Audio watermark embedding and extraction method and system

Abstract

Disclosed in the present invention are an audio watermark embedding and extraction method and system. The method comprises: acquiring audio data, calculating a masking threshold for each frequency point of the audio data, and preprocessing the audio data; performing encoding on the basis of watermark information to generate an audio watermark, and embedding the audio watermark into the audio data on the basis of the masking threshold of the audio data; and locating the position of the audio watermark in the audio data into which the audio watermark is embedded, and extracting the audio watermark on the basis of the masking threshold of the original audio data. In the present invention, an audio watermark is embedded on the basis of human auditory masking theory; the masking threshold of audio data is calculated, and the audio watermark is added to audio in a frequency domain on the basis of the masking threshold, thereby allowing the audio watermark to better resist these attacks, ensuring the robustness of the audio watermark, and ensuring that the detectability can be maintained even under harsh conditions. The present invention solves the problem whereby existing audio watermarks have insufficient robustness and are prone to affect the quality of audio itself.

Description

A method and system for embedding and extracting audio watermarks Technical Field

[0001] This invention relates to the field of audio processing technology, and in particular to an audio watermark embedding and extraction method and system. Background Technology

[0002] With the increasing prevalence of digitization and networking of multimedia content, copyright protection has become particularly important. Audio watermarking, as an effective means of copyright protection, can help confirm ownership of works and track illegal copying and distribution. However, traditional audio watermarking technologies face many challenges, such as insufficient robustness, impact on audio quality, poor detectability of watermarks, and low processing efficiency.

[0003] Therefore, there is a need for a robust watermark embedding method that does not affect audio quality. Summary of the Invention

[0004] To address the aforementioned issues, this invention provides an audio watermark embedding and extraction method. By calculating the masking threshold of the audio data, an audio watermark is added to the audio in the frequency domain based on the masking threshold. This solves the problem that existing audio watermarks lack robustness and easily affect the quality of the audio itself.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] An audio watermark embedding and extraction method includes the following steps:

[0007] S1. Acquire audio data, calculate the masking threshold for each frequency point of the audio data, and preprocess the audio data;

[0008] S2. Generate an audio watermark by encoding the watermark information, and embed the audio watermark into the audio data according to the masking threshold of the audio data;

[0009] S3. Retrieve the location of the audio watermark in the audio data with embedded audio watermark, and extract the audio watermark based on the masking threshold of the original audio data.

[0010] Furthermore, in step S1, the preprocessing of the audio data specifically includes: performing a fast Fourier transform on the audio data to obtain the audio spectrum.

[0011] Furthermore, in step S2, the specific implementation of encoding the audio watermark based on the watermark information includes converting the watermark information into a binary sequence.

[0012] Furthermore, in step S2, the process of encoding and generating an audio watermark based on the watermark information further includes adding redundant information to the audio watermark.

[0013] Furthermore, the redundant information is a forward error correction code, the encoding algorithm of the forward error correction code is the Reed-Solomon encoding algorithm, and the generation formula of the forward error correction code is:

[0014] ,

[0015] Where g(x) is the forward error correction code, m(x) is the watermark information polynomial, n is the watermark length, k is the information length, and r(x) is the redundancy code polynomial.

[0016] Further, in step S2, the specific implementation of embedding the audio watermark into the audio data according to the masking threshold of the audio data is as follows: search for the part of the audio data whose masking threshold is higher than a preset value, the part higher than the preset value is the masked part, divide the audio watermark into several segments according to the distribution of the masked part, embed the several segments of audio watermark into the masked part in sequence to obtain the audio spectrum of the embedded audio watermark, and perform an inverse fast Fourier transform on the audio data of the embedded audio watermark to obtain the audio data of the embedded audio watermark.

[0017] Furthermore, in step S3, the specific implementation method is as follows: after performing a fast Fourier transform on the audio data with embedded audio watermark, the masked part is resampled, the masked part after resampling is filtered according to the energy level to select the embedded audio watermark, and the embedded audio watermark is decoded to obtain the audio watermark.

[0018] Furthermore, the relationship between the intensity of the audio watermark and the masking threshold of the location of the audio watermark is detected, and the audio watermark is determined to have been modified based on the relationship.

[0019] Furthermore, step S2 also includes embedding the same audio watermark at multiple independent frequency domain locations.

[0020] Through the above technical solution, the present invention has the following beneficial effects: it solves the problem that existing watermarking technologies are prone to loss when faced with various signal processing operations. By embedding audio watermarks based on the masking theory of the human ear, the audio watermarks can better resist these attacks, ensuring their robustness and maintaining detectability even under harsh conditions. Attached Figure Description

[0021] Figure 1 is a schematic diagram of the overall process of an audio watermark embedding and extraction method according to the present invention.

