Motion-blur-free camera system splitting exposure time

Abstract

In this paper, a motion-blur-free camera system using multi-exposed images is proposed. Advances in image sensors enable a high-speed image readout, which makes it possible to capture multiple images within a normal exposure time. The proposed camera system captures three frames successively at long-short-long exposure times from the same scene. In order to obtain a high-quality image, we utilize the motion-based image merging algorithm that combines the shape information of objects in the under-exposed image captured at the short exposure time and the color information from two blurred images taken at the long exposure time. Simulation results show that the proposed multi-exposure camera system can provide high-quality images mitigating the motion blur.