Super-resolution video restoration system based on ZedBoard

In order to realize real-time super-resolution restoration of video image, a super-resolution video restoration system based on the ZedBoard programmable system on a chip was designed. The system includes the USB camera video acquisition based on video for Linux 2 (V4L2), the super-resolution restoration algorithm based on wavelet transform, and the graphical user interface and video output based on Qt. The bilinearity, double cube and wavelet transform methods were used to restore Lena image, and the peak signal noise ratio (PSNR) values between the original image and the restored images respectively are 29.516, 29.843 and 31.368,respectively. The experiment results show that the wavelet transform method proposed is better than traditional interpolation methods, and the super-resolution video restoration system based on ZedBoard is effective.