深圳大学学报理工版

为解决传统插值方法对采集的视频图像放大时易出现块效应和细节退化的问题,在高级RISC微处理器(advanced RISC microprocessor,ARM)平台上设计并搭建基于小波变换算法的嵌入式超分辨率复原系统.以ARM CortexTM-A8处理器(S5PV210)为核心搭建硬件平台,采用嵌入式Linux系统搭建软件开发环境,设计基于V4L2接口的视频图像采集程序.最后将基于小波变换的超分辨率复原算法植入系统.实验证明,该系统处理的视频图像,比传统插值方法处理的视频图像取得了更高的峰值信噪比,能够恢复出更多细节信息.

In order to overcome the disadvantage of the traditional methods for magnifying collected video images by interpolation, we design an embedded super-resolution restoration system based on advanced RISC microprocessor(ARM)and a wavelet transform algorithm. The hardware platform is an ARM CortexTM-A8 processor(S5PV210)as the core processor and the software development environment is embedded Linux. We implement a video image collecting program based on V4L2 and a super-resolution restoration algorithm based on wavelet transformation. Experimental results show that the peak signal to noise ratio(PSNR)is better than those processed by frequently used interpolation methods, and more detail in the images can be restored.

引言
1 系统软硬件平台的搭建
2 视频图像采集程序的设计
3 超分辨率复原处理算法
4 实验结果
5 结语

图1 系统总体结构图<br/>Fig.1 Overall structure of the system

图1 系统总体结构图
Fig.1 Overall structure of the system

图2 视频图像采集流程图<br/>Fig.2 Flow chart for video image collection

图2 视频图像采集流程图
Fig.2 Flow chart for video image collection

图3 本算法流程<br/>Fig.3 The calculation process

图3 本算法流程
Fig.3 The calculation process

表1 不同算法得到的PSNR值<br/>Table 1 The PSNR values from different methods dB

表1 不同算法得到的PSNR值
Table 1 The PSNR values from different methods dB

图4 Barbara图像实验结果<br/>Fig.4 Results of experiment of Barbara

图4 Barbara图像实验结果
Fig.4 Results of experiment of Barbara

表2 本算法在不同平台上得到的PSNR值Table 2 PSNR values of the proposed method on different platformsdB

图5 视频图像的超分辨率处理<br/>Fig.5 An example of super-resolution processing of a video image

图5 视频图像的超分辨率处理
Fig.5 An example of super-resolution processing of a video image

[1] Su Binghua,Jin Weiqi,Niu Lihong,et al.Super-resolution image restoration and progress[J]. Optical Technique,2001,27(1):6-9.(in Chinese)
[2] Zhang Jiang.Research on image restoration technology based on sparse representation model[D]. Harbin:Harbin Institute of Technology,2014.(in Chinese)
[3] Wang Jinjun,Zhu Shenghuo,Gong Yihong.Resolution enhancement based on learning the sparse association of image patches[J].Pattern Recognition Letters,2010,31(1):1-10.
[4] Zhuo Li,Wang Suyu,Li Xiaoguang,et al.Super-resolution restoration of images and videos[M].Beijing:People's Posts and Telecommunications Press,2008.(in Chinese)
[5] Wei Dongshan.Full manual for application development of embedded Linux[M].Beijing:People's Posts and Telecommunications Press,2008.(in Chinese)
[6] Lin Rimao,Su Binghua,Wu Xiaobing,et al. Image super resolution technique based on wavelet decomposition implemented on OMAP3530 platform[C]// Proceedings of the 3rd International Conference on Multimedia Information Networking and Security(MINES).Shanghai(China):IEEE Press,2011:69-72.
[7] Li Yucheng, Sun Peng.The design of embedded video-based vehicle tracking system[C]// Proceedings of 4th International Conference on Digital Manufacturing and Automation(ICDMA).Qindao(China):IEEE Press,2013: 1437-1440.
[8] Thuong L T,Khoa L C,Cao B T,et al.An improvement of curvelet based super-resolution image processing implemented on ARM AT91SAMRL [C]// Proceeding of the IEEE International Conference on Advanced Technologies for Communications(ATC).Ho Chi Minh City(Vietnam):IEEE Press,2013:380-385.
[9] Lu Yinli,Yu Hongli,Zhang Pengpeng.The implementation of embedded image acquisition based on V4L2 [C]// Proceeding of the IEEE International Conference on Electronic, Communications Control(ICECC).Ningbo(China):IEEE Press,2011:549-552.
[10] Wang Huiyang,You Xiangdong, Wang Ran. Design of image capture and transmission embedded system for remote monitoring[C]// Proceeding of the 8th International Conference on Information Science and Digital Content Technology(ICIDT).Jeju Island(South of Korea):IEEE Press,2012,3: 661- 664.
[11] Tham J Y,Shen Lixin,Lee S L.A general approach for analysis and application of discrete multiwavelet transforms[J].IEEE Transactions on Transactions on Signal Processing,2000,48(2):457- 464.
[12] Ashwini Devi S,Vasuki A.Image super resolution using fourier-wavelet transform[C]// Proceeding of the IEEE International Conference on Machine Vision and Image Progressing(MVIP).Taipei(China):IEEE Press,2012:109-112.
[13] Matsuo Y,Misu T,Sakaida S,et al.Video coding with wavelet image size reduction and wavelet super resolution reconstruction[C]// Proceeding of the 18th IEEE International Conference on Image Progressing(ICIP).Brussels:IEEE Press,2011:1157-1160.
[14] Chavez-Roman H, Ponomary V.Super resolution image generation using wavelet domain interpolation with edge extraction via a sparse representation[J].IEEE Geoscience and Remote Sensing Letters,2014,11(10): 1777-1781.
[15] Su Binghua.Research of super-resolution image resto-ration[D].Beijing:Beijing Institute of Technology, 2002.

备注

引言

1 系统软硬件平台的搭建

2 视频图像采集程序的设计

3 超分辨率复原处理算法

4 实验结果

5 结语

期刊信息

备注