A digital image stabilizer based on wavelet transform

Image stabilization is a necessary step to reduce the effect of camera motion when, as in this case, image sequences are acquired from a mobile platform. In this work, we use block-based motion estimation in the wavelet space and kalman filtering method to stabilize the sequence of the image taken from Outdoor Visual Surveillance System. We first estimate the translational motion by projecting the vertical and horizontal details of wavelet decomposition in level 2. These parameters are used as initial estimation of motion's parameters. Then, we use block-based motion estimation method to refine the precise motion parameters. Global motion correction by frame position smoothing (FPS) is based on obtaining correction vectors from the difference of original and low-pass filtered absolute frame positions. FPS executed in realtime using Kalman filtering. The simulation results have shown that the algorithm is reliable.