A robust DWT-based video watermarking algorithm

Abstract

In this paper, we propose a DWT-based video watermarking algorithm. It embeds multiple information bits into uncompressed video sequences. Its features include: i) embedding data in the LL subband; ii) reducing error probabilities of detection by exploiting the BCH code; iii) combating bursts of errors by using a novel 3-D interleaving technique; iv) developing an effective temporal synchronization technique (compared with the sliding correlation proposed by Hartung and Girod, our synchronization technique is more advance); and v) a valid secure key is needed in watermarking extraction. Experimental results demonstrate that the hidden data with the proposed algorithm is visually transparent, robust against MPEG-2 compression, scaling 2.0, and the detection error rate with 3-D interleaving is much lower than that without 3-D interleaving when frame loss rate is below 50%. The imperceptible high bit rate watermark thus embedded is robust and blind extracted. The experiments have demonstrated the effectiveness of the proposed algorithm.