胶片档案中降噪算法的研究

Research on denoising algorithms for film archives

胶片档案的修复一直是人们关注的重点,传统的人工物理或化学修复方法存在低效和不可逆性的缺点,因此数字化和图像处理技术成为胶片档案修复的热点。本文提出了一种针对彩色胶片档案影像的三维块匹配视频降噪算法。该算法使用连续阶导数的阈值滤波,并充分利用空域和频域特性,有效改善了硬阈值滤波引起的去噪图像上的“振铃”效应。此外,该算法还引入光流估计视频中的运动,在短时间内修复胶片档案画面。通过与其他滤波算法进行实验比较分析,证明了该算法在胶片档案修复中的可行性和实用性,并展示了其在降噪方面的优越性。本文的结果可用于提高胶片档案的质量,促进文化遗产的保护和传承。

The restoration of film archives has always been a focal point of people's attention,in order to avoid the low efficiency and irreversibility of traditional physical or chemical restoration methods,digital and image processing techniques have been adopted for the restoration of film archives.This paper proposes a 3D block matching video denoising algorithm for color film archive images.The algorithm uses threshold filtering of continuous order derivatives and fully exploits the characteristics of space and frequency domains,effectively improving the"ringing"effect on denoised images caused by hard threshold filtering.At the same time,the algorithm also introduces motion estimation in video,to repair the film archive images in a shorter time.Through experimental comparison and analysis of other filtering algorithms for film archive restoration,the feasibility and practicality of this algorithm in film archive restoration are demonstrated,and its superiority in noise reduction is demonstrated.The results of this paper can be used to improve the quality of film archives and promote the preservation and inheritance of cultural heritage.