基于空间自适应正则化的超分辨率重建算法

针对模糊图像的复原问题，从正则化技术克服问题病态性的思想出发，研究了一种有效的超分辨率重建算法。该算法充分考虑了图像的局部特性，引入了空间自适应加权矩阵，采用全局正则化参数与局部正则化参数矩阵相结合的方法，弥补了传统正则化方法所带来的正则化误差以及噪声放大误差。实验结果表明，该算法能够有效地减少重建误差，保护图像的细节信息。

Super resolution reconstruction algorithm based on spatial adaptive regularization

Aiming at the restoration of blurred image, an effective approach to super resolution image restoration was proposed based on regularization technique for dealing with ill-posed problem. The new algorithm gave full consideration to local characteristic of the image, and introduced spatial adaptive weighting matrix. By combining global regularization parameter with local regularization matrix, the new algorithm overcame regularization error and noise amplification error which were generated by regularization. The results of the experiments indicated that the new algorithm could decrease reconstruction error and protect the detail information in high resolution image.