Multimedia anti-counterfeiting method based on target reflected light polarization characteristic coding
By using a method based on the polarization characteristics of the target reflected light to generate a transmission key using polarization degree and polarization angle images, and combining multiple encryption processes to embed information and perform QR code image processing, the problem of insufficient reliability and interpretability of image anti-counterfeiting detection in existing technologies is solved. This enables accurate detection and positioning of image tampering areas and enhances the robustness and anti-attack capability of watermarks.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-01
- Publication Date
- 2026-04-10
AI Technical Summary
Existing image anti-counterfeiting technologies are insufficient in detection range, accuracy, and robustness when facing complex scenarios such as multiple types of targets, local tampering, and deep synthesis. They are unable to provide highly reliable authenticity judgments, and traditional digital watermarking technology cannot reflect evidence of modification at the physical or semantic level of the image. Deep learning detection methods lack interpretability.
By using a method based on the polarization characteristics of the reflected light of the target to generate a transmission key using polarization degree and polarization angle images, and combining an improved RGB-LSB algorithm and multiple encryption processes (XOR, Logistic chaotic mapping, Arnold mapping) to embed information, the key is converted into a QR code image. The receiver can then detect tampered areas by restoring polarization information and reconstructing light intensity images.
It achieves accurate detection and localization of image tampering areas, enhances the robustness and anti-attack capability of watermarks, improves the credibility and interpretability of multimedia anti-counterfeiting detection, and breaks through the limitations of deep learning detection's training data dependence and black box uninterpretability.
Smart Images
Figure CN121842333A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the fields of image processing and optical information security technology, specifically to a multimedia anti-counterfeiting method based on the encoding of the polarization characteristics of the reflected light from a target. Background Technology
[0002] With the rapid evolution of image processing and compositing technologies, the methods of tampering with digital images and video content are becoming increasingly complex, encompassing both traditional image editing operations and AI-based synthetic forgery techniques based on deep learning (such as face swapping and voice-driven video generation). In the field of image anti-counterfeiting, digital watermarking and depth detection are currently the mainstream technologies: digital watermarking achieves content verification and traceability by embedding authentication information into images, and is widely used in copyright protection; depth detection relies on neural network models to learn the differences between real and forged images to identify forgery traces. However, facing complex scenarios such as multiple types of targets, local tampering, and deep synthesis, existing methods still have limitations in terms of detection range, accuracy, and robustness, making it difficult to provide highly reliable authenticity judgments. Therefore, how to improve the adaptability and interpretability of image anti-counterfeiting technologies has become a key issue that urgently needs to be addressed in the field of image authenticity verification.
[0003] Currently, the field of image anti-counterfeiting mainly adopts two types of technical solutions: digital watermarking-based image anti-counterfeiting methods achieve anti-counterfeiting by embedding authentication information (such as identity identifiers, feature codes, etc.) into the image. The typical process includes information encoding, embedding, extraction, and verification. For example, the least significant bit (LSB) algorithm hides information by replacing the least significant bit of the image pixel value, taking advantage of the low sensitivity of the human eye to small grayscale changes. Deep learning-based image anti-counterfeiting detection methods learn the difference features between real and fake images by training a neural network model. The process includes constructing a training dataset containing real samples and fake samples (such as those generated by AI synthesis models), extracting features such as texture, edge, and frequency, and finally outputting the authenticity judgment. Some methods also introduce time series information or visualization mechanisms to assist in analysis.
[0004] However, both of these technologies have significant drawbacks in practical applications. Digital watermarking technology typically embeds information unrelated to the image content, only verifying the presence of the watermark and failing to directly reflect whether the image itself has been modified. Once the watermark is tampered with or deleted, it cannot provide evidence of physical or semantic modification of the image, and its ability to identify new tampering methods such as deepfakes is limited. Deep learning detection technology's performance is highly dependent on the diversity and representativeness of the training dataset, and it is prone to failure when faced with new types of forgery models, targets, or styles that have not been seen before. At the same time, most methods output the overall authenticity probability of the image, lacking explicit localization of local forgery areas, and the model is a "black box structure," making it difficult to trace the internal judgment criteria and resulting in insufficient interpretability, which limits its application in high-confidence scenarios. Summary of the Invention
[0005] To address the aforementioned problems in the existing technology, this invention provides a multimedia anti-counterfeiting method based on the polarization characteristics encoding of target reflected light.
