,et al. Single-image super-resolution via local learning[J]. International journal of machine learning and cybernetics,2011,2(1):15-23.
[5]GAO X B,ZHANG K B,TAO D C,et al. Image super-resolution with sparse neighbor embedding[J]. IEEE transactions on image processing,2012,21(7):3194-3205.
[6]BABACAN S D,MOLINA R,KATSAGGELOS A K. Variational bayesian super resolution[J]. IEEE transactions on image processing,2011,20(4):984-999.
[7]YANG J C,WRIGHT J,HUANG T,et al. Image super-resolution via sparse representation[J]. IEEE Transactions on image processing,2010,19(11):2861-2873.
[8]FREEMAN W T,JONES T R,PASZTOR E C. Example-based superresolution[J]. IEEE computer graphics and applications,2002,22(2):56-65.
[9]TRINH D H,LUONG M,DIBOS F,et al. Novel example-based method for super-resolution and denoising of medical images[J]. IEEE transactions on image processing,2014,23(4):1882-1895.
[10]JEONG S,YOON I,JEON J,et al. Multi-frame example-based super-resolution using locally directional self-similarity[C]//Proc of IEEE International Conference on Consumer Electronics,LV:IEEE,2015:631-632.
[11]VAPNIK V. Statistical learning theory[M]. New York:John Wiley and Sons,1998.
[12]WANG Z,BOVIK A C,SHEIKH H R,et al. Image quality assessment:from error measurement to structural similarity[J]. IEEE Transactions on image processing,2004,13(4):600-612.
[13]梅树立. 基于变分法和剪切波耦合算法的蝗虫切片保纹理图像降噪[J]. 农业工程学报,2016,32(17):152-159.
[14]朱志刚,林学訚,石定机. 数字图像处理[M]. 北京:电子工业出版社,2011.
[15]汤嘉立,左健民,黄陈蓉. 基于SVM预分类学习的图像超分辨率重建算法[J]. 计算机应用研究,2012,29(8):3151-3175.
[16]张卫国,李景妹. 改进的基于纹理特征的图像配准算法[J]. 计算机工程与应用,2016,52(6):214-218.
[17]曹杨,李晓光,王素玉,等. 基于预分类学习的超分辨率复原算法[J]. 数据采集与处理,2009,24(4):514-518.
[18]柳益君,朱广萍,钱进,等. 基于支持向量机的绿色战略选择模型研究[J]. 计算机仿真,2010,27(11):307-310.

Memo

Memo:
-
Last Update: 2018-11-19