摘要
为了提高重构图像或者视频的分辨率,提出把新型的基于光流法的图像配准算法应用于迭代反投影(IBP)超分辨率算法中。在所提出的方法中,基于光流法的图像配准算法用来提高图像配准的准确性。首先,为了得到像素级别的运动矢量,基于光流法的图像配准算法被用于估计图像间的运动矢量,以得到更加准确的运动矢量矩阵。接着,利用所获得的运动矢量矩阵结合迭代反投影算法重构高分辨率的图像。同时,由于基于光流法的图像配准能够很好地估计视频图像间的运动,所提出的方法同样适用于视频图像的超分辨。实验结果表明,提出的方法对于图像或者视频的超分辨率效果,在主观效果和客观评价上都有一定的提升。
To improve the spatial resolution of reconstructed images and videos, proposes a super-resolution (SR) reconstruction method using iteratire back projection with a new optical flow based image matching. In the proposed method, optical flow based image matching is em- ployed to improve the accuracy of image registration. First, optical flow based image matching is performed between images to obtain pix- el-level motion fields. Then, high-resolution (HR) images are reconstructed based on the motion fields using iterative back projection (IBP) algorithm. Since the optical flow based image matching can adapt to various types of motions in real video scenes, the proposed method can be used in video super resolution. Experimental results demonstrate that, subjective and objective quality improvements are obtained by the proposed algorithm, when applied to image or video sequence.
基金
国家自然科学基金(No.61102135)
关键词
图像超分辨
视频超分辨
光流法
迭代反投影
Image Super Resolution
Video Super Resolution
Optical Flow
herative Back Projection