Reversible matrix color image blind watermarking method based on energy aggregation characteristic

Abstract

Along with rapid propagation of digital media, copyright protection becomes an increasingly prominent problem. In order to effectively protect the copyright of a digital image, the invention discloses a reversible matrix color image blind watermarking method based on an energy aggregation characteristic, and the method utilizes the characteristic of a reversible matrix in a frequency domain to realize the high-energy aggregation of the matrix through the operation of first left multiplication and then right multiplication. The embedding of the digital watermark is completed by quantizing the high-energy aggregation coefficient, the robustness and invisibility of the watermark method are improved, the operation process is simplified, and the calculation complexity is reduced. The method has good robustness and invisibility, effectively solves the problem that the color image digital watermark operation speed is low, and is suitable for occasions where digital media copyright protection is carried out safely and efficiently.

Description

Technical Field

[0001] The invention belongs to the technical field of information security and relates to rapid copyright protection of large-capacity color digital images. Background Art

[0002] With the advent of the digital age, the creation, distribution and consumption of digital media content have undergone revolutionary changes. In particular, the popularization of Internet technology has enabled multimedia content such as digital images and videos to be rapidly disseminated around the world. Faced with the easy duplication and modification of digital content, traditional copyright protection methods seem to be powerless. Therefore, the development of effective digital watermarking technology has become an important means to protect digital media copyright. With the development of digital watermarking technology, the requirements for it are becoming higher and higher. Not only does it need to ensure the invisibility and robustness of the watermark, but it is also necessary to consider the computational efficiency of the algorithm and the feasibility of practical application. The reversible matrix color image blind watermarking scheme based on high energy aggregation characteristics was invented to meet these needs. Summary of the invention

[0003] The purpose of the present invention is to provide a reversible matrix color image blind watermarking algorithm based on energy aggregation characteristics, which is characterized by being implemented through a specific watermark embedding process and a watermark extraction process. The watermark embedding process is described as follows: Step 1: Take a picture of size 24-bit color watermark image Divided into three single color channels: red, green and blue , and each single color channel Key-based The Lorenz chaotic map is used to encrypt and convert each encrypted decimal pixel value into an 8-bit binary sequence; then, The binary sequences in are concatenated into a length of A string sequence ,in, Respectively represent the red, green and blue channels; Step 2: Make a picture of size 24-bit color host image Divide into 3 layered host images in the order of red, green and blue colors ; Then, each layered host image Divide into size non-overlapping image blocks; according to the layered watermark bit sequence Total length , using key-based The MD5 hashing pseudo-random algorithm selects non-repeating image blocks, where Respectively represent the red, green and blue channels; Step 3: Select one image block at a time , and use the reversible matrix to achieve high energy concentration according to formula (1); (1) in, represents the original image block matrix, Represents an image block The frequency domain matrix with high energy concentration is realized through the reversible matrix. is the side length of the image block matrix, represents a reversible matrix; Step 4: Layer the watermark bit sequence In the sequence, take out a watermark information to be embedded. According to the embedded watermark information and formula (2), the frequency domain coefficients in the upper left corner of the matrix are quantized to obtain the frequency domain coefficients containing the watermark ,in, Respectively represent the red, green and blue channels; (2) in, , Represents the judgment threshold in the quantization process, is the XOR function, is the remainder function, is the rounding function, Indicates the quantization step size; Step 5: According to the inverse transformation formula (3), the image block containing the watermark is obtained , and the watermarked image block matrix Update to its host image in layer The corresponding position in , where Respectively represent the red, green and blue channels; (3) in, is the matrix after embedding watermark, yes The inverse matrix of Step 6: Repeat steps 3 to 5 of this process until all watermark information is embedded, thus obtaining a layered host image with watermark. , and layer the red, green, and blue host images containing the watermark Reassemble and get the size Watermarked image ,in, Respectively represent the red, green and blue channels; The watermark extraction process is described as follows: Step 1: Set the size to Watermarked image Divided into 3 layers with watermark images , and each layered watermarked image Further divided into non-overlapping image patches, where Respectively represent the red, green and blue channels; Step 2: Select a watermarked image block in turn , and use the reversible matrix to achieve high energy concentration according to formula (4); (4) in, represents the watermarked image block matrix, Represents a watermarked image block The watermarked frequency domain matrix with high energy aggregation is realized through the reversible matrix. is the side length of the image block matrix, represents a reversible matrix; Step 3: Use formula (5) to extract the frequency domain matrix The watermark contained in ; (5) in, is a matrix The high energy coefficient in is the remainder function, is the rounding function towards zero, Indicates the quantization step size; Step 4: Repeat steps 2 and 3 of this process to extract the binary watermark bit sequence of each layer. , and then convert each 8-bit binary information into a decimal pixel value, where Respectively represent the red, green and blue channels; Step 5: Perform key-based analysis on the converted decimal pixels of each layer Decrypt the Lorenz chaotic map and obtain the layered watermark ,in, Respectively represent the red, green and blue channels; Step 6: Combine layered watermarks Form the final extracted watermark ,in, Represents the red, green and blue channels respectively.

