DCT BASED DIGITAL VIDEO WATERMARKING USING HUMAN VISUAL SYSTEM PROPERTIES

Abstract

The major thesis work carried out in this report is based on video image-In-image watermarking algorithms which also take in to consideration the Human visual system Properties to find efficient locations in video frames for robust and imperceptible watermarks. Here a new HVS is defined which is a optimized global masking map for hiding watermark signals by combining the frequency masking, spatial masking, and motion masking effects of HVS. After getting adequate locations for insertion of watermark in parent video frames another video sequence is embedded frame by frame as watermark. For embedding the video sequence two DCT based algorithms are used which do embedding in spatial domain by modifying the value of selected coefficients. The transform operation is carried out on 4*4 blocks of each frame for DCT .Later video scene characteristic detection to improve digital watermarking transparency is carried out by subjective test analysis using different video sequences an...