| Among the information obtained by human beings,the information obtained by human eyes access accounts for 80% of the total information.With the continuous development and progress of science and technology,images and videos are growing rapidly as information carriers.Due to a large amount of video data,video compression has become a key technology to store and transfer the videos.In addition,haze becomes common in people’s daily life and it is necessary to remove the haze from the video.However,there is almost no video compression method which removes the haze at the same time.Therefore,we did the following works:Firstly,we improve the video dehazing method based on temporal visual coherence,so that the video sequence after dehazing can obtain good details while preserving the visual coherence.Secondly,we propose a hazy video compression method based on motion estimation sharing which combining video dehazing method with video compression method.Through motion estimation,the computing resource sharing between video dehazing and video compression is realized.Therefore,we can reduce the video transmission bandwidth and storage space while effectively removing the haze to weaken the negative effects of haze on video capture and transmission.Last but not least,we make the optimum decision on video dehazing and video compression.By building the correlation function of dehazing result evaluation and ratedistortion performance and making an optimum decision on the motion vector and macroblock partition mode,we obtain good video dehazing result and video compression performance at the same time. |
PDF Full Text Request