Abstract
Abstract Electronic image stabilization technology for shipborne cameras enhances auxiliary navigation by providing critical status information and supporting decision-making in intelligent navigation systems. We have developed an optimized framework based on image feature points to improve the stability and operational efficiency of these cameras. This method facilitates rapid processing while maintaining high image stability through the integration of feature recognition, motion estimation, and image smoothing. Our approach significantly outperforms traditional methods in efficiency, reliability, and flexibility. Additionally, we introduced a weighted smoothing algorithm for handling motion trajectories and analyzed the impact of the ship's movement on the camera. Using a transformation compensation matrix, our stabilization experiments demonstrated a reduction in mean square error by 44–64% and an increase in peak signal-to-noise ratio by 9–17%.