[0006] The technical problem to be solved by this invention is achieved through the following technical solution: In a first aspect, the present invention provides a multimedia anti-counterfeiting method based on the polarization characteristics encoding of target reflected light, comprising: The sender acquires multi-angle polarization images and raw optical images of the target, generates a transmission key based on a hash algorithm, and sends the transmission key to the receiver. The sender generates a polarization degree image and a polarization angle image based on the multi-angle polarization image; The sender uses an improved RGB-LSB algorithm to embed the polarization angle image into the original optical image to obtain an intermediate transmission image; the improved RGB-LSB algorithm is an RGB channel-level replacement process performed based on the preset low-order bits of the grayscale value of the embedded target by the embedded object. The sender uses multiple encryption processes based on the polarization image to encrypt the intermediate transmitted image, resulting in the final encrypted image. The multiple encryption processes are based on XOR operation, Logistic chaotic mapping method and Arnold mapping method. The sender converts the final encrypted image into a QR code image, and then uses an improved RGB-LSB algorithm to embed the QR code image pixel by pixel into the intermediate transmission image to obtain the final transmission image, which is then sent to the receiver. The receiver receives the final transmitted image and the transmission key, and uses the transmission key to perform polarization information restoration processing on the final transmitted image to obtain the polarization degree restored image and the polarization angle restored image. The receiver uses the polarization degree restored image and the polarization angle restored image to generate a reconstructed light intensity image, and then uses the reconstructed light intensity image and the final transmitted image to perform tampering area detection processing to obtain the multimedia anti-counterfeiting detection result.
[0007] Optionally, the sender generates a polarization degree image and a polarization angle image based on the multi-angle polarization image, including: The sender uses the Stokes vector method to generate the degree of polarization and polarization angle of the target at each pixel based on the multi-angle polarization image; The degree of polarization and the angle of polarization are mapped to an 8-bit grayscale image, respectively, to obtain the corresponding degree of polarization image and angle of polarization image.
[0008] Optionally, the sender uses an improved RGB-LSB algorithm to embed the polarization angle image into the original optical image to obtain an intermediate transmitted image, including: The original optical image is represented as an 8-bit grayscale image in the RGB three channels to obtain the first representation result; The polarization angle image is represented as an 8-bit grayscale image to obtain the second representation result; The second representation result is split into three parts, resulting in the first split result, the second split result, and the third split result; The preset low-level gray values of the B channel, G channel and R channel in the first representation result are replaced by the first splitting result, the second splitting result and the third splitting result respectively, to obtain the replaced B channel, the replaced G channel and the replaced R channel. The replaced B channel, replaced G channel, and replaced R channel constitute the intermediate transmission image.
[0009] Optionally, the intermediate transmitted image is represented as: ; ; ; in, This represents the grayscale value of the B channel corresponding to the intermediate transmitted image. This represents the grayscale value of the G channel corresponding to the intermediate transmitted image. This represents the grayscale value of the R channel corresponding to the intermediate transmitted image. The first polarization angle image represents the... The grayscale value of the bit. The first part representing the original optical image The grayscale value of the B channel corresponding to the bit. The first part representing the original optical image The grayscale value of the G channel corresponding to the bit. The first part representing the original optical image The grayscale value of the R channel corresponding to the bit. Represents the number of binary bits. , The first part representing the original optical image The grayscale value of the B channel corresponding to the bit.
[0010] Optionally, the sender encrypts the intermediate transmitted image based on the polarization image using multiple encryption processes to obtain the final encrypted image, including: The polarization image and the intermediate transmitted image are XORed to obtain the first XOR encryption result. Use the transmission key to generate chaotic mapping parameters and Arnold mapping parameters; Chaotic images are generated based on chaotic mapping parameters and the Logistic chaotic mapping method. Perform an XOR operation between the first XOR encryption result and the chaotic image to obtain the second XOR encryption result; The pixel coordinates of the second XOR encryption result are scrambled and perturbed using the Arnold mapping parameters to obtain the final encrypted image.
[0011] Optionally, the final transmitted image is represented as: ; ; ; in, This represents the grayscale value of the B channel corresponding to the final transmitted image. This represents the grayscale value of the G channel corresponding to the final transmitted image. This represents the grayscale value of the R channel corresponding to the final transmitted image. The first polarization angle image represents the... The grayscale value of the bit. The first part representing the original optical image The grayscale value of the B channel corresponding to the bit. The first part representing the original optical image The grayscale value of the G channel corresponding to the bit. The first part representing the original optical image The grayscale value of the R channel corresponding to the bit. Represents the number of binary bits. This refers to the QR code image represented as a binary image. Optionally, the receiver receives the final transmitted image and the transmission key, and uses the transmission key to perform polarization information restoration processing on the final transmitted image to obtain a polarization degree restored image and a polarization angle restored image, including: The receiver receives the final transmitted image and transmission key; The polarization angles in the RGB channels of the final transmitted image are extracted using the inverse operation of the improved RGB-LSB algorithm to reconstruct the image. The chaotic mapping parameters and Arnold mapping parameters are recovered using the transmission key; Based on the chaotic mapping parameters and the Arnold mapping parameters, the polarization degree restored image is obtained from the final transmitted image by using the inverse operation of multiple encryption processes.
[0012] Optionally, the receiver uses the polarization degree restored image and the polarization angle restored image to generate a reconstructed light intensity image, and uses the reconstructed light intensity image and the final transmitted image to perform tampering area detection processing to obtain multimedia anti-counterfeiting detection results, including: The receiver constructs a reconstructed light intensity image based on the polarization degree reconstructed image and the polarization angle reconstructed image, using Fresnel's law of reflection and the Lambert reflection model. The reconstructed light intensity image and the final transmitted image were sequentially processed using dual Gaussian filtering, edge extraction, and Euclidean distance calculation to perform texture difference comparison analysis and obtain the comparison results. The comparison results that exceed the difference threshold are marked and used as the multimedia anti-counterfeiting detection results.
