A Low-Cost Rolling Shutter Distortion Correction Algorithm for Mobile Phone Camera

Abstract

In this paper, we propose a method to solve the problem of rolling shutter distortion exist in mobile phone camera. Conventional method uses full search to estimate motion vector and correct distort by image warping method. It results in large memory usage and high computational complexity. To solve this problem, the proposed method uses two frames with viewing screen size and local search motion estimation by using four searching windows. And instead of image warping, the proposed method utilizes the characteristic of rolling shutter of CMOS image sensor, and corrects raw data line by line in distorted photograph. The simulation results show that good quality is achieved in the corrected photographs by our proposed method.