Audio Watermarking is a method to insert a copyright marker on audio. This method inserts a watermark in the information form and in a way that does not damage the audio. This technique is one of the ways to solve the problem of copyright infringement. The embedded watermark has to meet the condition of not damaging the audio and must have robustness, imperceptibility, and good capacity. The data hiding technique use the combined method of Lifting Wavelet Transform (LWT), Fast Fourier Transform (FFT), QR Decomposition and Reconstruction, and Cartesian-Polar Transformation (CPT) based on Quantization Index Modulation (QIM) with the secured and compressed watermark using Compressive Sampling (CS) technique. The proposed scheme is blind Audio Watermarking as it no needs for original audio in the detection process. The combination of methods overcomes multiple attacks with guaranteed quality watermarking and high capacity. Compared to the existing technique, the data hiding technique can withstand LPF attacks, Resampling, Linear speed change (LSC), and MP3 compression. This proposed technique is also secured due to the coded watermark by a particular random key using CS. Combining CS and Audio Watermarking techniques can perform well in capacity, imperceptibility, security, and attack resistance.
Basis Pursuit with Sparsity Averaging for Compressive Sampling of Iris Images
