中文摘要：

为了用一组低质量、低分辨率图像来产生高质量、高分辨率图像,提出了一种基于Perona-Malik（P-M）扩散的超分辨率图像重建方法。首先分析了lp范数的稳健性以及P-M扩散保持图像纹理和边缘的特点;将两者相结合,并加入了抑制图像明亮特征的调整项;最后给出了迭代格式进行迭代求解。实验结果表明,本文方法的峰值信噪比（PSNR）平均提高了0.85 dB,图像质量也得到了提高。

英文摘要：

In order to combine a sequence of low-resolution noisy blurred images to produce a higher resolution image,a method of super-resolution image reconstruction is presented based on P-M（Perna-Malik） diffusion.We analyze the robust of lp norm.And,we present the algorithm to the super-resolution image reconstruction based on P-M diffusion,and a regulator called energy condensation integral is introduced to prevent the fattening of bright spots or linear structures.The simulation results confirm that the improvement of PSNR is 0.85 dB averagely,and the quality of the image is also enhanced.