基于混合噪声模型的图像超分辨率重建  

A Super-resolution Image Reconstruction Method Based on Hybrid Noise Model

作　　者：成怡[1,2] 佟晓宇 修春波[1,2] 

机构地区：[1]天津工业大学天津市电工电能新技术重点实验室,天津300387 [2]天津工业大学自动化系,天津300387

出　　处：《控制工程》2017年第9期1903-1907,共5页Control Engineering of China

基　　金：国家自然科学基金(61203302);天津市应用基础与前沿技术研究计划项目(14JCYBJC18900)

摘　　要：为了获得更加理想的图像超分辨率重建结果,提出了一种基于混合噪声模型的图像超分辨率重建方法。首先利用L1范数在突变区域较好地保持图像边缘信息,然后采用L2范数抑制图像平坦区域的噪声,最后采用多幅图像进行超分辨率重建仿真实验。仿真结果表明,相对于其他重建算法,该算法提高了图像重建结果峰值信噪比,降低了图像重建的误差,可以较好的平滑和保持图像边缘细节信息,同时加快了图像重建的速度,可以满足重建的实时性要求。In order to obtain good reconstruction results of image super-resolution, a novel reconstruction method of image super-resolution is proposed based on the hybrid noise model in this paper. Firstly, Ll norm minimization is used to preserve the edge message of images, and then L： norm minimization is used to smooth the noises in images, finally, the experimental experiments are carried out to test the performance. The results show that the proposed algorithm has obtained higher peak signal to noise ratio and reduced the root mean square error, it can keep and smooth the edge message compared with other algorithms. At the same time, the proposed algorithm fastens the reconstructing speed and can meet the real-time requirements of image super resolution reconstruction.

关 键 词：图像超分辨率重建 混合噪声模型 峰值信噪比 均方根误差 

分 类 号：TN51[电子电信]