Implementation method and system of high-quality L-shaped dot amplitude modulation screening digital watermark

Abstract

The invention discloses a high-quality L-shaped dot amplitude modulation screening digital watermark implementation method and system, and the method comprises the steps: inputting a carrier image color space, and outputting a monochromatic channel; splitting the halftone units to form a plurality of sub-dots, and calculating center coordinates of all the sub-dots; selecting and rejecting net points to obtain L-shaped net points; carrying out dot coding to obtain a halftone unit of an adjusted dot code; correcting and compensating to obtain a final adjusted and corrected halftone unit; screening to generate a final target image; the system comprises a file reading and preprocessing module, a halftone unit splitting module, a website selection and rejection module, an information coding module, a correction and compensation module and an amplitude modulation screening module. The method has the characteristics of high image definition, good hiding effect, flexible printing resolution and large watermark capacity.

Description

Technical Field

[0001] The present invention relates to the technical field of digital image watermarking, and in particular to a method and system for realizing high-quality L-type dot amplitude modulation screening digital watermarking. Background Art

[0002] Digital watermarking technology is a technology for hiding information in digital media. It embeds specific hidden information in digital media (such as images, audio, video, text, etc.). This information is imperceptible to the human visual or auditory system, but can be extracted by a specific decoder when needed. The core principle of digital watermarking technology is to embed hidden information into the carrier media without affecting the quality of the carrier media, so as to achieve functions such as authentication, copyright protection and traceability of digital content.

[0003] In the context of the rapid development of digital information and printing technology, books, illustrations, music, and advertisements have become important ways for people to obtain information and enjoy culture. However, with the development of network technology and electronic products, the access and copying of multimedia products have become extremely convenient, which has also led to a surge in pirated printing and infringement, causing huge losses to companies and individuals in related industries. In order to meet these challenges, digital watermark technology came into being. Product certification, copyright traceability and other property rights protection are increasingly valued, providing a powerful means to maintain the market and providing necessary protection and support for creators and companies.

[0004] In particular, the anti-printing and scanning digital watermarking technology for graphic printing has become one of the important directions in the field of digital watermarking research. However, the printing and scanning of digital products are subject to many limitations, such as the resolution of the printing equipment, the printing media used, the type of ink, etc., which may cause distortion or damage. Low-resolution printing equipment may cause the loss of image details and blurred edges. At the same time, the noise, color deviation or image distortion of the printed product will also affect the reliability and stability of the watermark. Certain physical effects in the printing process will also affect the performance of the watermark, which puts higher requirements on traditional watermarking technology.

[0005] The amplitude modulation watermarking technology based on phase modulation is an important development branch of graphic printing watermarking technology. Compared with traditional watermarking technology, the core idea of ​​this technology is to embed watermarks by fine-tuning the dot position of the halftone image, and use the interaction between the grating and the dots to form an effective pattern for watermark detection. Therefore, the use of amplitude modulation watermarking technology is not only affected by image quality, paper quality and printing equipment accuracy, but also limited by grating design.

[0006] The invention patent application with publication number CN117834791A proposed a latent image method of dual-channel amplitude modulation screening at different angles, which improved the image hiding effect to a certain extent. However, this method is not perfect, the physical grating is scattered, and the watermark extraction quality is poor. At the same time, the simple circular dots and fixed screening angles make the watermark area and the non-watermark area have a large visual difference, so it can still be understood as a visible watermark, which is difficult to achieve effective protection.

[0007] The difficulty in solving existing problems and defects lies in how to ensure the balance between the image quality of the target image and the concealment of the watermark image, so as to meet the technical requirements of high-quality printed image watermarks. At present, there is no good processing algorithm to reduce the visual difference between the boundary of the watermark area and the non-watermark area. At the same time, improving the embedding capacity of the watermark is also an important breakthrough in the technology. Summary of the invention

[0008] In order to overcome the deficiencies of the above-mentioned prior art, the purpose of the present invention is to provide a method and system for realizing high-quality L-dot amplitude modulation screening digital watermarking. The technology splits ordinary dots in amplitude modulation screening into a plurality of dots to form a dot group with a square dot cluster shape, and then discards some dots to obtain the remaining dots to form an "L"-shaped dot cluster shape, namely, L-dots, thereby improving the quality effect of embedded watermarks; analyzing the situation pattern generated by the combination of L-dots at the edge of the watermark, giving appropriate dot compensation, solving the overlap and missing problems generated in the critical area between the watermark area and the non-watermark area, effectively improving the concealment of the watermark, and achieving the goal of gradual smoothing effect under high visual clarity.

