结合像素流和最优化方法的视频着色系统

Video colorization system based on pixel stream and optimization method

早期电影受技术限制,多以黑白影片的方式呈现,因此,复现真实色彩场景,满足人们对影像色彩的需求日渐成为图像处理领域中的重要研究方向之一。现有的上色工具多数是为一张普通黑白图像着色,而针对视频连续帧的着色是缺乏系统应用的。利用基于像素流的视频着色方法,一定程度上解决了着色中出现的大幅度运动及遮挡问题;针对着色过程中出现瑕疵的视频帧,利用pyqt与最优化视频着色方法,自动对着色图像进行局部修正,省去了大量的手动操作。系统设计上采用MVC架构,整体分为用户登录模块,UI界面设计模块,算法模块和功能处理模块,有效的实现了界面与业务逻辑的分离。实验结果表明,本文设计的视频着色系统性能稳定,便捷高效,且具有更好的交互效果。

Early films were limited by technology and were mostly presented in black and white. Therefore, recreating the real color scenes and satisfying people′s demand for image color has become one of the important research directions in the field of image processing. Most of the existing coloring tools are for coloring a normal black and white image, while the coloring for continuous frames of video is lack of system application. The pixel coloring based video coloring method is used to solve the large-scale motion and occlusion problems in coloring to some extent. For the video frames that appear in the coloring process, the pyqt and the optimized video coloring method are used to automatically color the image. Partial corrections eliminate a lot of manual operations. The system design adopts MVC architecture, which is divided into user login module, UI interface design module, algorithm module and function processing module, which effectively realizes the separation of interface and business logic. The experimental results show that the video coloring system designed in this paper has stable performance, convenient and efficient, and has better interaction effect.