SECURE WATERMARKING ALGORITHM WITH HARDWARE IMPLEMENTATION UTILIZING PHASE CONGRUENCY AND SINGULAR VALUE DECOMPOSITION TECHNIQUES

Abstract

ABSTRACT “SECURE WATERMARKING ALGORITHM WITH HARDWARE IMPLEMENTATION UTILIZING PHASE CONGRUENCY AND SINGULAR VALUE DECOMPOSITION TECHNIQUES” The present invention provides secure watermarking algorithm with hardware implementation utilizing phase congruency and singular value decomposition techniques. The effortless accessibility of digital information and the simplicity of the digital systems have left the contents over the digital media extremely insecure. The need for mechanisms to protect such information is undeniable. Digital watermark based information hiding is a prospective means for copyright protection, authentication, integrity verification and intellectual property right protection. Phase congruency technique works on the principle that perceptually significant image features have effect at spatial locations, where the essential Fourier components are maximally in phase with one another. An adaptive digital watermarking algorithm for better performance in multi-parametric solution space is developed here. This algorithm is developed for hiding the copyright information by means of phase congruency and SVD supported information hiding technique. Figure 1

Specification

Description:TECHNICAL FIELD
[0001] The present invention relates to the field of computer science, and more particularly, the present invention relates to the secure watermarking algorithm with hardware implementation utilizing phase congruency and singular value decomposition techniques.
BACKGROUND ART
[0002] The following discussion of the background of the invention is intended to facilitate an understanding of the present invention. However, it should be appreciated that the discussion is not an acknowledgment or admission that any of the material referred to was published, known, or part of the common general knowledge in any jurisdiction as of the application's priority date. The details provided herein the background if belongs to any publication is taken only as a reference for describing the problems, in general terminologies or principles or both of science and technology in the associated prior art.
[0003] In Phase congruency, the feature points are more related to phase than the amplitude of a signal. Hence, phase congruency reflects the frequency domain behavior of an image. The phase congruency results invariant quantities to measure the image feature and it remains constant throughout the image. The dimensionless value of Phase Congruency varies between 0 and 1, where 1 indicates most significant feature and 0 indicates not significant. Local energy model postulates features are perceived in the location of high phase congruency.
[0004] The watermark data in this project is an image of a fingerprint which is assumed to be a unique watermark data. The watermark data is embedded using well known SVD technique in the phase congruency mapping points on the host image. It is perceived that the visual quality of the image has not deteriorated to an unacceptable level on embedding the data. The watermark is invisible and the algorithm is categorized as a non-blind digital watermarking algorithm as original image is used during the detection process to detect the watermark.
[0005] In light of the foregoing, there is a need for Secure watermarking algorithm with hardware implementation utilizing phase congruency and singular value decomposition techniques that overcomes problems prevalent in the prior art associated with the traditionally available method or system, of the above-mentioned inventions that can be used with the presented disclosed technique with or without modification.
[0006] All publications herein are incorporated by reference to the same extent as if each individual publication or patent application were specifically and individually indicated to be incorporated by reference. Where a definition or use of a term in an incorporated reference is inconsistent or contrary to the definition of that term provided herein, the definition of that term provided herein applies, and the definition of that term in the reference does not apply.
OBJECTS OF THE INVENTION
[0007] The principal object of the present invention is to overcome the disadvantages of the prior art by providing secure watermarking algorithm with hardware implementation utilizing phase congruency and singular value decomposition techniques.
[0008] Another object of the present invention is to provide secure watermarking algorithm with hardware implementation utilizing phase congruency and singular value decomposition techniques that develops and implements the hardware of a highly secured and authenticated watermarking algorithm based on feature detection property of phase congruency and SVD to protect the copyright of digital contents.
[0009] Another object of the present invention is to provide secure watermarking algorithm with hardware implementation utilizing phase congruency and singular value decomposition techniques that utilizes the lesser prominent image feature regions for hiding the copyright information by means of phase congruency and SVD scheme.
[0010] Another object of the present invention is to provide secure watermarking algorithm with hardware implementation utilizing phase congruency and singular value decomposition techniques, wherein the functional simulation was performed using Xilinx ISE 14.3 simulator and the algorithm was implemented on high end FPGA device.
