A New Approach For Dark Video Denoising And Enhancement Based On NLM And Tone Mapping

There are some problems in low light video, for example, high noise and low contrast, because of low light in environment of shooting. This kind of video gives some obstacles to people who are watching and distinguishing it. This paper presents a novel three-stage algorithm for low light video denoising and enhancement, by exploiting the Non Local Means (NLM) method and a new proposed tone mapping operator. The first stage does the spatial and temporal NLM based denoising respectively, and follows a new adaptively weighted average of the space and time filtered results. The second stage applies our proposed tone mapping technique to transform very dark video into more watchable. The third stage performs further noise suppression by using the NLM video filter in the Y channel, and by using the neighborhood image filter in the Cr channel and in the Cb channel.The overall approach transforms unperceivable videos into more visible ones, effectively reduces very high noise, and all in all, produces high quality restored image sequences outperforming the recent state-of-the-art results. Additionally, the first and third stages can be combined as a two-step filtering scheme for normal-light videos: the novel denoising solution achieves a heavy noise removal, while reducing motion blur artifacts and preserving image details.