Provided by: International Journal of Computer Science and Mobile Computing (IJCSMC)
Topic: Big Data
Date Added: May 2014
A reversible watermarking scheme is one of the improvements in histogram shifting modulation that modifies local precision of image content. In existing schemes large number of images can't be hidden i.e. capacity is less and distortion is more so the authors propose dynamic histogram shifting to insert large number of images into a single host images and with the less distortion. For the same capacity, achieved Peak Signal-to-Noise Ratio (PSNR) is more than 63dB. Invariant image classification is to classify the parts of the images and embedder, extractor remain synchronize with the watermarked image. Classification of original images derives from image itself.