Audio Watermarking with Patchwork and DCT

Abstract

Digital audio watermarking plays a crucial role in preserving multimedia content from illegitimate copying and reproduction. Audio watermarking has a relatively good perception quality level for the patchwork process. The concerns between protection, robustness, and imperceptibility are contemporary research areas, which are still significant issues. Proposed research explores, a patchwork approach for signal processing attacks, using discrete cosine transforms (DCT)within audio watermarking slant. At first, the watermarking audio clip is fragmented up into alike portion and their sub-portions, followed by its coefficient calculation. After removing the coefficient associated with high frequency, the residual coefficients are utilized for the formation of equal length of frame pairs. Using specific embedding criteria and secure information, key watermarks are inserted into audio. The adjustments are made pertaining, the selection conditions chosen through the embedding procedure, so that, the recognition of watermarked pair of frames, is made at the extraction step. To remove the watermark from watermarked audio, the private data key is used. The explored audio watermarking technique is applied and evaluated for security, robustness, security, imperceptibility, and data payload under signal processing attacks.