Audio Watermarking Using DWT and SVD for Robust Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing audio watermarking systems are vulnerable to attacks and lack robustness in embedding and detecting watermarks in audio signals, especially when subjected to intentional or unintentional distortions, making it difficult to ascertain the type, strength, and location of manipulation without the original host signal.
Innovation Solution
A computer-implemented system using Discrete Wavelet Transform (DWT) and Singular Value Decomposition (SVD) for embedding and detecting watermarks in audio signals, which divides audio signals into frames, applies multi-level DWT, arranges coefficients, performs SVD, embeds watermarks in eigenvalues, and uses inverse DWT to create a watermarked signal, and subsequently detects watermarks by correlating eigenvalues with pre-stored values to authenticate the embedded watermark.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional watermarking methods are used to embed watermarks in audio signals, then the watermarking process is simple, but the system is vulnerable to attacks and lacks robustness against distortions
Solution Approach 1:
The audio signal is divided into multiple frames, and each frame undergoes multi-level DWT decomposition to generate detailed coefficients at different levels. This segmentation approach allows the watermark to be embedded in a distributed manner across multiple frequency sub-bands, enhancing robustness against attacks while maintaining manageable system complexity through modular processing of each frame independently
Solution Approach 2:
The patent transforms the audio signal from the time domain to the frequency domain using multi-level DWT, creating a multi-dimensional representation with approximation coefficients and detailed coefficients at different decomposition levels. This dimensional transformation provides additional degrees of freedom for robust watermark embedding in the frequency domain while preserving the ability to reconstruct the original signal
2Measurement precision
If watermark is embedded strongly to ensure detectability, then the watermark can be detected easily, but the watermark becomes audible and degrades audio quality
Solution Approach 1:
The patent embeds watermark information selectively in specific frequency sub-bands (detailed coefficients at different DWT levels) rather than uniformly across the entire audio spectrum. By targeting specific local regions in the frequency domain where the human ear is less sensitive, the system achieves reliable watermark detection while minimizing audible degradation of audio quality
Solution Approach 2:
The system adjusts the embedding strength parameter (alpha) dynamically based on the local characteristics of the audio signal and the specific frequency sub-band being processed. This allows optimization of the watermark-to-audio ratio to achieve detectability while staying below the threshold of human perception in each local region
3Measurement precision
If the original host audio signal is required for watermark detection, then precise detection can be achieved, but the system cannot detect watermarks in distributed or modified copies without the original
Solution Approach 1:
The patent uses synchronization patterns as intermediary signals that are embedded along with the watermark information. These synchronization patterns serve as reference markers that enable the detection system to locate and extract watermark information from watermarked signals even when the original host signal is unavailable, by providing a basis for correlation-based detection
Solution Approach 2:
The system embeds synchronization patterns and reference information preliminarily during the watermark embedding process, preparing the watermarked signal with built-in detection capabilities. This preliminary action enables the detection system to autonomously locate and verify watermark presence without requiring the original host signal, enhancing adaptability for distributed copy detection
Data Source
AI summary
A computer implemented system for audio watermarking for providing robust and blind audio watermarking. The system comprises a watermark embedding system wherein an audio signal is divided into audio frames, multi-level District Wavelet Transform (DWT) is applied on each frame, followed by Singular Value Decomposition (SVD) and embedding the watermark, further followed by inverse SVD and inverse DWT to get watermarked audio frames which are combined to generate a watermarked audio signal. The system further comprises watermark extracting detection system wherein the watermarked audio signal which may be attacked and/or modified is divided into watermarked audio frames, multilevel DWT is applied on each watermarked audio frame, followed by SVD, extracting the embedded watermarked, correlating the extracted watermark with pre-stored watermarks, calculating Peak to Sidelobe ratio (PSR) from the correlation coefficient arrays and finally comparing each PSR with a threshold to authenticate the embedded watermark.