[0022] Figure 2 is a schematic diagram of an audio watermark embedding and extraction system according to an embodiment of the present invention. Detailed Implementation

[0023] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0024] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0025] Example 1

[0026] Referring to Figure 1, an audio watermark embedding and extraction method includes the following steps:

[0027] S1. Acquire audio data, calculate the masking threshold for each frequency point of the audio data, and preprocess the audio data;

[0028] S2. Generate an audio watermark by encoding the watermark information, and embed the audio watermark into the audio data according to the masking threshold of the audio data;

[0029] S3. Retrieve the location of the audio watermark in the audio data with embedded audio watermark, and extract the audio watermark based on the masking threshold of the original audio data.

[0030] Masking is a common psychoacoustic phenomenon determined by the human ear's frequency discrimination mechanism. It refers to the phenomenon where a relatively weak sound is not perceived by the ear when near a stronger sound; that is, it is masked by the louder sound. The stronger sound is called the masker, and the weaker sound is called the masked sound. Within the auditory region, the human ear's response to sound varies with frequency, with the most sensitive frequency range being 2-4 kHz. Outside this range, the auditory sensitivity gradually decreases. The lowest sound pressure level that the human ear can just hear is called the hearing threshold, which is a function of sound frequency. The human ear cannot hear sounds with sound pressure levels below the hearing threshold. For example, if a pure tone signal is used as the target and its sound pressure level is below the hearing threshold (i.e., the quiet threshold), the human ear cannot hear it. Therefore, we can utilize this characteristic to embed audio watermark signals at locations with higher thresholds, thus achieving the goal of hiding information within the audio without affecting the user's normal listening experience.

[0031] In an optional embodiment, the preprocessing of the audio data in step S1 specifically includes: performing a fast Fourier transform on the audio data to obtain the audio spectrum.

[0032] The time-domain signal is converted into a frequency-domain signal by Fourier transform, which makes it easier to embed the watermark signal in the frequency domain.

[0033] In an optional embodiment, in step S2, the specific implementation of encoding the audio watermark based on the watermark information includes converting the watermark information into a binary sequence.

[0034] In an optional embodiment, step S2, which involves encoding the audio watermark based on the watermark information, further includes adding redundant information to the audio watermark.

[0035] Redundant information is used to add verification information to audio watermarks to improve their robustness.

[0036] The watermark information is a binary sequence. To enhance robustness, we need to add error correction codes to it. Reed-Solomon codes are a commonly used forward error correction code that can be used to increase the watermark's resistance to interference.

[0037] This invention applies the Reed-Solomon coding algorithm and uses the RS(255,239) coding scheme, where 255 is the codeword length and 239 is the information length. This means that every 239 bits of information will generate a 255-bit coding block, including 16 bits of redundant code for error correction.

[0038] In an optional embodiment, the redundant information is a forward error correction code, the encoding algorithm of the forward error correction code is the Reed-Solomon encoding algorithm, and the generation formula of the forward error correction code is:

[0039] ,

[0040] Where g(x) is the forward error correction code, m(x) is the watermark information polynomial, n is the watermark length, k is the information length, and r(x) is the redundancy code polynomial.

[0041] In an optional embodiment, in step S2, the specific implementation of embedding the audio watermark into the audio data according to the masking threshold of the audio data is as follows: searching for the portion of the audio data whose masking threshold is higher than a preset value, the portion higher than the preset value is the masked portion, dividing the audio watermark into several segments according to the distribution of the masked portion, embedding the several segments of audio watermark into the masked portion in sequence to obtain the audio spectrum of the embedded audio watermark, and performing an inverse fast Fourier transform on the audio data of the embedded audio watermark to obtain the audio data of the embedded audio watermark.

[0042] Watermarks are embedded in regions of the frequency domain protected by masking effects, avoiding important audio feature areas and minimizing the impact on audio quality. Specifically, the steps include:

[0043] Using a masking threshold M(f) and a predefined factor α, the watermark embedding strength W(f) = α * M(f) is calculated, where α represents the expansion factor of the embedding region.

[0044] The encoded watermark information is added to the selected spectral component by a ratio β, where β is the embedding strength (0-1), i.e., S'(f) = S(f) + β* W(f), where S(f) is the original spectrum and S'(f) is the spectrum after embedding the watermark.

[0045] In an optional embodiment, step S3 is specifically implemented as follows: after performing a fast Fourier transform on the audio data with the embedded audio watermark, the masked part is resampled, the resampled masked part is filtered according to the energy level to select the embedded audio watermark, and the embedded audio watermark is decoded to obtain the audio watermark.

[0046] In an optional embodiment, the relationship between the audio watermark intensity and the masking threshold of the audio watermark location is detected, and the audio watermark is determined to have been modified based on the relationship.