[0013] Optionally, chaotic mapping parameters and Arnold mapping parameters are generated using the transmission key, including: The transmission key is processed by bidirectional hashing to generate a first hash value and a second hash value; The first hash value and the second hash value are concatenated to obtain a mixed hash value; A chaotic seed is generated using a hybrid hash value, and the chaotic seed is used as the initial value for pre-iteration processing using the Logistic chaotic mapping method to obtain the initial values of the chaotic mapping parameters and the Arnold mapping parameters. The initial values of the chaotic mapping parameters and the initial values of the Arnold mapping parameters are processed by a nonlinear perturbation algorithm to generate the chaotic mapping parameters and the Arnold mapping parameters.
[0014] Secondly, the present invention provides a multimedia anti-counterfeiting device based on target reflected light polarization characteristic encoding, comprising: a processor, a storage medium and a bus, wherein the storage medium stores machine-readable instructions executable by the processor. When the multimedia anti-counterfeiting device based on target reflected light polarization characteristic encoding is running, the processor communicates with the storage medium via the bus, and the processor executes the machine-readable instructions to perform the steps of the multimedia anti-counterfeiting method based on target reflected light polarization characteristic encoding described in the first aspect above.
[0015] This invention provides a multimedia anti-counterfeiting method based on the polarization characteristics encoding of target reflected light, comprising: a sender acquiring a multi-angle polarization image of the target and an original optical image, generating a transmission key based on a hash algorithm, and sending the transmission key to a receiver; the sender generating a polarization degree image and a polarization angle image based on the multi-angle polarization image; the sender embedding the polarization angle image into the original optical image using an improved RGB-LSB algorithm to obtain an intermediate transmission image; the improved RGB-LSB algorithm is an RGB channel-level replacement process performed on the preset low-order bits of the grayscale value of the embedded target based on the embedded object; the sender encrypting the intermediate transmission image using multiple encryption processes based on the polarization degree image to obtain a final encrypted image; the multiple encryption processes... The encryption process is based on XOR operation, Logistic chaotic mapping, and Arnold mapping. The sender converts the final encrypted image into a QR code image, and embeds the QR code image pixel by pixel into the intermediate transmission image using an improved RGB-LSB algorithm to obtain the final transmission image, which is then sent to the receiver. The receiver receives the final transmission image and the transmission key, and uses the transmission key to restore the polarization information of the final transmission image, obtaining a polarization degree restored image and a polarization angle restored image. The receiver uses the polarization degree restored image and the polarization angle restored image to generate a reconstructed light intensity image, and uses the reconstructed light intensity image and the final transmission image to perform tampering area detection processing to obtain the multimedia anti-counterfeiting detection result. In this invention, firstly, polarization degree images and polarization angle images are used as physical feature encodings to directly associate the embedded information with the optical properties of the original optical image. Once the original optical image is modified, its polarization characteristics will become abnormal, thus providing physical evidence of tampering. Secondly, the improved RGB-LSB algorithm, combined with multiple encryption methods (XOR, chaotic mapping, Arnold mapping), enhances the robustness and anti-attack capability of the watermark, avoiding the defects of traditional watermarks being easily deleted or tampered with. At the same time, the receiver can restore and reconstruct the light intensity image through polarization information, and can explicitly locate local forgery areas, breaking through the limitations of deep learning detection methods that rely on training data and are unexplainable by the "black box," thereby improving the credibility and interpretability of multimedia anti-counterfeiting detection.
[0016] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description
[0017] Figure 1 A flowchart illustrating a multimedia anti-counterfeiting method based on the polarization characteristics encoding of target reflected light, provided in an embodiment of the present invention; Figure 2 An exemplary schematic diagram illustrates the process of replacing pixels in the RGB channels of the original optical image using a polarization angle image and a QR code image; Figure 3An exemplary schematic diagram illustrates the overall process by which the sender encrypts the original optical image; Figure 4 An exemplary schematic diagram illustrates the overall process by which the receiver decrypts the final transmitted image. Figure 5 This is a schematic diagram of the structure of a multimedia anti-counterfeiting device based on the polarization characteristics encoding of target reflected light, provided as an embodiment of the present invention. Detailed Implementation
[0018] The purpose of this invention is to construct an anti-counterfeiting method that integrates polarization information perception, multiple encryption embedding, reversible decryption, and image physical consistency comparison. This method enables accurate detection and location of tampered areas in images or videos, while possessing good information security, concealment, and interpretability. By utilizing polarization imaging to obtain the polarization degree and polarization angle information of each pixel in the image and embedding it as an encryption feature into the image, combined with optical modeling to achieve light intensity image reconstruction and comparison, the comprehensive detection capability against traditional tampering and deepfakes is significantly improved, achieving stable, efficient, and physically interpretable verification of the authenticity of multimedia content.
