High Capacity Video Watermarking based on DWT-DCT-SVD
C. N. Sujatha, P. Satyanarayana · 2015
Abstract - In this paper, a new non-blind high capacity video watermarking algorithm based on DWT, DCT and SVD has been proposed. In this method, the watermarking data can be embedded in singular values of DCT coefficients of middle and high frequency subbands (LH, HL, HH) in DWT domain of selected group of frames. The embedded watermarks are extracted with inverse process of embedding. The amount of watermark insertion is controlled by the Peak Signal to Noise Ratio and Mean square error of the watermarked frame with original frame. Experimental results show that the watermarked video is indistinguishable from original video and also state that the proposed algorithm is imperceptible with a good watermark capacity at various scaling factors. From the results, the effect of scaling factor on video quality is clearly observed. The similarity between inserted and extracted watermark is estimated by correlation factor. The further extension of selected method was presented in conclusion.