Abstract. Jitter or unintentional motion during image capture, poses a critical problem for any image processing application. Video stabilization is a technique used to correct images against unintentional camera motion. We propose a simple and fast video stabilization algorithm that can be used for real time pre-processing of images, which is especially useful in automotive vision applications. Corner and edge based features have been used for the proposed stabilization method. An affine model is used to estimate the motion parameters using these features. A scheme to validate the features and a variant of iterative least squares algorithm to eliminate the outliers is also proposed. The motion parameters obtained are smoothed using a moving average filter, which eliminates the higher frequency jitters obtained due to unintentional motion. The algorithm can be used to correct translational and rotational distortions arising in the video due to jitter.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.