[0004] This method uses the maximum coefficient of the image block matrix obtained by reversible matrix calculation to realize a method of color digital watermark embedding and blind extraction; this method has good watermark invisibility and robustness, and is simple and fast. BRIEF DESCRIPTION OF THE DRAWINGS

[0005] Figure 1 (a) Figure 1 (b) are two original color host images.

[0006] Figure 2 (a) Figure 2 (b) are two original color watermarked images.

[0007] Figure 3 (a) Figure 3 (b) is to Figure 2 (a) Figure 2 (b) The watermarks shown are embedded into the host image Figure 1 (a) Figure 1 The structural similarity SSIM values ​​of the watermarked images obtained after (b) are 0.9845 and 0.9755, and the peak signal-to-noise ratio PSNR values ​​are 40.1674dB and 40.1647dB, respectively.

[0008] Figure 4 (a) Figure 4 (b) are respectively from Figure 3 (a) Figure 3 The normalized cross-correlation coefficient NC values ​​of the watermarks extracted in (b) are 1.0000 and 1.0000 respectively.

[0009] Figure 5 (a) Figure 5 (b) Figure 5 (c) Figure 5 (d) Figure 5 (e) Figure 5 (f) is to Figure 3 The watermarks extracted from the watermarked image shown in (a) after being attacked with JPEG compression (70), JPEG2000 (4:1), salt and pepper noise (0.2%), median filtering (3×3), scaling (400%), and shearing (12.5%), have normalized cross-correlation coefficients NC values ​​of 0.9997, 0.9992, 0.9944, 0.9687, 1.0000, and 0.9773, respectively.

[0010] Figure 6 (a) Figure 6 (b) Figure 6 (c) Figure 6 (d) Figure 6 (e) Figure 6 (f) is to Figure 3The watermarks extracted from the watermarked image shown in (b) after being attacked with JPEG compression (70), JPEG2000 (4:1), salt and pepper noise (0.2%), median filtering (3×3), scaling (400%), and shearing (12.5%), have normalized cross-correlation coefficients NC values ​​of 1.0000, 0.9985, 0.9928, 0.9862, 0.9979, and 0.9740, respectively. DETAILED DESCRIPTION