[0011] Another object of the present invention is to provide secure watermarking algorithm with hardware implementation utilizing phase congruency and singular value decomposition techniques, wherein the experimental analysis discloses the robustness of the algorithm as it stands against various attacks like compression, geometrical distortion and common signal processing operations.
[0012] The foregoing and other objects of the present invention will become readily apparent upon further review of the following detailed description of the embodiments as illustrated in the accompanying drawings.
SUMMARY OF THE INVENTION
[0013] The present invention relates to secure watermarking algorithm with hardware implementation utilizing phase congruency and singular value decomposition techniques.
[0014] The effortless accessibility of digital information and the simplicity of the digital systems have left the contents over the digital media extremely insecure. The need for mechanisms to protect such information is undeniable. Digital watermark based information hiding is a prospective means for copyright protection, authentication, integrity verification and intellectual property right protection. Phase congruency technique works on the principle that perceptually significant image features have effect at spatial locations, where the essential Fourier components are maximally in phase with one another. An adaptive digital watermarking algorithm for better performance in multi-parametric solution space is developed here. This algorithm is developed for hiding the copyright information by means of phase congruency and SVD supported information hiding technique. Performance evaluation of the algorithm is performed using simulation in Matlab environment in terms of PSNR,NCC and SSIM index. Hardware realization up to the RTL schematic level has been performed using high syntax VHDL code and Xilinx ISE 14.3 simulator and high performance FPGA board. The experimental analysis establishes better robustness of the proposed algorithm as it stands against various attacks (like compression, geometrical distortion, and common signal processing operations) along with better data hiding capacity. Comparison with some modern techniques recommends that proposed idea can be an effective tool for copyright protection and authentication.
[0015] In the present work, we have developed and implemented the hardware of a highly secured and authenticated watermarking algorithm based on feature detection property of phase congruency and SVD to protect the copyright of digital contents. The scheme utilizes the lesser prominent image feature regions for hiding the copyright information by means of phase congruency and SVD scheme. The architecture was modeled using VHDL and synthesized using Virtex-7 technology. Functional simulation was performed using Xilinx ISE 14.3 simulator and the algorithm was implemented on high end FPGA device. The chip was tested using hardware co-simulation at the Maximum clock frequency of 351.457MHz with minimum period of 2.845 ns. Finally, the processor is presented as a synthesizable module. The experimental analysis discloses the robustness of the algorithm as it stands against various attacks like compression, geometrical distortion and common signal processing operations. Additionally, data hiding capacity and imperceptibility of the algorithm is also superior. Comparisons with some modern techniques recommend that the proposed idea can be effective tool for copyright protection and authentication.
[0016] While the invention has been described and shown with reference to the preferred embodiment, it will be apparent that variations might be possible that would fall within the scope of the present invention.
BRIEF DESCRIPTION OF DRAWINGS
[0017] So that the manner in which the above-recited features of the present invention can be understood in detail, a more particular description of the invention, briefly summarized above, may have been referred by embodiments, some of which are illustrated in the appended drawings. It is to be noted, however, that the appended drawings illustrate only typical embodiments of this invention and are therefore not to be considered limiting of its scope, for the invention may admit to other equally effective embodiments.
[0018] These and other features, benefits, and advantages of the present invention will become apparent by reference to the following text figure, with like reference numbers referring to like structures across the views, wherein:
[0019] Fig. 1 Proposed watermark embedding scheme.
[0020] Fig. 2 Operational flow chart for feature Point extraction.
[0021] Fig. 3 Operational flow chart for Embedding Algorithm
[0022] Fig. 4(a) Host image Lena (512×512).
[0023] Fig. 4(b) Fingerprint 345×345 converted to 128×128.
[0024] Fig. 4(c) Non-Compressed stego.
[0025] Fig. 5 The proposed watermark Extraction scheme.
[0026] Fig. 6 Operational flow chart for Watermark Extraction.
[0027] Fig.7 Graph shows the NCC values with rotation up to 180o.
[0028] Fig.8. Increasing SSIM value as it approaches towards the original image (Lena).
