Feature based video stabilization based on boosted HAAR Cascade and representative point matching algorithm

The success of handheld video capturing devices has further fueled the need of improved video stabilization. The videos often contain many foreground facial features like eyes, nose etc. These foreground features can be considered as feature points and may be used to stabilize videos. This paper proposes an innovative and effective digital video stabilization technique, which utilizes foreground features present in the video to produce consistent and stabilized output. It uses successive stages of Boosted HAAR cascade and representative point matching digital motion stabilization algorithm to identify and stabilize the video. The feature based tracking of object improves motion estimation accuracy between two frames thereby increasing the correlation calculation and compensation motion vector. This work achieves a significantly smoother sequence after the motion compensation. It also improves the robustness, precision and quality of the video when compared to traditional digital stabilization algorithms. The simulation results compared with pre-existing techniques reflect distinct improvements in Inter-Frame Transformation Fidelity values and Structural Similarity Index along with lesser standard deviation between image frames. (C) 2020 Elsevier B.V. All rights reserved.