Abstract
We present a technique for effective lens aberration correction using a low cost mask based light field camera that uses locally optimal masks. Present physical methods for lens aberration corrections require complicated hardware installations on the cameras. Lens aberration correction in software without active optics is an ill-posed problem since the open lens blur causes frequency information loss. We present an extremely low cost modification in the camera by inserting a plastic transparency based mask which converts the ill-posed de blurring problem into a well-posed one allowing efficient de blurring of the lens aberration blur. We use a theoretical field curvature aberration model for designing the optimal transparency masks which are locally optimal. We perform the MSE analysis for our lens aberration correction technique and show that its performance is up to 35 db better than the standard software based correction techniques. We present the hardware built, theoretical model, and synthetic results for aberration correction.
© 2010 Optical Society of America
PDF ArticleMore Like This
Ren Ng and Patrick M. Hanrahan
WB2 International Optical Design Conference (IODC) 2006
Kathrin Berkner and Sapna Shroff
CMD4 Computational Optical Sensing and Imaging (COSI) 2011
Akihiko Utsugi and Kenichi Ishiga
ITuB5 International Optical Design Conference (IODC) 2010