图像超分辨率重构(super resolution,SR)是指利用计算机将一幅低分辨率图像(low resolution,LR)或图像序列进行处理,恢复出高分辨率图像(high resolution,HR)的一种图像处理技术。HR意味着图像具有高像素密度,可以提供更多的细节,这些细节往往在应用中起到关键作用。
低分辨率图像的信噪比(SNR)和图像之间的配准误差对超分辨率重构效果的影响是该技术应用在遥感领域的瓶颈。
The influence of Signal-to-noise ratio (SNR) of low-resolution images and registration error among them on the super-reconstruction of images is the bottle-neck in the field of remote sensing.
该文分别从成像过程和压缩过程两个方面讨论了超分辨率重构过程需要考虑的因素,同时指出了运动估计和运动补偿在超分辨率重构中的重要性。
In this paper, we introduce the main factor during imaging and compression and discuss the importance of motion estimation and compensation in super-resolution construction.
但是,由于超分辨率图像重构本身是一个病态求逆过程,在重构图像中所产生的边缘振荡效应往往很难在重构过程中予以消除。
However, due to the ill-posed nature of the super-resolution image reconstruction, the ringing artifacts can not be easily removed in the reconstruction process.
应用推荐