PDF下载 分享
[1]陈光拓,孙 洋,杜雨洺,等.基于小波的超分辨率算法研究及FPGA实现[J].成都信息工程大学学报,2017,(06):601-604.[doi:10.16836/j.cnki.jcuit.2017.06.005]
　CHEN Guang-tuo,SUN Yang,DU Yu-ming,et al.Research and FPGA Implementation of Super-resolution Algorithm based on Wavelet Transform[J].Journal of Chengdu University of Information Technology,2017,(06):601-604.[doi:10.16836/j.cnki.jcuit.2017.06.005]
点击复制

基于小波的超分辨率算法研究及FPGA实现

成都信息工程大学学报[ISSN:1006-6977/CN:61-1281/TN] 卷: 期数: 2017年06期 页码: 601-604 栏目: 电子信息科学与技术 出版日期: 2017-05-30
Title:
Research and FPGA Implementation of Super-resolution Algorithm based on Wavelet Transform
文章编号:
2096-1618(2017)06-0601-04
作者:
陈光拓1;  孙 洋1;  杜雨洺1;  杨学博2
(1.成都信息工程大学电子工程学院,四川 成都 610225; 2.电子科技大学,四川 成都 611731)
Author(s):
CHEN Guang-tuo1;  SUN Yang1;  DU Yu-ming1;  YANG Xue-bo2
(1.College of Electronic Engineering,Chengdu University of Information Technology,Chengdu 610225,China; 2.University of Electronic Science and Technology of China, Chengdu 611731,China)
关键词:
超分辨率;  小波;  边缘估计;  边缘加强
Keywords:
super resolution;  wavelet;  edge enhancement;  interpolation
分类号:
TN911.73
DOI:
10.16836/j.cnki.jcuit.2017.06.005
文献标志码:
A
摘要:
由于超分辨率研究较火热,且在各个方面应用广泛,尤其是在家庭娱乐方面需求较大。如果利用纯软件的方法,处理速度会过慢,不适合于具体应用。利用基于小波效果较好的超分辨率算法,提出硬件实现方式,并给出各部分的实现方法以及设计实现。最后将整体以及各部分仿真,利用FPGA验证其方法的正确性。利用此方法增加图像分辨率可以在每秒30帧的情况下处理1024×1024大小图像,然而1024×1024大小的图像通过软件实现则需要较长时间。得到的硬件架构简练,使其更加容易应用到实际中。通过升级(并行处理)可以处理4 K图像。
Abstract:
The research of super-resolution is hot and it is widely used in many fields,especially in family entertainment.If you use a pure software approach,the processing speed will be too slow,so it is not suitable for specific applications.In this paper,a method of hardware implementation is proposed based on the super-resolution algorithm with better effect,and the implementation method and design implementation of each part are given.Finally,the whole and each part of the design are simulated,and the correctness of the method is verified by FPGA.This method can process 1024×1024 size images at 30 frames per second when increasing image resolution.However,1024×1024 size images require a much more longer time by software implementation.The hardware architecture of this article is concise,making it easier to apply to practice.4 K images can be processed by upgrading(parallel processing).

参考文献/References:

[1] 钟宝江,陆志芳,季家欢.图像插值技术综述[J].数据采集与处理,2016,31(6):1083-1096.
[2] 刘刚,戴明.基于区域分割自适应的超分辨率算法[J].微电子学与计算机,2012,29(1):76-79.
[3] 汤晓莉,韩睿,郭若杉.基于边缘指导的快速动态超分辨率算法[J].微电子学与计算机,2016,33(10):167-172.
[4] Akbarzadeh S,Ghassemian H,Vaezi F.An efficient single image super resolution algorithm based on wavelet transforms[C].Iranian Conference on Machine Vision and Image Processing.IEEE,2015:111-114.
[5] 公茂法,王志文,于江,等.基于FPGA小波变换核的设计[J].电测与仪表,2014,51(8):103-106.
[6] 赵阳,李国林,谢翔,等.一种邻域嵌入超分辨率算法[J].微电子学与计算机,2017,34(2):10-14.
[7] 谭会生.多级二维整数小波变换的FPGA实现研究[J].电子测量与仪器学报,2009(7):29-34.
[8] 刘怡,黄自力,王经纬,等.FPGA双线性插值图像变换系统的设计与实现[J].中国测试技术,2008,34(3):65-67.
[9] Matsuo Y,Iwasaki S,Yamamura Y,et al.Wavelet domain image super-resolution from digital cinema to ultrahigh definition television by dividing noise component[C].Visual Communications and Image Processing(VCIP),2012 IEEE.IEEE,2012:1-6.

相似文献/References:

[1]陈光拓,孙 洋,杜雨洺,等.一种改进的基于边缘加强超分辨率算法[J].成都信息工程大学学报,2017,(06):609.[doi:10.16836/j.cnki.jcuit.2017.06.007]
　CHEN Guang-tuo,SUN Yang,DU Yu-ming,et al.An Improved Super Resolution Algorithm based on Edge Enhancement[J].Journal of Chengdu University of Information Technology,2017,(06):609.[doi:10.16836/j.cnki.jcuit.2017.06.007]

备注/Memo

收稿日期:2017-09-07

更新日期/Last Update: 2017-05-30