Multiframe Super-Resolution of Color Images Based on Cross Channel Prior

Color images have a wider range of applications than gray images. There are two ways to extend the traditional super-resolution reconstruction method color images: Super resolution reconstructs each channel image individually; Change RGB bands into YCrCb bands, then luminance component and interpolates chrominance components.These algorithms cannot effectively utilize property that edges textures similar in channels, results those methods may lead artifacts. Aiming solve these problems, we propose new based on cross prior. First, prior is proposed describe similarity gradient channels. Then, for via combination methods. Finally, channels alternately. The experimental show could suppress generation artifacts improve quality reconstructed