[0029] Fig. 9(a): Test bench simulation result for entire image blocks
[0030] Fig. 9(b) Test bench simulation result for entire water marked image blocks
[0031] Fig. 9(c) Test bench simulation result for the decoder circuit of the water marked image blocks.
[0032] Fig.10 RTL Schematic view of processor incorporating the watermark embedding algorithm.
[0033] Fig.11 RTL Schematic view of the decoder module using Xilinx 14.3 Plan ahead tools
DETAILED DESCRIPTION OF THE INVENTION
[0034] While the present invention is described herein by way of example using embodiments and illustrative drawings, those skilled in the art will recognize that the invention is not limited to the embodiments of drawing or drawings described and are not intended to represent the scale of the various components. Further, some components that may form a part of the invention may not be illustrated in certain figures, for ease of illustration, and such omissions do not limit the embodiments outlined in any way. It should be understood that the drawings and the detailed description thereto are not intended to limit the invention to the particular form disclosed, but on the contrary, the invention is to cover all modifications, equivalents, and alternatives falling within the scope of the present invention as defined by the appended claim.
[0035] As used throughout this description, the word "may" is used in a permissive sense (i.e. meaning having the potential to), rather than the mandatory sense, (i.e. meaning must). Further, the words "a" or "an" mean "at least one" and the word "plurality" means "one or more" unless otherwise mentioned. Furthermore, the terminology and phraseology used herein are solely used for descriptive purposes and should not be construed as limiting in scope. Language such as "including," "comprising," "having," "containing," or "involving," and variations thereof, is intended to be broad and encompass the subject matter listed thereafter, equivalents, and additional subject matter not recited, and is not intended to exclude other additives, components, integers, or steps. Likewise, the term "comprising" is considered synonymous with the terms "including" or "containing" for applicable legal purposes. Any discussion of documents, acts, materials, devices, articles, and the like are included in the specification solely for the purpose of providing a context for the present invention. It is not suggested or represented that any or all these matters form part of the prior art base or were common general knowledge in the field relevant to the present invention.
[0036] In this disclosure, whenever a composition or an element or a group of elements is preceded with the transitional phrase "comprising", it is understood that we also contemplate the same composition, element, or group of elements with transitional phrases "consisting of", "consisting", "selected from the group of consisting of, "including", or "is" preceding the recitation of the composition, element or group of elements and vice versa.
[0037] The present invention is described hereinafter by various embodiments with reference to the accompanying drawing, wherein reference numerals used in the accompanying drawing correspond to the like elements throughout the description. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiment set forth herein. Rather, the embodiment is provided so that this disclosure will be thorough and complete and will fully convey the scope of the invention to those skilled in the art. In the following detailed description, numeric values and ranges are provided for various aspects of the implementations described. These values and ranges are to be treated as examples only and are not intended to limit the scope of the claims. In addition, several materials are identified as suitable for various facets of the implementations. These materials are to be treated as exemplary and are not intended to limit the scope of the invention.
[0038] The present invention relates to secure watermarking algorithm with hardware implementation utilizing phase congruency and singular value decomposition techniques.
[0039] In Phase congruency, the feature points are more related to phase than the amplitude of a signal. Hence, phase congruency reflects the frequency domain behavior of an image. The phase congruency results invariant quantities to measure the image feature and it remains constant throughout the image. The dimensionless value of Phase Congruency varies between 0 and 1, where 1 indicates most significant feature and 0 indicates not significant. Local energy model [10] postulates features are perceived in the location of high phase congruency.
[0040] The watermark data in this project is an image of a fingerprint which is assumed to be a unique watermark data. The watermark data is embedded using well known SVD technique in the phase congruency mapping points on the host image. It is perceived that the visual quality of the image has not deteriorated to an unacceptable level on embedding the data. The watermark is invisible and the algorithm is categorized as a non-blind digital watermarking algorithm as original image is used during the detection process to detect the watermark.
[0041] A novel and highly robust phase-congruency and SVD based digital watermarking algorithm has been implemented in this work. SVD is a basically a factorization of a real or complex matrix. Usually, for an m×n real or complex matrix the singular value decomposition, M is the factorization of the form, M = USVT , Where U is a m×m real or complex unitary matrix, S is an m×n rectangular diagonal matrix with non-negative real numbers on the diagonal and VT is an n×n real or complex unitary matrix.
