Title :
Removing rolling shutter distortion using optical flow and RANSAC
Author :
Woo Jin Jeong ; Young Shik Moon
Author_Institution :
Dept. of Comput. Sci. & Eng., Hanyang Univ., Seoul, South Korea
Abstract :
CMOS camera modules are used for smartphone. But CMOS camera modules have a problem of rolling shutter effect. In this paper, we propose an efficient method of removing the rolling shutter effect. We first compute the global motion vector by using the optical flow and the camera motion estimation algorithm. Then we estimate the velocity for each row of sensor array. Finally the distorted image is realigned. Experimental results show that the proposed method is more robust and faster than the previous method.
Keywords :
CMOS image sensors; image sequences; motion estimation; optical distortion; optical elements; sensor arrays; smart phones; CMOS camera modules; RANSAC; camera motion estimation algorithm; global motion vector; optical flow; rolling shutter distortion; sensor array; smartphone; Arrays; CMOS integrated circuits; Cameras; Computer vision; Optical distortion; Optical imaging; Videos;
Conference_Titel :
Consumer Electronics (ICCE), 2015 IEEE International Conference on
Conference_Location :
Las Vegas, NV
Print_ISBN :
978-1-4799-7542-6
DOI :
10.1109/ICCE.2015.7066435