【摘要】 本文论证了超分辨率图像复原计算中的两个性质,并基于此在MAP(Maximum A Posteriori)框架下提出了一种新的纹理自适应算法.算法首先根据低分辨率图像和高分辨率图像近似计算的可类比性质计算初始图像,使初始图像的质量更高,并根据超分辨率复原图像阶跃边缘的陡坡性质,将三边滤波正则化应用于迭代运算中,更好地保护了图像的陡坡和屋顶边缘.算法可根据图像的纹理自动计算初始图像融合参数以及正则化函数中的梯度阈值等参数,解决了以往超分辨率图像复原算法参数调整复杂的问题.实验结果表明,本文算法在没有人工参与的情况下,重建图像的客观评价和主观质量均有明显提高.更多还原

【Abstract】 Two properties of Super-Resolution calculation for image super-resolution reconstruction are proved,and a new texture adaptive super-resolution algorithm is proposed based on MAP(Maximum a posteriori)model.In Super-Resolution reconstruction process,an initial high resolution image with higher quality is calculated firstly by approximate calculation,and then trilateral model is applied as a regularization term in MAP model to preserve slope edge and roof edge.The fusion parameter and regularization parameter can be adaptively computed by image texture without manual adjustment.Simulation results confirm the effectiveness of this algorithm and demonstrate its superiority to other super-resolution algorithms.更多还原