[0042] The operational block diagram for the embedding mechanism is shown in Fig.1. We have designed phase congruency block to incorporate the feature point extraction algorithm and the watermark embedding block incorporating the watermark embedding algorithm.
[0043] A. Feature Point Extraction: We determine the feature points of the original watermark image for embedding purpose, henceforth instead of the original image; watermark is the feature region of the original watermark. This procedure is adopted to reduce the amount of bits required for watermarking on the original image. Our algorithm is mostly suitable for fingerprinting application as different fingerprint have its own unique features. That is, it can be applied for copyright protection of multiple owners, each having its unique watermark. The proposed feature point extraction and watermark embedding algorithm are described in Fig.2 and Fig.3 respectively.
[0044] The detailed method for choosing the points or regions for hiding the information is based on the following steps:
[0045] A. Feature Point extraction Algorithm:
- Step1: Find the phase congruent map having same dimension of the watermark.
- Step2: Find the total number of non-zero points of the phase congruent map.
- Step 3: Let n2 number of non-zero points determined. Now these non-zero locations are utilized to fetch gray scale values in particular order from the original watermark into a matrix Mw of dimension n×n.
- Step 4: I f n×n is the order of the watermark matrix,
- then n2 number of points in the original image is required to embed the data.
- Feature map of the original image is determined.
- Numerical value of n2 is within limit and justified.
- Step 5: The particular value of phase congruency of the host image which is mapped between 0-255 is determined above which the number of points are exactly or just exactly above the n2 value.
- [Initialize a Phase congruency value G=128]
- Step 6: If the number of points > n2
- then n2 number of location are fetched in a particular order which increases the security.
- If number of points = n2
- then, it doesn't need the ordering.
[0046] That is, it depends on the amount of feature points of the host.
[0047] B. Embedding Algorithm: Singular Value Decomposition (SVD) is a powerful technique of watermark embedding [11-12]. Let M be a grayscale image matrix with size of m × n. Every real matrix M can be decomposed into three matrices as M = USVT by SVD, where U and V are orthogonal matrices with size of m × m and n × n respectively S is a diagonal matrix with the same size as matrix M. Fig. 3 describes the operational flow chart for Embedding Algorithm.
[0048] B. Embedding Algorithm:
- Step 1: SVD is performed on both the matrices
Mo =UoSoVo and Mw = UwSwVw
- Step 2: Now a new matrix
Sem = So + q*Sw is defined. Here |q| ˂= 1.
- Step 3: Value of q is determined in such a way that the matrix elements of Mem = UoSemVo , shall not exceed 510 gray scale level or less than -255 gray scale value.
- Step 4: The negative values (if any) of the matrix Mem is adjusted and kept within the grayscale range.
- Step 5: Now this change on the matrix Mo which is made via Mem continue to effect on the corresponding points of the original image from where the matrix Mo is fetched.
[0049] The information about n2, the particular phase congruency level and the order of data extraction along with some parameters [q,C] are kept secret for the watermark detection.
[0050] In the algorithm it is considered, Min (Mem (i,j) ) = A & Max(Mem (i,j)) = B. Now, a value 'C' is determined as Max (|A|,|B|) = C . The Mem matrix is adjusted following a rule as: if( A <= 0) then Mem (i,j) = Mem(i,j) + C &if( B >= 255) then Mem (i,j) = Mem (i,j) - C in order to abolish the negative values and keep the values within the grayscale range.
[0051] The algorithm is highly secure as G, C & q are kept secret. These values in turn depend on the feature information of both the host and the watermark. So if fingerprint as a unique watermark is used then it would be highly secure and difficult to intercept.
[0052] In this work, host image Lena 512×512 in Fig. 4(a) is considered. The finger print of 345×345 in Fig. 4 (b) is converted to achieve a unique watermark of 128×128 using Phase Congruency.
