In this paper, we propose an efficient blind audio watermarking detection method for spread spectrum watermarking system. The proposed method considers the detection problem of the watermark as a matter of Blind Source Separation (BSS) between the audio signal and the watermark one. Thus, we aim to improve detection step by employing Independent Subspace Analysis (ISA) and Empirical Mode Decomposition (EMD) theories. The resulting technique called UISA (Underdetermined Independent Subspace Analysis) permits extracting the embedded watermark with low binary error rate and high bit rate.
展开▼