[0019] The present invention will be further described in detail below with reference to specific embodiments, but the implementation of the present invention is not limited thereto.
[0020] To improve the credibility and interpretability of multimedia anti-counterfeiting detection, this invention provides a multimedia anti-counterfeiting method based on the polarization characteristics encoding of target reflected light. Figure 1 A flowchart illustrating a multimedia anti-counterfeiting method based on the polarization characteristics encoding of target reflected light, as provided in an embodiment of the present invention, is shown below. Figure 1 As shown, it includes: S101. The sender acquires the multi-angle polarization image and the original optical image of the target, generates a transmission key based on a hash algorithm, and sends the transmission key to the receiver.
[0021] In this embodiment, a split-focus plane polarization camera is used to image and acquire the target, obtaining multi-angle polarization images at four polarization angles: 0°, 45°, 90°, and 135°: I0, I 45 I 90 I 135 An optical camera is used to acquire the original optical image of the same target. Additionally, in this embodiment, a hash algorithm can be used to randomly generate a 256-bit high-entropy master key as the transmission key. The number of bits in the transmission key can be flexibly set according to requirements; this embodiment does not limit this.
[0022] S102. The sender generates a polarization degree image and a polarization angle image based on the multi-angle polarization image.
[0023] Optionally, S102 may specifically include: The sender uses the Stokes vector method to generate the degree of polarization and polarization angle of the target at each pixel based on the multi-angle polarization image; The degree of polarization and the angle of polarization are mapped to an 8-bit grayscale image, respectively, to obtain the corresponding degree of polarization image and angle of polarization image.
[0024] S103. The sender uses an improved RGB-LSB algorithm to embed the polarization angle image into the original optical image to obtain an intermediate transmission image.
[0025] Among them, the improved RGB-LSB algorithm is an RGB channel-level replacement process performed on the preset low-order bits of the grayscale value of the embedded target based on the embedded object.
[0026] Optionally, S103 may specifically include: The original optical image is represented as an 8-bit grayscale image in the RGB three channels to obtain the first representation result; The polarization angle image is represented as an 8-bit grayscale image to obtain the second representation result; The second representation result is split into three parts, resulting in the first split result, the second split result, and the third split result; The preset low-level gray values of the B channel, G channel and R channel in the first representation result are replaced by the first splitting result, the second splitting result and the third splitting result respectively, to obtain the replaced B channel, the replaced G channel and the replaced R channel. The replaced B channel, replaced G channel, and replaced R channel constitute the intermediate transmission image.
[0027] Wherein, the first indicates the result as follows: ; ; ; in, This represents the grayscale value of the R channel corresponding to the original optical image. This represents the grayscale value of the G channel corresponding to the original optical image. This represents the grayscale value of the B channel corresponding to the original optical image. The first part representing the original optical image The grayscale value of the B channel corresponding to the bit. The first part representing the original optical image The grayscale value of the G channel corresponding to the bit. The first part representing the original optical image The grayscale value of the R channel corresponding to the bit. Represents the number of binary bits. . The second representation result is as follows: ; in, This represents the polarization angle image after being represented as an 8-bit grayscale image. The first polarization angle image represents the... The grayscale value of the bit.
[0028] In this embodiment, The eight bits of binary data are divided into three parts. , as well as By replacing the last few bits of the RGB three channels of the original optical image with these three parts respectively, the intermediate transmission image is obtained. The result after replacement is: The intermediate transmitted image is represented as follows: ; ; ; in, This represents the grayscale value of the B channel corresponding to the intermediate transmitted image. This represents the grayscale value of the G channel corresponding to the intermediate transmitted image. This represents the grayscale value of the R channel corresponding to the intermediate transmitted image. The first part representing the original optical image The grayscale value of the B channel corresponding to the bit.
[0029] S104. The sender uses multiple encryption processes based on the polarization image to encrypt the intermediate transmitted image, thus obtaining the final encrypted image.
[0030] Among them, the multiple encryption processing is based on XOR operation, Logistic chaotic mapping method and Arnold mapping method.
[0031] Optionally, S104 may specifically include: The polarization image and the intermediate transmitted image are XORed to obtain the first XOR encryption result. Use the transmission key to generate chaotic mapping parameters and Arnold mapping parameters; Chaotic images are generated based on chaotic mapping parameters and the Logistic chaotic mapping method. Perform an XOR operation between the first XOR encryption result and the chaotic image to obtain the second XOR encryption result; The pixel coordinates of the second XOR encryption result are scrambled and perturbed using the Arnold mapping parameters to obtain the final encrypted image.