[0011] The purpose of the present invention is to provide a reversible matrix color image blind watermarking algorithm based on energy aggregation characteristics, which is characterized by being implemented through a specific watermark embedding process and a watermark extraction process. The watermark embedding process is described as follows: Step 1: Take a picture of size 24-bit color watermark image Divided into three single color channels: red, green and blue , and each single color channel Key-based The Lorenz chaotic map is used to encrypt and convert each encrypted decimal pixel value into an 8-bit binary sequence (for example, the decimal pixel value 185 is converted into a binary value of 10111001); then, The binary sequences in are concatenated into a length of A string sequence ,in, Respectively represent the red, green and blue channels; Step 2: Make a picture of size 24-bit color host image Divide into 3 layered host images in the order of red, green and blue colors ; Then, each layered host image Divide into size non-overlapping image blocks; according to the layered watermark bit sequence Total length , using key-based The MD5 hashing pseudo-random algorithm selects non-repeating image blocks, where Respectively represent the red, green and blue channels; Step 3: Select one image block at a time , and use the reversible matrix to achieve high energy concentration according to formula (1); (1) in, represents the original image block matrix, Represents an image block The frequency domain matrix with high energy concentration is realized through the reversible matrix. is the side length of the image block matrix, represents a reversible matrix; Here, let the selected image block , , , then we get ; Step 4: Layer the watermark bit sequence In the sequence, take out a watermark information to be embedded. According to the embedded watermark information and formula (2), the frequency domain coefficients in the upper left corner of the matrix are quantized to obtain the frequency domain coefficients containing the watermark ,in, Respectively represent the red, green and blue channels; (2) in, , Represents the judgment threshold in the quantization process, is the XOR function, is the remainder function, is the rounding function, represents the quantization step size; here, let , watermark information w='1', , can be obtained , and then we can get ; Step 5: According to the inverse transformation formula (3), the image block containing the watermark is obtained , and the watermarked image block matrix Update to its host image in layer The corresponding position in , where Respectively represent the red, green and blue channels; (3) in, is the matrix after embedding watermark, yes Here, let the matrix after embedding watermark be , , , can be obtained ; Step 6: Repeat steps 3 to 5 of this process until all watermark information is embedded, thus obtaining a layered host image with watermark. , and layer the red, green, and blue host images containing the watermark Reassemble and get the size Watermarked image ,in, Respectively represent the red, green and blue channels; The watermark extraction process is described as follows: Step 1: Set the size to Watermarked image Divided into 3 layers with watermark images , and each layered watermarked image Further divided into non-overlapping image patches, where Respectively represent the red, green and blue channels; Step 2: Select a watermarked image block in turn , and use the reversible matrix to achieve high energy concentration according to formula (4); (4) in, represents the watermarked image block matrix, Represents a watermarked image block The watermarked frequency domain matrix with high energy aggregation is realized through the reversible matrix. is the side length of the image block matrix, Represents a reversible matrix; Here, let the selected watermarked image block , , , then we get ; Step 3: Use formula (5) to extract the frequency domain matrix The watermark contained in ; (5) in, is a matrix The high energy coefficient in is the remainder function, is the rounding function towards zero, represents the quantization step size; here, let , , we can get the watermark information w * ='1'; Step 4: Repeat steps 2 and 3 of this process to extract the binary watermark bit sequence of each layer. , and then convert each 8-bit binary information into a decimal pixel value, where Respectively represent the red, green and blue channels; Step 5: Perform key-based analysis on the converted decimal pixels of each layer Decrypt the Lorenz chaotic map and obtain the layered watermark ,in, Respectively represent the red, green and blue channels; Step 6: Combine layered watermarks Form the final extracted watermark ,in, Represents the red, green and blue channels respectively.

[0012] This method uses the maximum coefficient of the image block matrix obtained by reversible matrix calculation to realize a method of color digital watermark embedding and blind extraction; this method has good watermark invisibility and robustness, and is simple and fast.

[0013] Verification of the effectiveness of the present invention In order to prove the effectiveness of the present invention, Figure 1 (a) Figure 1 (b) The two images shown are of different sizes. The 24-bit standard image is used as the host image, and the Figure 2 (a) Figure 2 (b) Two 24-bit color images of size 32×32 are used as digital watermarks for verification.

[0014] Figure 3 (a) Figure 3 (b) is to Figure 2 (a) Figure 2 (b) The watermarks shown are embedded into the host image Figure 1 (a) Figure 1 The structural similarity SSIM values ​​of the watermarked images obtained after (b) are 0.9845 and 0.9755, respectively, and the peak signal-to-noise ratio PSNR values ​​are 40.1674dB and 40.1647dB, respectively; Figure 4 (a) Figure 4 (b) are respectively from Figure 3 (a) Figure 3 The normalized cross-correlation coefficient NC values ​​of the watermarks extracted in (b) are 1.0000 and 1.0000 respectively.

[0015] Figure 5 (a) Figure 5 (b) Figure 5 (c) Figure 5 (d) Figure 5 (e) Figure 5 (f) is to Figure 3 The watermarked image shown in (a) is extracted after being attacked with JPEG compression (70), JPEG2000 (4:1), salt and pepper noise (0.2%), median filtering (3×3), scaling (400%), and shearing (12.5%). The normalized cross-correlation coefficient NC values ​​are 0.9997, 0.9992, 0.9944, 0.9687, 1.0000, and 0.9773, respectively.

