摘要
凸集投影(POCS)是一种把多帧低分辨率图像重构为高分辨率图像的算法,但是该算法在图像的高频信息恢复和降噪方面不是很理想。在本文中,在凸集投影(POCS)算法的基础上利用小波变换来提取出隐藏在低分辨图像中的高频信息和降低低分辨图像的噪声,可以更好地恢复出图像的细节并提高信噪比。实验仿真的结果表明,此方法在细节重构方面确实更优于凸集投影(POCS)算法。
Projection on Convex Sets（POCS） is a algorithm which products high-resolution image from a set of lowresolution images, but doesn＇ t perform very well on the reconstruction of high frequency information and depressing noise. In this paper, wavelet transform is utilized to extract high frequency information hidden and depress the noise in the low resolution images based on POCS, thus the detail information of images and SNR are better than the results of normal POCS. The results of simulation confirm that the method in this paper is more effective than POCS algorithm.
出处
《中国西部科技》
2009年第25期48-50,共3页
Science and Technology of West China
关键词
超分辨
凸集投影
小波变换
Super-resolution
POCS （Projection on Convex Sets）
Wavelet transform
