An Adaptive Anti-HSI Audio Watermarking Algorithm In DCT Domain

The audio watermarking algorithm embeds specific meaning information(related information of the audio creator,audio file download and dissemination records)into the audio file,and does not affect the quality of the original audio file itself;in the case of copyright disputes,it can The watermark information embedded in the audio is extracted correctly,even if the audio embedded in the watermark is attacked by signal processing,the correctness of the watermark extraction will not be affected.Embedding a watermark in the audio can realize audio copyright tracking.Its effectiveness depends not only on the imperceptibility of the algorithm after the watermark is embedded,but also on the robustness of the extraction.There are checks and balances between the two.Whether the two can achieve a better balance is the determining factor of whether the audio watermarking algorithm can be practical.In recent years,watermarking algorithms based on Discrete Cosine Transform(DCT)have received widespread attention.The DCT transform has high computational efficiency.After the audio is transformed,the energy is concentrated and conserved.Aiming at the audio watermark in the DCT domain,this paper studies adaptive embedding technology and Host Signal Interference(HSI)technology during extraction to ensure that the embedded watermark is imperceptible and maximizes the robustness of watermark extraction.The adaptive embedding technology determines the embedding parameters adaptively according to the host audio.Signal-Noise Ratio(SNR)is often used to embed the imperceptible parameter of the watermark.The current audio watermarking algorithm based on SNR adaptation does not give an answer to whether there is room for further increasing the upper limit of the embedding parameters;for the algorithm for processing HSI in the audio watermark extraction in the DCT domain,the current research does not use HSI to be active in watermark extraction.Function,the robustness of the algorithm has room for further improvement.Based on the above research status of adaptive audio watermarking algorithms,this paper draws on the idea of adaptive image watermarking algorithms in the DCT domain,and proposes an adaptive anti-HSI audio watermarking algorithm,which maximizes the robustness of the watermark under the premise of ensuring that the watermark is imperceptible.The main work content and innovations of this paper are as follows:(1)Propose an adaptive function relationship that increases the upper limit of embedding parameters.When establishing the functional relationship between embedding parameters and SNR,the principle of conservation of energy in DCT is taken into consideration,and the upper limit of embedding parameters is increased,so as to better balance the imperceptibility and robustness of the watermark;on this basis,it is suitable for Watermarked audio segments,adaptively determine the embedding parameters of each audio segment according to the determined functional relationship,so that the audio file obtains a consistent SNR,and through simulation experiments on five different types of audio,the effectiveness of this algorithm in terms of imperceptibility is verified.(2)Propose a strategy for introducing HSI control factors to improve the robustness of the algorithm.Through theoretical analysis,it can be known that HSI has both positive and negative effects on the results of watermark extraction.The introduction of HSI control factors can better eliminate the negative effects of HSI on watermark extraction.At the same time,through the appropriate value of HSI control factors,the maximum It retains its positive effect on watermark extraction to a certain degree,and further improves the robustness of the algorithm.Through simulation experiments on five different types of audio,the effectiveness of this algorithm in terms of robustness is verified.The results of simulation experiments on five different types of audio show that from the perspective of imperceptibility,the SNR of the algorithm in this paper is greater than that of the existing algorithm.The average SNR of the algorithm in this paper is compared with the A-SNR algorithm,the OPT-SNR algorithm,The MO-PN algorithm and the DCT-GA algorithm have improved by 4.6,3.7,3.2 and 1.3;in terms of the algorithm's robustness to conventional attacks,the average bit error rate(BER)of the algorithm in this paper is significantly better than that of A-SNR algorithm and OPT-SNR algorithm.For MP3 attacks(64kbps),the BER of this algorithm is 12.84% and 2.12% higher than that of MO-PN algorithm and DCT-GA algorithm,respectively.The above experimental results show that this algorithm has good performance The imperceptibility of the embedded watermark also improves the robustness against attacks,making the two achieve a higher level of balance.This paper uses the above-mentioned innovative algorithm as the key technology to realize a digital audio watermark embedding and robust extraction system,including user login,original audio selection,watermark information generation,watermark embedding,add attack and watermark extraction six modules.