Full frame video motion detection and stabilization using mosaicing and deblurring

Abstract

Now-a-days, most of the videos captured are either from mobile phones or handheld video cameras. These videos are mostly shaky with undesired motion. The concept of video stabilization aims at removing this annoying shaky motion from the videos. In this process we estimate the motion of the camera, regenerate the motion of camera trajectory by removing the shaky component and complete the video with improvements like inpainting to fill the empty image areas. The pixel information of the nearby pixels is used to estimate the intensity of the missing pixels and filling the video frames. Though we can obtain a stabilized video, there are some parameters which can be improved such as key point detection, deblurring. So that we can get a superior quality video. Mosaicing is a parameter which is used to correct geometric deformations, video registration using video data and/or camera models and eliminating seams from video mosaics. The quality of the video after removing the shaky effect is further enhanced by using a deblurring algorithm. The smoothness of the video is improved by eliminating motion blur. It transfers and interpolates sharper image pixels from neighbouring frames to increase the sharpness of the frame. Using this process we create superior quality videos and reduce distracting vibrations and it is also used for improvising image quality in surveillance cameras.