Improving quality of images in print-scan channel

A Printed photograph cannot be used to recover the original digital image with improved accuracy because of distortions introduced during the process of scanning the printed material. This paper describes a method to recover the original digital images from the print-scan channel with improved quality. We solve this approximation problem by formulating a method for approximating the original digital image by combining both the scan of printed photograph as well as digital auxiliary information kept together with the print. The method consists of two processes namely, encoding and decoding. In the process of encoding Auxiliary information is generated from the original digital image and it is kept together with print. Digital auxiliary information consists of small amount of digital data which is composed of sampled color pixel values to enable accurate image reconstruction. In the process of decoding the auxiliary information in the printed photograph is used along with the scan of printed photograph. Thus it provides a closer approximation of the original digital image and reducing all distortions introduced in scanning. Image registration is done by using the auxiliary information and then registered image gets color corrected using the sampled color pixel values. The experimental results confirm that quality of reconstructed image gets improved while using reduced amount of auxiliary information.