[0032] Since the polarization angle image has already been embedded into the intermediate transmitted image in the above embodiments, it is necessary to further embed the polarization degree image in this embodiment. Therefore, the polarization degree image and the intermediate transmitted image are first XORed to obtain the first XOR encryption result. XOR operation is a simple yet effective basic technique in image encryption. Its core is to confuse and encrypt image data through logical operations on binary bits, which is suitable for the binary image encryption requirements of this invention. Furthermore, a Logistic chaotic mapping is used to generate multiple random numbers, and these generated random numbers are used to form a chaotic image for encryption. The core of this algorithm lies in generating complex chaotic results through nonlinear recursive equations. When the chaotic mapping parameters... When the system is within the interval [3.57, 4], it enters a chaotic state (chaotic mapping parameters). Specifically, this can be achieved using a transmission key (and it exhibits high sensitivity to initial conditions, meaning that even small changes in initial values can lead to significant differences in the iteration results). When the chaotic mapping parameters are substituted... Then, the random data results are arranged sequentially to obtain a chaotic image. .
[0033] Arnold mapping parameters When the value is between 20 and 100, the image encryption effect is good. (Arnold mapping parameter) The specific value can be generated based on the transmission key. The specific generation method is described in the following embodiments and will not be repeated in this embodiment.
[0034] The final encrypted image is the result of the final encryption of the polarization degree image. At this point, if the polarization angle image, the parameters of the chaotic mapping, and the Arnold mapping parameters are not obtained simultaneously, it will be difficult to crack the accurate polarization degree information.
[0035] Optionally, chaotic mapping parameters and Arnold mapping parameters are generated using the transmission key, including: The transmission key is processed by bidirectional hashing to generate a first hash value and a second hash value; The first hash value and the second hash value are concatenated to obtain a mixed hash value; A chaotic seed is generated using a hybrid hash value, and the chaotic seed is used as the initial value for pre-iteration processing using the Logistic chaotic mapping method to obtain the initial values of the chaotic mapping parameters and the Arnold mapping parameters. The initial values of the chaotic mapping parameters and the initial values of the Arnold mapping parameters are processed by a nonlinear perturbation algorithm to generate the chaotic mapping parameters and the Arnold mapping parameters.
[0036] In this embodiment, when the transmission key is 256 bits, two independent 512-bit hash values can be generated through bidirectional hashing (SHA-3-512 and BLAKE2b). The first hash value... Second hash value ,Will and After concatenation, a 1024-bit mixed hash value is obtained. .Will Divided into 16 64-bit blocks , , ..., Calculate dynamic weights for each block. The weight coefficients are obtained by normalizing the dynamic weights of the 16 64-bit blocks. ,pass Generate weighted sum reuse Generate Chaos Seeds : ; in, Indicates the first 64-bit blocks For the first The value of a 64-bit block Represent it in decimal.
[0037] by Using the initial values, a Logistic chaotic mapping method is used for pre-iteration processing (the Logistic mapping parameter is initially set to 4, skipping the first 1000 transients) to obtain the initial values of the three parameters. Subsequently, a nonlinear perturbation algorithm is used to process the initial values of the three parameters through nonlinear perturbation mapping to generate the parameters required by the multiple encryption algorithm. , and ,in, and All are used as parameters for chaotic mapping. As the first parameter of the chaotic mapping As the second parameter of the chaotic mapping.
[0038] S105. The sender converts the final encrypted image into a QR code image, and embeds the QR code image pixel by pixel into the intermediate transmission image using an improved RGB-LSB algorithm to obtain the final transmission image, and then sends the final transmission image to the receiver.
[0039] Optionally, the final transmitted image is represented as: ; ; ; in, This represents the grayscale value of the B channel corresponding to the final transmitted image. This represents the grayscale value of the G channel corresponding to the final transmitted image. This represents the grayscale value of the R channel corresponding to the final transmitted image. The first polarization angle image represents the... The grayscale value of the bit. The first part representing the original optical image The grayscale value of the B channel corresponding to the bit. The first part representing the original optical image The grayscale value of the G channel corresponding to the bit. The first part representing the original optical image The grayscale value of the R channel corresponding to the bit. Represents the number of binary bits. This refers to the QR code image represented as a binary image. Figure 2 An exemplary diagram illustrates the process of replacing pixels in the RGB channels of the original optical image using a polarization angle image and a QR code image. For example... Figure 2 As shown, the 5, 6, and 7-bit grayscale values of the polarization angle image are used. 2, 3, and 4-bit grayscale values and 0 and 1 bit grayscale values The last three gray values (layers) of the R channel, the last three gray values of the G channel, and the second to last and third to last gray values of the B channel of the original optical image are replaced respectively. The gray value of the QR code image is then used to replace the last gray value of the G channel to obtain the final transmitted image.
[0040] Figure 3 An exemplary schematic diagram illustrates the overall process by which the sender encrypts the original optical image, such as... Figure 3 As shown, the sender first generates a chaotic image based on the transmission key, and then performs an XOR operation between the polarization degree image and the polarization angle image to obtain the final encrypted image. After generating a QR code image from the final encrypted image, the final transmitted image is obtained based on the polarization angle image, the original optical image, and the QR code image, combined with an improved RGB-LSB algorithm.
[0041] S106. The receiver receives the final transmitted image and the transmission key, and uses the transmission key to perform polarization information restoration processing on the final transmitted image to obtain the polarization degree restored image and the polarization angle restored image.