[0009] In order to achieve the above object, the technical solution adopted by the present invention is:

[0010] A method for implementing high-quality L-dot amplitude modulation screening digital watermarking includes the following steps:

[0011] Step 1, converting the carrier image color space: converting the original photo in the input RGB color space into a carrier image in the CMYK color space suitable for printing, and outputting a monochrome channel;

[0012] Step 2, splitting the halftone unit: generating an initial halftone unit according to the image resolution and screen line number of the carrier image obtained in step 1; further splitting the initial halftone unit into a plurality of sub-halftone units, and at the same time, splitting the dots on the halftone unit so that they are distributed in the corresponding sub-halftone units to form a plurality of sub-halftone dots, and calculating the center coordinates of all the sub-halftone dots;

[0013] Step 3, dot selection operation: discard some of the sub-dots obtained in step 2, so that the remaining dots form an "L" shape, thereby obtaining L-shaped dots;

[0014] Step 4, dot coding operation: according to the L-shaped dots obtained in step 3, determine the aggregation shape of each dot, calibrate the dots according to whether they contain watermark information, and adjust the coding of the dots accordingly according to the watermark information required to be included, so as to obtain the halftone unit of the adjusted dot coding;

[0015] Step 5, correction and compensation operation: Based on the adjusted halftone unit of the dot code obtained in step 4, according to the overlap or missing of the image edge and the watermark edge, the corresponding "black point problem" or "white point problem" correction and compensation strategy is adopted to obtain the final adjusted and corrected halftone unit;

[0016] Step 6, AM screening operation: Based on the final adjustment and the corrected halftone unit obtained in step 5, screen is added to generate the final target image.

[0017] The step 1 skips converting the carrier image color space by inputting a CMYK picture;

[0018] In the step 2 of splitting the halftone adjustment unit, a split coefficient M is introduced, and one halftone adjustment unit is split into halftone adjustment units of M×M sub-grid points. The screen pitch p in the x direction is divided into M parts, and the screen pitch q in the y direction is divided into M parts. The calculation formula for the coordinates corresponding to the sub-grid point center is formula (1.1), which is expressed as follows:

[0019]

[0020] Among them, P x is the x coordinate of the dot center mapped to the original image, P y It is the y coordinate of the dot center mapped to the original image, and p and q are the lengths of the horizontal side projection of the dot in the x and y directions respectively.

[0021] In the dot selection operation of step 3, the number of sub-dots in the L-type dot is dynamically generated according to the number of target image lines and the number of extracted grating lines;

[0022] In the dot selection operation, a selection coefficient m is introduced. There are multiple m mesh adjustment units under the same split coefficient M for dot selection operation, so that the remaining dots form different L-shaped dots and present different aggregation shapes.

[0023] In the dot coding operation of step 4, there are four coding modes for the actual coding of the remaining L-shaped dots: code 00, i.e. the gradient dots on the lower left are discarded; code 01, i.e. the gradient dots on the lower right are discarded; code 10, i.e. the gradient dots on the upper left are discarded; code 11, i.e. the gradient dots on the upper right are discarded.

[0024] The specific method of step 5 is:

[0025] During the correction and compensation operation, optimization is performed in three aspects: the halftone unit cut off due to the image edge, the color loss caused by the dot selection operation, and the edge overlap or loss caused by the special dot shape combination.

[0026] For the halftone unit cut off due to the image edge: recalibrate the halftone unit range of the screen to optimize the halftone unit cut off due to the image edge; the optimization formula is as follows:

[0027] Assuming that the original image length is w, the height is h, the x-axis projection length of the halftone unit is p, and the y-axis projection length of the halftone unit is q, the extended screening operation processing range is:

[0028]

[0029] For the color loss caused by dot selection: determine the dot compensation amount based on the grayscale value of the original image and the grayscale value of the image after the selection operation, and apply the compensation amount evenly to all the sub-dots of the L-type dot, so as to optimize the color loss caused by the selection operation after splitting the dot. The optimization formula is as follows:

[0030] Assume that the grayscale increment of the subnet point is d v The total overflow is d′, the cut-off coefficient is m, the number of subnets in the blank area is t, and the compensation formula is:

[0031]

[0032] For edge overlap or missing caused by special dot shape combination: According to the structural characteristics of L-type dot, this L-type dot encoding will produce white dots or black dots; the corresponding sub-dot compensation amount obtained according to different dot situation modes is evenly applied to all sub-dots of L-type dot; the optimization strategy is as follows:

[0033] Black dot problem: first determine the split coefficient M and the cut-off coefficient m of the gradient dot, then calculate the grayscale increment R on the boundary line and the number of sub-dots t that need to be corrected, and calculate the dot compensation amount λ accordingly, and finally traverse t sub-dots and modify their actual thresholds; the calculation formula (1.4) for the dot compensation amount λ for the black dot problem is:

[0034]

[0035] White dot problem: The calculation formula of the dot compensation λ for white dot problem is (1.5). The meaning of each coefficient is the same as that of the black dot problem, and the expression is:

[0036]

[0037] Among them, M is the splitting coefficient of the gradient dot, m is the cut-off coefficient, R is the grayscale increment on the boundary line, t is the number of sub-dots that need to be corrected, and λ is the dot compensation amount.

