一种基于改进频域图像配准技术的超分辨力图像处理方法

为提高光电成像系统的空间分辨力,提出了一种基于改进的频率域图像配准技术的超分辨力图像处理方法。首先利用改进的频域图像配准方法估算出低分辨力图像之间的微位移量,然后采用Papoulis-Gerchberg超分辨力处理方法完成图像复原。利用不同重构方法进行了仿真及实验研究,给出了评价参数。模拟和实际显微热图像的处理结果表明:该算法可使图像质量得到改善,分辨的细节更多,可有效地提高光电成像系统的空间分辨力;处理算法简单,计算量小,可实现快速处理。该算法还可应用于其他不可控光学微扫描成像系统中,具有广泛的应用前景。

A method of superresolution image processing based on the improved frequency-domain image registration

In order to improve the spatial resolution of the designed photoelectric imaging system,based on the improved frequency-domain image registration techniques,a Papoulis-Gerchberg superresolution reconstruction method is presented. The micro displacement between low resolution images is estimated by the improved frequency-domain image registration techniques,and the image restoration is done with the Papoulis-Gerchberg superresolution image processing method. Different reconstruction methods are applied to reconstruct the simulated image and the real thermal microscope image,the evaluation parameters are given. Both the evaluation parameters and the visual effect show that the proposed algorithm is effective. The image spatial resoluion and quality are improved and the detail is more,the algorithm proposed is sample,fast and has low computational complexity. The algorithm can also be applied into the uncontrolled optical microscanning imaging systems and has wide application prospect.