Home About Login Current Archives Announcements Editorial Board
Submit Now For Authors Call for Submissions Statistics Contact
Home > Archives > Volume 20, No 17 (2022) > Article

DOI: 10.14704/NQ.2022.20.17.NQ88005

Effective Digital Audio Watermarking Using Dwt And Neural Networks

Abhijit Patil, Dr. Ramesh Shelke

Abstract

Watermarking is the process in which a digital signal is added with another secret digital signal. Digital audio watermarking has been widely used in many applications such as copyright protection, tamper detection, piracy prevention, content authentication, etc. The audio watermarking process has to satisfy many properties such as robustness, imperceptibility, and security. There are many classical as well as hybrid techniques available in the literature to achieve these properties. However, it’s difficult to achieve all the properties at the highest level using a single technique. Echo hiding and pitch shifting approaches are being used from beginning. Some new techniques are being designed that make use of machine learning and deep learning algorithms, “bio-inspired algorithms” such as swarm intelligence algorithms [18], and genetic algorithms, AI-based techniques such as simulated annealing can also be used for optimization in watermarking process. Arnold scrambling and use of cryptographic algorithms can be used for increasing the security of watermarks. However, it is observed that many modern approaches are not giving efficient results in embedding and extraction of watermark with minimum bit error. Finding optimal locations for the watermark bits embedding into host signal is a challenge. In this paper, we are discussing some of the important hybrid and novel techniques used for digital audio watermarking. We proposed and demonstrated the work using a custom-designed simple backpropagation neural network. Our focus is to demonstrate the usefulness of neural network architectures in the subject of study

Keywords

Audio Watermarking, Neural Network, DWT, Robustness, Imperceptibility, Deep Learning.

Full Text

PDF

References

?>