联合运动估计和帧选择的多帧超分辨率重建算法  

Joint Motion Estimation and Frame Selection Algorithm for Multi-frame Super-resolution Reconstruction

作　　者：薛翠红[1] 于明[1] 高韬[2,3] 阎刚[1] 贾超[1] 李果[4] 

机构地区：[1]河北工业大学计算机科学与软件学院,天津300401 [2]河北省电子信息产品监督检验院,石家庄050071 [3]河北省工业和信息化厅,石家庄050051 [4]北京理工大学管理与经济学院,北京100081

出　　处：《武汉理工大学学报》2012年第2期129-134,共6页Journal of Wuhan University of Technology

基　　金：国家自然科学基金(71102174);河北省教育厅重点基金(ZD200911)

摘　　要：多帧图像超分辨率重建中,连续各帧图像间的精确匹配和帧的选择具有非常重要的意义。提出了自适应帧选择原则,设计了联合光子流运动估计配准和超分辨率重建的方法。首先采用光子流运动估计算法计算各帧间的运动估计,设计一种自适应帧选择方法丢弃一些帧间运动较大的帧,再通过亚像素图像配准,计算得到相对精确的运动估计参数,最后结合最大后延概率方法进行图像超分辨率计算,并充分考虑两次迭代所得图像向量的差值对下次迭代算法的影响。实验结果表明,该方法不仅可以实现亚像素级的精确配准,还可以使重建后图像在视觉效果和峰值信噪比上都得到更好的效果。Registration of consecutive frames and selection of frame is quite essential in multi-frame image super-resolution.An adaptive frame selection principle is proposed.A method which joints the optical flow algorithm for motion estimation registration and the super-resolution is designed.First,using the optical flow algorithm to calculate the inter-frame motion estimation,designing an adaptive frame selection method to discard some of the larger inter-frame motion frames,and then through sub-pixel image registration to calculate the accurate motion estimation parameters,and finally combine the MAP method for image super-resolution which take into account two iterations of the difference between the resulting image vectors of the next iteration algorithm.Experimental results show that this method not only achieve sub-pixel accurate registration,but also achieve better results in the visual effects and the peak signal to noise ratio.

关 键 词：超分辨率重建 自适应帧 运动估计 光流法 最大后延概率 

分 类 号：TN919.8[电子电信—通信与信息系统]