基于分区曝光融合的不均匀亮度视频增强  

Non-uniform Illumination Video Enhancement Based on Zone System and Fusion

作　　者：张聿 刘世光[1] Zhang Yu;Liu Shiguang(School of Computer Science and Technology, Tianjin University, Tianjin 300350)

机构地区：[1]天津大学计算机科学与技术学院,天津300350

出　　处：《计算机辅助设计与图形学学报》2017年第12期2317-2322,共6页Journal of Computer-Aided Design & Computer Graphics

基　　金：国家自然科学基金(61170118;61672375)

摘　　要：视频增强技术可以使低曝光视频中的细节显示出来,有效地改进视频质量,为视频的后期处理提供保证.针对直接将不均匀亮度图像增强方法运用到视频处理中容易出现时空不一致现象的问题,提出一种基于分区曝光融合的不均匀亮度视频增强方法.为了实现视频帧的均匀曝光,首先对输入视频帧采用分区曝光法计算不同的曝光区域,对不同的曝光区域进行多尺度曝光校正;然后对每一个区域融合其不同曝光程度的区域;最终形成完整的曝光良好的视频帧.对于视频帧之间时空一致性的保持,结合前一帧曝光分区的方法能更加准确地计算相邻帧之间的对应关系,将前一帧的相应的像素传播到后面的帧中.实验结果表明,相比于目前的视频增强方法,该方法可以使不均匀亮度视频均匀曝光并且更好地保持视频场景的自然性,同时保持视频的时空一致性.Video enhancement is used to reveal hidden details in underexposed videos and improve video quality,which can contribute to video post processing.It will introduce temporal inconsistency by directly applying image enhancement on video.In this paper,we present a non-uniform illumination video enhancement method based on zone system and fusion.We first apply the zone system for exposure evaluation.We then remap each region using a series of tone mapping curves to generate multi-exposure regions.Next,we fuse the regions into a well-exposed video frame.Finally,in order to keep temporal consistency,we temporally propagate the zone regions from the former frame to the current frame using our zone system-based distance metric.Experimental results have shown that the enhanced video exhibits uniform exposure,keeps better naturalness,and preserves temporal consistency.

关 键 词：不均匀亮度视频 视频增强 分区曝光法融合 自然性保持 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]