Font Size: a A A

Research Of Super-resolution Restoration Of Video Images Based On Embedded ARM

Posted on:2016-03-24Degree:MasterType:Thesis
Country:ChinaCandidate:W S QiuFull Text:PDF
GTID:2308330464956911Subject:Optical Engineering
Abstract/Summary:PDF Full Text Request
Along with the development of digital media technology, a large number of video and image information influx into daily lives and various fields of the national economy,and the people would like to obtain more and more high-resolution images. The super-resolution image restoration technology can improve the image’s resolution and get more details and information without changing the existing imaging equipment. So it becomes one of the hot research areas in recent years. The development of super-resolution enhancement technology depends on the super-resolution enhancement algorithm and its hardware implementation. They are distinct, self-contained but closely related. The purpose of implementing the super-resolution enhancement algorithms in hardware is to enhance the processing speed to reach the real-time effect and the realization in the embedded equipments. It can be embedded to exited system to improve its performance or integrated to a new system to improve the performance and competitiveness of the new production.Based on research of exited super resolution(SR) restoration algorithm, this paper selected an algorithm based on wavelet transform, and wrote C program, which is proved to be better than the traditional interpolation algorithm by experiment. Then this paper built soft and hard system of SR video on Tiny210 and Zed Board both with ARM core. Hard system was made up of USB camera as input module, ARM processor as core algorithm module, and LCD as output module. Embedded Linux system was transplanted and video collected and display system based on V4L2 was designed. And the SR algorithm was transplanted into the video system to achieve the SR restoration progress and display of the collected video images from camera. Finally the effect and performance of both platforms was compared. The results show that these two platforms have same effect of SR restoration which is close to PC platform and they are better than the traditional interpolation algorithm. The Zed Board platform has better performance on speed than Tiny210 platform for its NEON coprocessor integrated on ARM inner core.
Keywords/Search Tags:Super-resolution restoration, wavelet transformation, ARM, Tiny210, Zed Board
PDF Full Text Request
Related items