[0053] Firstly, we give the test image or the host, a fingerprint used as a watermark (without the payload reduction) and the watermarked image. The watermark is bypassed to a phase congruency function which only retains the outline or the important feature region of the watermark. This reduces the payload to an acceptable level consistent with the capacity of the host image. The stego Fig.4(c) followed by the embedding process do not undergo any perceptible visual distortion.
[0054] C. Watermark Extraction Algorithm:
- Step 1: The watermarked image is reconfigured to its original shape (if scaled and rotated) using the original image.
- Step 2: Here Phase Congruency map is used to find the watermarked feature region.
- Step 3: The gray scale values of the feature points where the watermark data are embedded are kept on a matrix [G value is previously stored].
- Step 4: Gray Scale values are fetched to a matrix Mat in a predetermined order using the embedding locations of the undistorted stego.
- Step 5: Mat matrix is adjusted to Mat1 considering the negative or the out of bound gray scale values while embedding the watermark.
- Step 6: The SVD is performed on the given matrix
Mat1 = Uat1 Sat1 Vat1 .
Then the tampered watermarked matrix is given by, Swat = ( Sat1- So)/q.
where q is previously determined and |q| ˂= 1.
- Step 7: The tampered watermarked matrix is given by, Mwat = Uw SwatVw . Now Mwat is compared with Mw to detect the watermark.
[0055] Extraction algorithm uses the previously stored secret values G, C and q to extract the watermark along with the So and Mo values from the original image. Finally, the extracted image is compared with the watermark Mw. It is a very simple but highly secured algorithm showing sufficient resistance against various attacks, which we will verify following some experiments in Matlab environment and simulation results determining the performance evaluation matrices like, PSNR, NCC and SSIM values. Non-negative real numbers are used for SVD technique.
[0056] II. EXPERIMENTAL RESULTS IN MAT LAB: Feature regions of both the image (landscape.jpg) and watermark (wat.jpg) have been determined. The feature region of the wat.jpg is now the watermark to be embedded. All attacks are stimulated after the image is stored in .jpg format. The results of the extracted watermark under different attacks following the extraction algorithm are shown in the Table I.
[0057] The results after stimulating different kind of attack generally encountered in watermarking process or during transmission process evidently show us the strong robustness of this algorithm. This algorithm especially exhibits its good results when encountered with comparatively difficult geometric attack like rotation and cropping, scale and rotation etc.
[0058] Structural Similarity Index (SSIM) is a method to measure the similarity of two images, original and the extracted one. When the watermark is extracted, it is compared with the original one using it the SSIM index. SSIM is designed to improve peak signal-to- noise ratio (PSNR) and mean squared error. The PSNR, NCC (normalized correlation coefficient) and SSIM value of the extracted watermark under different attack are listed in Table 1. In this work, NCC values against almost all possible attacks are near about 95%, which validates the efficiency of the proposed algorithm. For non- compressed file format (.tiff ), we find a approximately full recovery of the embedded watermark with NCC = 0.99 - 1 and SSIM approximately 1 in all cases.
[0059] A graph is shown in Fig.7 describing different NCC values with rotation up to 180o. Table 4 illustrate the amount of cropping and the Corresponding NCC, SSIM and the PSNR quality of the stego. Results of other attacks are described in tabular form in Table 1.
[0060] In our design and implementation part we have followed the process step by step to achieve our desired goal. After designing the block level processor, we have implemented each module using VHDL code and finally incorporated all to provide the final processor model as single chip processor. The original image has been considered as several image blocks. Each image block, a form of binary bit stream is checked by a phase congruency detector whether it is less or greater than the predetermined standard value or not. If it is greater than the standard value the comparator output will be high else low. These values determine the featured image to be watermarked. We have considered a standard image value. In Fig. 9(a), the output of the entire image block has been reflected as testbench simulation result. We observe the output Y0 to Y7 for image rows image_in0 to image_in7. so, the entire picture has been mapped in a particular manner.
[0061] The Embedding algorithm, as discussed in Section III, has been performed on the featured image bits and we achieve the embedded image blocks. In Fig. 9(b), the simulation results show the results of the water marked image blocks w_mark0 to w_mark7. Fig. 9(c) shows the test bench simulation results for the decoder circuit of the watermark algorithm. The algorithm has been explained in Section III.
