Error reduction in reconstruction of kinoform CGH patterns for a hologram ID tag system

In this paper, we propose a hologram reconstruction algorithm with techniques which can effectively reduce reconstruction errors caused from reconstruction process and external damages of the kinoform CGH ID tag. The 11- level kinoform computer generated hologram (CGH) patterns which are generated by modified simulated annealing are recorded on a texture paper by a photo printer, and retrieved by a digital camera. Through suitable reconstruction program and image processing in computer, the captured hologram patterns on the ID tag are reconstructed with a low error rate of about 4%, compared to the original data prior to the kinoform CGH pattern. It is also shown that our robust reconstruction algorithm can considerably reduce reconstruction errors for damaged tags. The original image has been actually reconstructed at a low error rate of about 8% and 12% from a 25% and 50% damaged tag, respectively.