提出了基于小波融合的残差金字塔超分辨率方法.算法包括2部分:1) 通过残差金字塔分别估计多幅低分辨率图像的高频信息;2) 对获得的多幅高分辨率图像进行小波融合.图像的低频部分采用能量加权法融合,高频部分采用最大值规则融合,融合后经过小波反变换得到最终的高分辨率图像.研究结果表明,该算法避免了大量的迭代运算,降低了计算复杂度,使大尺寸图像的超分辨率成为可能.
The image residual pyramid super-resolution reconstruction based on wavelet fusioin is proposed. The algorithm consists of two parts:l) the high frequency information of several low resolution images is estimated by residual pyramid respectively;2) the high resolution images obtained by residual pyramid are fused in wavelet domain. In course of the fusion, the energy method is used to fuse the low frequency components, and the high frequency wavelet coefficients are fused by the maximum frequency rule. Finally,the inverse wavelet transform is used to obtain the high resolution image. The proposed algorithm avoids the application of iterative method, reduces the complexity of calculation and can be applied to large dimension low-resolution images. The results show the proposed approach has succeeded in obtaining a high-resolution image with a high neck signAl-to-noise ratio and good visual aualitv.