[0042] Optionally, S106 may specifically include: The receiver receives the final transmitted image and transmission key; The polarization angles in the RGB channels of the final transmitted image are extracted using the inverse operation of the improved RGB-LSB algorithm to reconstruct the image. The chaotic mapping parameters and Arnold mapping parameters are recovered using the transmission key; Based on the chaotic mapping parameters and the Arnold mapping parameters, the polarization degree restored image is obtained from the final transmitted image by using the inverse operation of multiple encryption processes.
[0043] Figure 4 An exemplary schematic diagram illustrates the overall process of the receiver decrypting the final transmitted image, as shown below. Figure 4 As shown, after obtaining the final transmitted image, the receiver uses the inverse process of the improved RGB-LSB algorithm to recover the polarization angle restored image and the QR code restored image. The QR code restored image is then used to obtain the final encrypted restored image, and the chaotic mapping parameters are recovered using the transmission key. These chaotic mapping parameters are then combined with the final encrypted restored image to generate the polarization degree restored image. Furthermore, if the receiver obtains an incorrect transmission key, a garbled image will be generated, meaning the restoration of the polarization angle and polarization degree will be impossible.
[0044] S107. The receiver uses the polarization degree restored image and the polarization angle restored image to generate a reconstructed light intensity image, and uses the reconstructed light intensity image and the final transmitted image to perform tampering area detection processing to obtain the multimedia anti-counterfeiting detection result.
[0045] Optionally, S107 may specifically include: The receiver constructs a reconstructed light intensity image based on the polarization degree reconstructed image and the polarization angle reconstructed image, using Fresnel's law of reflection and the Lambert reflection model. The reconstructed light intensity image and the final transmitted image were sequentially processed using dual Gaussian filtering, edge extraction, and Euclidean distance calculation to perform texture difference comparison analysis and obtain the comparison results. The comparison results that exceed the difference threshold are marked and used as the multimedia anti-counterfeiting detection results.
[0046] To enable comparison between the reconstructed light intensity image and the final transmitted image, and to accurately identify and locate any tampered areas, preprocessing of both images is necessary. Specifically, a double Gaussian filter is applied to both images. The first Gaussian filter uses a larger standard deviation, while the second uses a smaller standard deviation. This double Gaussian filtering smooths out subtle noise in both the reconstructed light intensity image and the final transmitted image, preventing high-frequency interference in subsequent gradient calculations. Then, the Sobel operator is used to extract edges from both the filtered reconstructed light intensity image and the final transmitted image, yielding the final transmitted edge image. and reconstructed light intensity edge image .
[0047] Finally, calculate Each point in the middle The Euclidean distance to each point in the interval, and then for For each point in the array, find its position in the array. The nearest neighbor in the middle, if The point in the middle corresponds to If the nearest neighbor in the data is greater than the difference threshold, then... The point in the code is marked as the result of the multimedia anti-counterfeiting detection.
[0048] In addition, in order to clearly and accurately mark the tampered areas in the final transmitted image, morphological processing such as scatter removal and dilation operations were performed on the point set at the marked location. At the same time, the connected regions were analyzed and fitted, ultimately achieving the goal of marking the tampered areas in the transmitted image.
[0049] This invention provides a multimedia anti-counterfeiting method based on the polarization characteristics encoding of target reflected light, comprising: a sender acquiring a multi-angle polarization image of the target and an original optical image, generating a transmission key based on a hash algorithm, and sending the transmission key to a receiver; the sender generating a polarization degree image and a polarization angle image based on the multi-angle polarization image; the sender embedding the polarization angle image into the original optical image using an improved RGB-LSB algorithm to obtain an intermediate transmission image; the improved RGB-LSB algorithm is an RGB channel-level replacement process performed on the preset low-order bits of the grayscale value of the embedded target based on the embedded object; the sender encrypting the intermediate transmission image using multiple encryption processes based on the polarization degree image to obtain a final encrypted image; multiple encryption... The encryption process is based on XOR operation, Logistic chaotic mapping, and Arnold mapping. The sender converts the final encrypted image into a QR code image and embeds the QR code image pixel by pixel into the intermediate transmission image using an improved RGB-LSB algorithm to obtain the final transmission image, which is then sent to the receiver. The receiver receives the final transmission image and the transmission key, and uses the transmission key to restore the polarization information of the final transmission image, obtaining a polarization degree restored image and a polarization angle restored image. The receiver uses the polarization degree restored image and the polarization angle restored image to generate a reconstructed light intensity image, and uses the reconstructed light intensity image and the final transmission image to perform tampering area detection processing to obtain the multimedia anti-counterfeiting detection result. In this embodiment of the invention, firstly, polarization degree images and polarization angle images are used as physical feature encodings to directly associate the embedded information with the optical properties of the original optical image. Once the original optical image is modified, its polarization characteristics will become abnormal, thus providing physical evidence of tampering. Secondly, the improved RGB-LSB algorithm, combined with multiple encryption methods (XOR, chaotic mapping, Arnold mapping), enhances the robustness and anti-attack capability of the watermark, avoiding the defects of traditional watermarks being easily deleted or tampered with. At the same time, the receiver can restore and reconstruct the light intensity image through polarization information, and can explicitly locate the local forgery area, breaking through the limitations of deep learning detection methods that rely on training data and are unexplainable by the "black box," thereby improving the credibility and interpretability of multimedia anti-counterfeiting detection.