[0038] In the AM screening operation of step 6, AM screening is performed only on the sub-grid points where calibration exists.

[0039] In the AM screening operation of step 6, the sub-dot shapes include rectangular dots, circular dots, rhombus dots, star dots, diamond dots, wavy dots or cross dots.

[0040] The present invention provides a high-quality L-type dot amplitude modulation screening digital watermarking implementation system, comprising:

[0041] File reading and preprocessing module, used to read the target image and watermark image and perform preprocessing;

[0042] A halftone unit splitting module is used to generate an initial halftone unit according to the image resolution and screen line number of the acquired carrier image; further split the initial halftone unit into a plurality of sub-halftone units, and at the same time, split the dots on the halftone unit so that they are distributed in the corresponding sub-halftone units to form a plurality of sub-dots, and calculate the center coordinates of all the sub-dots;

[0043] The dot selection module is used to change the sub-dot selection after splitting into L-shaped dots, while ensuring that the overall screen count of the L-shaped watermark dots is the same as the dots before splitting;

[0044] An information encoding module, used for encoding the information of the watermark image into the target image;

[0045] Correction and compensation module, used to correct image edge distortion, image color distortion and overlap or missing of watermark edge area;

[0046] AM screening module for generating halftone target images with watermarks to obtain full-color prints for plate making.

[0047] The file reading and preprocessing module includes:

[0048] A target image reading module is used to select and read a target image from a computer file system;

[0049] A watermark image reading module is used to select and read a watermark image from a computer file system;

[0050] A CMYK generation module, used to convert an original image in RGB color space into a carrier image in CMYK color space using an ICC file;

[0051] A channel splitting module is used to split the carrier image into four monochrome channels;

[0052] The watermark image binarization module is used to convert the watermark image into a binary image.

[0053] The mesh adjustment unit splitting module includes:

[0054] Splitting parameter generation module: used to calculate the splitting coefficient M and sub-grid point coordinates p (x, y) according to the printing accuracy of the target image and the number of extracted raster lines;

[0055] Dot splitting module: used to split the original dot into M 2 sub-dots, and the dot group now appears in a rectangular shape.

[0056] The network point selection module includes:

[0057] Selection parameter generation module: used to calculate the selectable range of subnet points and obtain the selection coefficient m;

[0058] Network point selection operation module: used to select some network points in the network point group to make it an L-shaped network point group, and the number of network points formed by discarding some network points is m 2 A rectangular grid of dots.

[0059] The information encoding module comprises:

[0060] Parameter input module: select the embedding direction of watermark according to the watermark image;

[0061] Dot encoding module: encodes the L-shaped dots in the printing plate according to the horizontal watermark image or the vertical watermark image.

[0062] The correction and compensation module comprises:

[0063] Image edge correction module: used to expand the halftone unit and recalibrate the halftone unit range for screening;

[0064] Dot color compensation module: used to calculate the total grayscale before and after splitting dots, and compensate the grayscale value for the sub-dots;

[0065] Watermark edge analysis module: According to the dot code of each dot and the dot codes near it, analyze and obtain the situation mode of the dot combination;

[0066] Watermark edge compensation module: According to different situation modes, introduce dot compensation to correct the overlap or missing image structure between dots.

[0067] The amplitude modulation screening module comprises:

[0068] Threshold matrix generation module: generates the corresponding dot threshold matrix according to the set sub-dot shape;

[0069] Screen generation module: generates a halftone target image for printing based on the threshold matrix;

[0070] File storage module: used to store the halftone target image to the user's specified location;

[0071] Plate making and printing module: used for plate making and printing high-quality full-color prints with invisible watermarks.

[0072] Compared with the prior art, the present invention has the following beneficial effects:

[0073] 1. The present invention proposes an implementation method and system for L-type dot amplitude modulation screening digital watermarking, which improves the fineness of the target image by splitting the mesh modulation unit. The mesh modulation unit is split into several sub-mesh modulation units, thereby improving the fineness of the target image embedded with the watermark. At the same time, by selecting the dots in an "L" shape, the problem of hiding the amplitude modulation watermark with a low line count in a high-resolution image is effectively solved, and the visual effect of the watermarked image with low parallax is achieved, thereby ensuring the clarity of the target image.

[0074] 2. The present invention uses dot coding to hide a 2-bit information unit in a single dot, thereby realizing the existence of watermarks horizontally and vertically on a monochrome channel, and effectively improving the embedding capacity of the amplitude modulation watermark algorithm.

