Spatio-temporal Just Noticeable Distortion Model Guided Video Watermarking

Perceptual video water mar kilo: needs to take In advantage of the results from human visual system (FINIS) studies Since motion is a specific feature of video. temporal HVS properties need to be taken into account In this paper. we exploit a combined Spatio-Temporal Just Noticeable Distortion (JND) model which incorporates spatial CSF. temporal modulation factor, retinal velocity, luminance adaptation and contrast masking to :wide watermarking for digital video The proposed water mar king scheme where visual models representive additional accurate perceptual visibility threshold are fully used to determine scene-cloven upper bounds on watermark insertion. allows W. to provide the maximum strength transparent watermark Experimental results confirm the improved performance of our combined Spatio-Temporal JND model guided watermarking scheme Our Spatio-Temporal JND model guided watermarking scheme which allows higher injected-watermark enemy without jeopardizing picture quality performs much better on robustness than other algorithms based on the relevant existing perceptual models