基于跨阶段特征融合算法的混合部位运动图像超分辨率修复方法  

A Super-resolution Restoration Method for Moving Images of Mixed Part

作　　者：朱磊[1] ZHU Lei(Department of Physical Education,Wuhu Vocational and Technical College,Wuhu 241003,China)

机构地区：[1]芜湖职业技术学院体育教学部,安徽芜湖241003

出　　处：《长春大学学报》2024年第6期25-31,共7页Journal of Changchun University

基　　金：安徽省高校哲学社会科学研究项目(2023AH052372)。

摘　　要：混合部位运动图像存在模糊、失真、低分辨率等问题,且峰值信噪比差值较小。提出基于跨阶段特征融合算法的混合部位运动图像超分辨率修复方法。结合非线性映射关系重构像素点,采用跨阶段特征融合算法,完成混合部位运动图像超分辨率修复。实验结果表明:得到完整的混合部位运动图像修复结果,表现出的峰值信噪比差值较大,修复后的图像分辨率更优,满足了混合部位运动图像的现实应用需求。There are some problems such as blur,distortion and low resolution in moving images of mixed parts,and the difference of peak signal-to-noise ratio is small.A super-resolution restoration method for moving images of mixed parts based on cross-stage feature fusion algorithm is proposed.The nonlinear mapping relationship is used to reconstruct the pixels,and the cross-stage feature fusion algorithm is used to complete the super-resolution restoration of the moving images of mixed parts.The experimental results show that the complete restoration results of moving images of mixed parts are obtained,the difference of peak signal-to-noise ratio is large,and the resolution of repaired images is better,which meets the practical application requirements of moving images of mixed parts.

关 键 词：运动图像 跨阶段特征融合算法 混合部位图像修复 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]