[0016] Figure 6 (a) Figure 6 (b) Figure 6 (c) Figure 6 (d) Figure 6 (e) Figure 6 (f) is to Figure 3 The watermarked image shown in (b) is extracted after being attacked with JPEG compression (70), JPEG2000 (4:1), salt and pepper noise (0.2%), median filtering (3×3), scaling (400%), and shearing (12.5%). The normalized cross-correlation coefficient NC values ​​are 1.0000, 0.9985, 0.9928, 0.9862, 0.9979, and 0.9740, respectively.

[0017] The algorithm has been run nearly 10,000 times on the platform 2.30GHZ CPU, 32.00GB RAM, Win 11, MATLAB (R2021a). The average embedding time of the digital watermark is 0.5490 seconds, the average extraction time is 0.2031 seconds, and the total time is 0.7521 seconds.

[0018] In summary, the embedded color image digital watermark has good invisibility and meets the invisibility requirement of the watermark algorithm; at the same time, the color image digital watermark extracted from various attacked images has good identifiability and high NC value, which shows that the method has strong robustness.

Claims

1. A reversible matrix color image blind watermarking algorithm based on energy aggregation characteristics, characterized by This is achieved through a specific watermark embedding process and a watermark extraction process. The watermark embedding process is described as follows: Step 1: Take a picture of size 24-bit color watermark image Divided into three single color channels: red, green and blue , and each single color channel Key-based The Lorenz chaotic map is used to encrypt and convert each encrypted decimal pixel value into an 8-bit binary sequence; then, The binary sequences in are concatenated into a length of A string sequence ,in, Respectively represent the red, green and blue channels; Step 2: Make a picture of size 24-bit color host image Divide into 3 layered host images in the order of red, green and blue colors ; Then, each layered host image Divide into size non-overlapping image blocks; according to the layered watermark bit sequence Total length , using key-based The MD5 hashing pseudo-random algorithm selects non-repeating image blocks, where Respectively represent the red, green and blue channels; Step 3: Select one image block at a time , and use the reversible matrix to achieve high energy concentration according to formula (1); （1） in, represents the original image block matrix, Represents an image block The frequency domain matrix with high energy concentration is realized through the reversible matrix. is the side length of the image block matrix, represents a reversible matrix; Step 4: Layer the watermark bit sequence In the sequence, take out the watermark information to be embedded. According to the embedded watermark information and formula (2), the frequency domain coefficients in the upper left corner of the matrix are quantized to obtain the frequency domain coefficients containing the watermark ,in, Respectively represent the red, green and blue channels; （2） in, , Represents the judgment threshold in the quantization process, is the XOR function, is the remainder function, is the rounding function, Indicates the quantization step size; Step 5: According to the inverse transformation formula (3), the image block containing the watermark is obtained , and the watermarked image block matrix Update to its host image in layer The corresponding position in , where Respectively represent the red, green and blue channels; （3） in, is the matrix after embedding watermark, yes The inverse matrix of Step 6: Repeat steps 3 to 5 of this process until all watermark information is embedded, thus obtaining a layered host image with watermark. , and layer the red, green, and blue host images containing the watermark Reassemble and get the size Watermarked image ,in, Respectively represent the red, green and blue channels; The watermark extraction process is described as follows: Step 1: Set the size to Watermarked image Divided into 3 layers with watermark images , and each layered watermarked image Further divided into non-overlapping image patches, where Respectively represent the red, green and blue channels; Step 2: Select a watermarked image block in turn , and use the reversible matrix to achieve high energy concentration according to formula (4); （4） in, represents the watermarked image block matrix, Represents a watermarked image block The watermarked frequency domain matrix with high energy aggregation is realized through the reversible matrix. is the side length of the image block matrix, represents a reversible matrix; Step 3: Use formula (5) to extract the frequency domain matrix The watermark contained in ; （5） in, is a matrix The high energy coefficient in is the remainder function, is the rounding function towards zero, Indicates the quantization step size; Step 4: Repeat steps 2 and 3 of this process to extract the binary watermark bit sequence of each layer. , and then convert each 8-bit binary information into a decimal pixel value, where Respectively represent the red, green and blue channels; Step 5: Perform key-based analysis on the converted decimal pixels of each layer Decrypt the Lorenz chaotic map and obtain the layered watermark ,in, Respectively represent the red, green and blue channels; Step 6: Combine layered watermarks Form the final extracted watermark ,in, Represents the red, green and blue channels respectively.

Images