[0047] When embedding an audio watermark, the embedding strength has a certain coefficient relationship with the masking threshold of the audio watermark location. We can determine whether the audio watermark has been tampered with during transmission by observing the weakening of the audio watermark strength.

[0048] In an optional embodiment, step S2 further includes: embedding the same audio watermark at multiple independent frequency domain locations.

[0049] By embedding multiple identical watermarks, the audio watermarks are preserved in different locations even if the audio is partially damaged due to compression or other reasons during transmission, ensuring the integrity of copyright protection.

[0050] Example 2

[0051] Referring to Figure 2, an audio watermark embedding and extraction system includes:

[0052] The audio processing module is used to acquire audio data, calculate the masking threshold for each frequency point of the audio data, and preprocess the audio data.

[0053] The watermark embedding module is used to encode and generate an audio watermark based on the watermark information, and to embed the audio watermark into the audio data according to the masking threshold of the audio data.

[0054] The watermark extraction module is used to retrieve the location of the audio watermark in the audio data with embedded audio watermarks, and extract the audio watermark based on the masking threshold of the original audio data.

[0055] The embodiments disclosed in this specification are merely illustrative of one aspect of the invention, and the scope of protection of the invention is not limited to these embodiments. Any other functionally equivalent embodiments fall within the scope of protection of the invention. Those skilled in the art can make various other corresponding changes and modifications based on the technical solutions and concepts described above, and all such changes and modifications should fall within the scope of protection of the claims of this invention.

Claims

1. A method for embedding and extracting audio watermarks, characterized in that, Includes the following steps: S1. Acquire audio data, calculate the masking threshold for each frequency point of the audio data, and preprocess the audio data; S2. Generate an audio watermark by encoding the watermark information, and embed the audio watermark into the audio data according to the masking threshold of the audio data; S3. Retrieve the location of the audio watermark in the audio data with embedded audio watermark, and extract the audio watermark based on the masking threshold of the original audio data.

2. The audio watermark embedding and extracting method of claim 1, wherein, In step S1, the preprocessing of the audio data specifically includes: performing a fast Fourier transform on the audio data to obtain the audio spectrum.

3. The method for embedding and extracting audio watermarks according to claim 2, characterized in that, In step S2, the specific implementation of encoding the audio watermark based on the watermark information includes converting the watermark information into a binary sequence.

4. The audio watermark embedding and extracting method of claim 3, wherein, In step S2, the process of encoding and generating an audio watermark based on the watermark information further includes adding redundant information to the audio watermark.

5. The audio watermark embedding and extracting method of claim 4, wherein, The redundant information is a forward error correction code, the encoding algorithm of which is the Reed-Solomon encoding algorithm, and the generation formula of the forward error correction code is: ， Where g(x) is the forward error correction code, m(x) is the watermark information polynomial, n is the watermark length, k is the information length, and r(x) is the redundancy code polynomial.

6. The audio watermark embedding and extracting method of claim 2, wherein, In step S2, the specific implementation of embedding the audio watermark into the audio data according to the masking threshold of the audio data is as follows: search for the part of the audio data whose masking threshold is higher than a preset value, the part higher than the preset value is the masked part, divide the audio watermark into several segments according to the distribution of the masked part, embed the several segments of audio watermark into the masked part in sequence to obtain the audio spectrum of the embedded audio watermark, and perform inverse fast Fourier transform on the audio data of the embedded audio watermark to obtain the audio data of the embedded audio watermark.

7. The audio watermark embedding and extracting method of claim 6, wherein, In step S3, the specific implementation method is as follows: after performing a fast Fourier transform on the audio data with embedded audio watermark, the masked part is resampled, the masked part after resampling is filtered according to the energy level, the embedded audio watermark is selected, and the embedded audio watermark is decoded to obtain the audio watermark.

8. The method for embedding and extracting audio watermarks according to claim 7, characterized in that, The relationship between the intensity of the audio watermark and the masking threshold of the location of the audio watermark is detected, and the audio watermark is determined to have been modified based on the relationship.

9. The method for embedding and extracting audio watermarks according to claim 1, characterized in that, Step S2 also includes embedding the same audio watermark at multiple independent frequency domain locations.

10. An audio watermark embedding and extraction system, characterized by, include: The audio processing module is used to acquire audio data, calculate the masking threshold for each frequency point of the audio data, and preprocess the audio data. The watermark embedding module is used to encode and generate an audio watermark based on the watermark information, and to embed the audio watermark into the audio data according to the masking threshold of the audio data. The watermark extraction module is used to retrieve the location of the audio watermark in the audio data with embedded audio watermarks, and extract the audio watermark based on the masking threshold of the original audio data.

Images