[0050] The method provided in this embodiment of the invention can be applied to electronic devices. Specifically, the electronic device can be a desktop computer, a portable computer, a smart mobile terminal, a server, etc., and this embodiment of the invention does not limit the application to such devices.
[0051] Figure 5A schematic diagram of a multimedia anti-counterfeiting device based on target reflected light polarization characteristic encoding provided in an embodiment of the present invention includes: a processor 510, a storage medium 520, and a bus 530. The storage medium 520 stores machine-readable instructions executable by the processor 510. When the multimedia anti-counterfeiting device based on target reflected light polarization characteristic encoding is running, the processor 510 and the storage medium 520 communicate via the bus 530. The processor 510 executes the machine-readable instructions to perform the steps of the above-described method embodiment. Specific implementation methods and technical effects are similar and will not be repeated here.
[0052] The storage medium may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage device. Optionally, the storage medium may also be at least one storage device located remotely from the aforementioned processor.
[0053] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.
[0054] It should be noted that the terms "first," "second," etc., are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Rather, they are merely examples of apparatuses and methods consistent with some aspects of the invention.
[0055] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Furthermore, those skilled in the art can combine and integrate the different embodiments or examples described in this specification.
[0056] Although the invention has been described herein in conjunction with various embodiments, those skilled in the art, by reviewing the accompanying drawings and the disclosure, will understand and implement other variations of the disclosed embodiments in carrying out the claimed invention. In this description, the word "comprising" does not exclude other components or steps, "a" or "an" does not exclude a plurality, and "a plurality" means two or more, unless otherwise explicitly specified. Furthermore, while different embodiments may describe certain measures, this does not mean that these measures cannot be combined to produce good results.
[0057] The above description, in conjunction with specific preferred embodiments, provides a further detailed explanation of the present invention. It should not be construed that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art, various simple deductions or substitutions can be made without departing from the inventive concept, and all such modifications and substitutions should be considered within the scope of protection of the present invention.
Claims
1. A multimedia anti-counterfeiting method based on the polarization characteristics encoding of target reflected light, characterized in that, include: The sender acquires multi-angle polarization images and raw optical images of the target, generates a transmission key based on a hash algorithm, and sends the transmission key to the receiver. Based on the multi-angle polarization image, the sender generates a polarization degree image and a polarization angle image; The sender uses an improved RGB-LSB algorithm to embed the polarization angle image into the original optical image to obtain an intermediate transmission image; the improved RGB-LSB algorithm is an RGB channel-level replacement process performed on the preset low-order bits of the grayscale value of the embedded target based on the embedded object; The sender encrypts the intermediate transmitted image based on the polarization image using multiple encryption processes to obtain the final encrypted image; the multiple encryption processes are based on XOR operation, Logistic chaotic mapping method and Arnold mapping method. The sender converts the final encrypted image into a QR code image, and embeds the QR code image pixel by pixel into the intermediate transmission image using the improved RGB-LSB algorithm to obtain the final transmission image, and then sends the final transmission image to the receiver. The receiver receives the final transmitted image and the transmission key, and uses the transmission key to perform polarization information restoration processing on the final transmitted image to obtain a polarization degree restored image and a polarization angle restored image. The receiver uses the polarization degree restored image and the polarization angle restored image to generate a reconstructed light intensity image, and uses the reconstructed light intensity image and the final transmitted image to perform tampering area detection processing to obtain multimedia anti-counterfeiting detection results.
2. The multimedia anti-counterfeiting method based on the polarization characteristics encoding of target reflected light according to claim 1, characterized in that, The sender generates a polarization degree image and a polarization angle image based on the multi-angle polarization image, including: Based on the multi-angle polarization image, the sender uses the Stokes vector calculation method to generate the polarization degree and polarization angle of the target at each pixel; The degree of polarization and the angle of polarization are mapped to 8-bit grayscale images respectively, to obtain the degree of polarization image and the angle of polarization image.
3. The multimedia anti-counterfeiting method based on the polarization characteristics encoding of target reflected light according to claim 1, characterized in that, The sender uses an improved RGB-LSB algorithm to embed the polarization angle image into the original optical image to obtain an intermediate transmission image, including: The original optical image is represented as an 8-bit grayscale image in the RGB three channels to obtain the first representation result; The polarization angle image is represented as an 8-bit grayscale image to obtain the second representation result; The second representation result is split into three parts to obtain the first split result, the second split result, and the third split result; The preset low-level gray values of the B channel, G channel and R channel in the first representation result are replaced by the first splitting result, the second splitting result and the third splitting result respectively, to obtain the replaced B channel, the replaced G channel and the replaced R channel; The replaced B channel, the replaced G channel, and the replaced R channel constitute the intermediate transmission image.