[0062] The proposed processor design has been implemented using Xilinx ISE 14.3 simulation tools and verified its functionality using FPGA boards which provide a comprehensive, high-performance development and demonstration platform. Very high speed FPGA are used for new generation module for advance electronic system and very much suitable for portable module. It is highly integrated and has high speed connectivity with superior bandwidth. We have successfully downloaded the synthesized processor module on this board and verified its functional activity as the alternative controller board. In the following Table 6, the FPGA implementation results can be observed.
[0063] After simulation, synthesis and hardware implementation process, the RTL schematic of the synthesizable modules are obtained, shown in Fig.10&11.
[0064] Fig. 11, we observe the RTL Schematic view of the processor incorporating the watermark embedding algorithm. The Decoder of the watermarked image has been designed using reverse algorithm. The RTL schematic view of the decoder circuit has been shown in Fig.12 using Xilinx 14.3 Plan ahead tools.
[0065] In the present work, we have developed and implemented the hardware of a highly secured and authenticated watermarking algorithm based on feature detection property of phase congruency and SVD to protect the copyright of digital contents. The scheme utilizes the lesser prominent image feature regions for hiding the copyright information by means of phase congruency and SVD scheme. The architecture was modeled using VHDL and synthesized using Virtex-7 technology. Functional simulation was performed using Xilinx ISE 14.3 simulator and the algorithm was implemented on high end FPGA device. The chip was tested using hardware co-simulation at the Maximum clock frequency of 351.457MHz with minimum period of 2.845 ns. Finally, the processor is presented as a synthesizable module. The experimental analysis discloses the robustness of the algorithm as it stands against various attacks like compression, geometrical distortion and common signal processing operations. Additionally, data hiding capacity and imperceptibility of the algorithm is also superior. Comparisons with some modern techniques recommend that the proposed idea can be effective tool for copyright protection and authentication.
[0066] Various modifications to these embodiments are apparent to those skilled in the art from the description and the accompanying drawings. The principles associated with the various embodiments described herein may be applied to other embodiments. Therefore, the description is not intended to be limited to the 5 embodiments shown along with the accompanying drawings but is to be providing the broadest scope consistent with the principles and the novel and inventive features disclosed or suggested herein. Accordingly, the invention is anticipated to hold on to all other such alternatives, modifications, and variations that fall within the scope of the present invention and appended claims. , Claims:CLAIMS
We Claim:
1) A method for embedding a secure digital watermark in a host image, the method comprising:
- Extracting feature points from the host image using a phase congruency algorithm that identifies significant image features independent of amplitude, based on phase congruency values;
- Mapping phase congruency values between 0 and 1 to identify significant feature points in the host image;
- Embedding a watermark image, represented as a matrix of singular values, in the host image using a Singular Value Decomposition (SVD)-based technique, such that the visual quality of the host image is preserved;
- Generating a watermarked image that conceals the watermark in feature points of the host image based on selected phase congruency values and singular value adjustments; and
- Securing embedding locations and watermark extraction parameters, including phase congruency level, singular value scaling factors, and watermark ordering, to prevent unauthorized extraction of the watermark.
2) The method as claimed in claim 1, wherein the watermark is a fingerprint image processed to extract unique features, such that the watermark is identifiable as unique to an individual or entity, enhancing authentication and copyright protection.
3) The method as claimed in claim 1, wherein the embedding further comprises:
- Performing SVD on both the host image matrix and the watermark image matrix, obtaining corresponding singular matrices.
- Adjusting the singular values of the host image matrix based on a scaling factor applied to the watermark matrix's singular values.
- Limiting adjusted singular values within a predefined grayscale range to prevent degradation of the watermarked image's visual quality.
4) The method as claimed in claim 1, wherein the extraction of the embedded watermark comprises:
- Retrieving feature points from the watermarked image using the same phase congruency values utilized in the embedding process.
- Reapplying SVD to these feature points and utilizing stored parameters to reconstruct the watermark image.
- Comparing the extracted watermark with the original watermark image using similarity indices such as SSIM, PSNR, and NCC to validate authenticity and watermark integrity.

Documents