摘要
为了克服终端设备硬件条件限制或传输过程中带宽限制而导致的获取图像和视频分辨率质量较低的问题,图像超分辨率技术作为计算机视觉领域中提升图像质量的重要技术,利用已知的低分辨率图像可恢复出高分辨率图像,现已广泛应用于卫星遥感、数字娱乐、视频监控等多个方面。围绕基于压缩感知理论的超分辨率方法,详细介绍将压缩感知应用于超分辨率的理论基础,列举几种典型的基于压缩感知的超分辨率方法,并在几种常用数据集上进行仿真对比。最后,对图像超分辨率的未来研究方向进行展望。
Due to the limitation of hardware or bandwidth in the process of transmission,the image and video resolution is usually of low quality.As an important technology to improve image quality in the computer vision,super-resolution can reconstruct high-resolution images from known low-resolution images,which has been widely used in satellite remote sensing,digital entertainment,video monitoring and other fields.The theory foundation of compressed sensing based super-resolution is discussed in detail.Some typical super-resolution methods based on compressed sensing are listed,and the experimental results are compared on several common data sets.Finally,the conclusion is drawn and the future research direction is prospected.
作者
李莹华
乔杨歌
刘颖
卢津
王富平
LI Yinghua;QIAO Yangge;LIU Ying;LU Jin;WANG Fuping(Key Laboratory of Electronic Information Application Technology for Crime Scene Investigation,Ministry of Public Security,Xi'an University of Posts and Telecommunications,Xi'an 710121,China;Center for Image and Information Processing,Xi'an University of Posts and Telecommunications,Xi'an 710121,China;International Joint Research Center for Wireless Communication and Information Processing Technology of Shaanxi Province,Xi'an University of Posts and Telecommunications,Xi'an 710121,China)
出处
《西安邮电大学学报》
2021年第2期87-95,共9页
Journal of Xi’an University of Posts and Telecommunications
关键词
压缩感知
超分辨率
字典学习
稀疏表示
自相似性
compressed sensing
super-resolution
dictionary learning
self-similarity