4. The multimedia anti-counterfeiting method based on the polarization characteristics encoding of target reflected light according to claim 3, characterized in that, The intermediate transmitted image is represented as follows: ； ； ； in, This represents the grayscale value of the B channel corresponding to the intermediate transmitted image. This represents the grayscale value of the G channel corresponding to the intermediate transmitted image. This represents the grayscale value of the R channel corresponding to the intermediate transmitted image. The first polarization angle image The grayscale value of the bit. The first part representing the original optical image The grayscale value of the B channel corresponding to the bit. The first part representing the original optical image The grayscale value of the G channel corresponding to the bit. The first part representing the original optical image The grayscale value of the R channel corresponding to the bit. Represents the number of binary bits. , The first part representing the original optical image The grayscale value of the B channel corresponding to the bit.
5. The multimedia anti-counterfeiting method based on the polarization characteristics encoding of target reflected light according to claim 1, characterized in that, The sender, based on the polarization image, employs multiple encryption processes to encrypt the intermediate transmitted image, obtaining a final encrypted image, including: Perform an XOR operation on the polarization image and the intermediate transmission image to obtain the first XOR encryption result; The transmission key is used to generate chaotic mapping parameters and Arnold mapping parameters; A chaotic image is generated based on the chaotic mapping parameters and the Logistic chaotic mapping method; Perform an XOR operation between the first XOR encryption result and the chaotic image to obtain the second XOR encryption result; The pixel coordinates of the second XOR encryption result are scrambled and perturbed using the Arnold mapping parameters to obtain the final encrypted image.
6. The multimedia anti-counterfeiting method based on the polarization characteristics encoding of target reflected light according to claim 1, characterized in that, The final transmitted image is represented as follows: ； ； ； in, This represents the grayscale value of the B channel corresponding to the final transmitted image. This represents the grayscale value of the G channel corresponding to the final transmitted image. This represents the grayscale value of the R channel corresponding to the final transmitted image. The first polarization angle image The grayscale value of the bit. The first part representing the original optical image The grayscale value of the B channel corresponding to the bit. The first part representing the original optical image The grayscale value of the G channel corresponding to the bit. The first part representing the original optical image The grayscale value of the R channel corresponding to the bit. Represents the number of binary bits. This refers to the QR code image represented as a binary image.
7. The multimedia anti-counterfeiting method based on the polarization characteristics encoding of target reflected light according to claim 1, characterized in that, The receiver receives the final transmitted image and the transmission key, and uses the transmission key to perform polarization information restoration processing on the final transmitted image to obtain a polarization degree restored image and a polarization angle restored image, including: The receiver receives the final transmitted image and the transmission key; The polarization angles in the RGB channels of the final transmitted image are extracted and the image is restored using the inverse operation of the improved RGB-LSB algorithm. The chaotic mapping parameters and Arnold mapping parameters are recovered using the transmission key; Based on the chaotic mapping parameters and the Arnold mapping parameters, the polarization degree restored image is obtained from the final transmitted image by using the inverse operation of the multiple encryption process.
8. The multimedia anti-counterfeiting method based on the polarization characteristics encoding of target reflected light according to claim 1, characterized in that, The receiver uses the polarization degree restored image and the polarization angle restored image to generate a reconstructed light intensity image, and uses the reconstructed light intensity image and the final transmitted image to perform tampering area detection processing to obtain multimedia anti-counterfeiting detection results, including: The receiver constructs the reconstructed light intensity image based on the polarization degree restored image and the polarization angle restored image, using Fresnel's law of reflection and the Lambert reflection model. The reconstructed light intensity image and the final transmitted image are sequentially processed using dual Gaussian filtering, edge extraction, and Euclidean distance calculation to perform texture difference comparison analysis and obtain the comparison results. The comparison results that are greater than the difference threshold are marked and used as the multimedia anti-counterfeiting detection results.
9. The multimedia anti-counterfeiting method based on the polarization characteristics encoding of target reflected light according to claim 5, characterized in that, The step of generating chaotic mapping parameters and Arnold mapping parameters using the transmission key includes: The transmission key is processed by bidirectional hashing to generate a first hash value and a second hash value; The first hash value and the second hash value are concatenated to obtain a mixed hash value; The mixed hash value is used to generate a chaotic seed, and the chaotic seed is used as the initial value to perform pre-iteration processing using the Logistic chaotic mapping method to obtain the initial values of the chaotic mapping parameters and the initial values of the Arnold mapping parameters. The initial values of the chaotic mapping parameters and the initial values of the Arnold mapping parameters are processed by a nonlinear perturbation algorithm to generate the chaotic mapping parameters and the Arnold mapping parameters.
10. A multimedia anti-counterfeiting device based on the polarization characteristics encoding of target reflected light, characterized in that, include: The device includes a processor, a storage medium, and a bus. The storage medium stores machine-readable instructions executable by the processor. When the multimedia anti-counterfeiting device based on the polarization characteristics of target reflected light is running, the processor communicates with the storage medium via the bus. The processor executes the machine-readable instructions to perform the steps of the multimedia anti-counterfeiting method based on the polarization characteristics of target reflected light as described in any one of claims 1-9.