【摘要】 利用正交离散小波变换,通过最小化广义交叉确认函数求取小波变换渐近最优阈值,与L-R超分辨力图像复原算法结合,提出了基于正交离散小波变换的L-R法(WL-R)和单次运算小波变换的L-R法(SWL-R).算法能够较好地复原信噪比较低的图像细节,SWL-R只需1次处理过程,在保证复原处理快速有效的基础上,减少了L-R算法的大量迭代运行时间,适合于实时化的超分辨力图像处理,具有广泛的应用前景.更多还原

【Abstract】 The new wavelet image restoration (WL-R) algorithm and the new single wavelet image restoration algorithm (SWL-R) based on orthogonal discrete wavelet transform are proposed by using the orthogonal discrete wavelet transform and generalized cross validation in evaluating the optimal threshold δ, and combining with Luck-Richardson super-resolution image restoration algorithm (L-R). The simulation experiments and processing results show that the proposed algorithms could retain image details better, and be more suitable to the low signal-noise-ratio images. Especially SWL-R needs operating only once, under the guarantee of the rapid and effective restoration processing, it could also reduce the operation time a lot in comparison with L-R. They are suitable to the real time processing of super-resolution image restoration, with an extensive application foreground.更多还原