摘要
随着技术的发展,常需将已有的隔行视频转换成高质量的逐行视频。本文提出一种基于Navier-Stokes(N-S)方程的视频去隔行方法。将基于N-S方程的图像修复模型应用到视频空间,由理想三维流体方程,建立视频的局部约束条件;然后,根据其各向异性特征,导出单个像素点的修复方程;最后,通过对隔行视频间隔区域的逐点修复,实现视频去隔行。实验表明,本文方法能够较好的抑制运动虚像和边沿锯齿。
With the development of technology,the existing interlaced video needs to be converted into high-quality progressive video. A Navier-Stokes (N-S) equation based is video de-interlacing method proposed. The image inpainting model based on N-S equations is applied to the video space. According to the ideal three-dimensional fluid equation,a local constraint condition of video is established,which is anisotropic and can be used to derive a single pixel inpainting equation. Finally,video de-interlacing is achieved through the pixel inpainting in the interlaced region point by point. The experimental results show that,the proposed method has a better performance in suppressing virtual image and serrated edges than some other methods.
出处
《信号处理》
CSCD
北大核心
2010年第7期1073-1078,共6页
Journal of Signal Processing