[0075] 3. The present invention is specially optimized for the image edge and the watermark edge, and adopts correction and compensation operations to solve the problems of color distortion, dot overlap and dot missing, so as to enhance the image watermark hiding effect and realize the generation of high-quality halftone target image.

[0076] In summary, the present invention realizes high-quality target images by splitting ordinary dots in AM screening into L-shaped dots composed of several sub-dots, ensures the embedding effect of watermark image, solves the drawback that AM watermark extraction depends on the limitation of designed grating, and achieves the effect of retaining high-resolution image details; expands the watermark capacity of AM watermark by dot coding; and further improves the embedding transparency of watermark by correcting and compensating for image edge problems and watermark edge problems, thereby ensuring the printing quality of target images. Therefore, the present invention has the characteristics of high image clarity, good hiding effect, flexible printing resolution, and large watermark capacity. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] Figure 1 It is a flow chart of the method of the present invention.

[0078] Figure 2 It is a system structure diagram of the present invention.

[0079] Figure 3 These are four codes of the L-shaped dots of the present invention, wherein: Figure 3 (a) is the schematic diagram of number 00, Figure 3 (b) is the schematic diagram of No. 01, Figure 3 (c) is the schematic diagram of number 10, Figure 3 (d) is the schematic diagram of number 11.

[0080] Figure 4 Schematic diagram of different cut-off coefficients of the L-type network point under the same splitting coefficient of the present invention, wherein: Figure 4 (a) is a schematic diagram of the splitting coefficient M being 3 and the cut-off coefficient m being 1. Figure 4 (b) is a schematic diagram of the splitting coefficient M being 3 and the cut-off coefficient m being 2.

[0081] Figure 5 is the target image generated by the L-type dot watermark algorithm of the present invention and the conventional amplitude modulation watermark algorithm, wherein, Figure 5 (a) is the overall effect diagram of the conventional algorithm. Figure 5 (b) is the local effect diagram of the conventional algorithm. Figure 5 (c) is the overall effect diagram of the L-type dot algorithm. Figure 5 (d) is the local effect diagram of the L-type algorithm.

[0082] Figure 6 It is a comparison diagram of the effects of the edge compensation method in the present invention.

[0083] Figure 7 is a comparison diagram of the effect of the color compensation method in the present invention, wherein: Figure 7 (a) is a schematic diagram of the target image for conventional processing. Figure 7 (b) is a schematic diagram of the L-type dot algorithm without color compensation. Figure 7 (c) is a schematic diagram after color compensation using the L-type dot algorithm.

[0084] Figure 8 are two watermark images extracted from the halftone target image generated by the L-type dot watermark in the present invention, wherein: Figure 8 (a) is the horizontal channel watermark information extraction diagram, Figure 8 (b) is the vertical channel watermark information extraction diagram.

[0085] Fig. 9 These are white dots and black dots generated by the combination of the L-type dots and another dot code in the present invention.

[0086] Fig.10 These are nine kinds of situation modes generated by the combination of the L-shaped dot of the present invention and the coding of the surrounding four dot. DETAILED DESCRIPTION

[0087] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments. However, the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0088] The present invention relates to a method and system for realizing high-quality L-type dot amplitude modulation screening digital watermarking, such as Figure 1 Shown is a flow chart of the method.

[0089] In this embodiment, two watermark images are embedded into a single channel of the target image, and the horizontal and vertical watermark images are extracted respectively by a grating extraction method, including the following steps:

[0090] Step 1, converting the carrier image color space;

[0091] Step 1.1, read the original image. If the original image is in RGB color space, it needs to be converted into a carrier image in CMYK color space; if the original image is in CMYK color space, it will not be processed. This embodiment reads an RGB image with a length of ω and a height of h, and performs color space conversion on the image. Among them, the RGB profile uses sRGB2014.icc, and the CMYK profile uses JapanColor2001Coated.icc;

[0092] Step 1.2, read two watermark images, embed them as horizontal watermark and vertical watermark respectively, transform the watermark images into grayscale images, and binarize the grayscale values ​​into 0 or 255, where the binarization threshold parameter is 128.

[0093] Step 2: Split the halftone adjustment unit:

[0094] Step 2.1, calculate the split coefficient M according to the input image resolution p and the target image screen number P. Considering that the split coefficient M is also related to the screen order C, the calculation formula of the split coefficient M is:

[0095]

[0096] Step 2.2, according to the split coefficient M, split all the halftone units in the target image into M×M sub-halftone units, and obtain M×M sub-halftone points;

[0097] Step 2.3, according to formula (1.1), calculate the center coordinates of all sub-network points.

[0098] The halftone unit splitting operation in step 2 may be replaced by a halftone dot splitting operation.

[0099] Step 3, point selection operation: From M 2 Select m from the subnets 2The sub-network points are discarded, and the remaining M points that form an "L" shape are taken. 2 -m 2 Sub-dots. Set the screening mark to obtain the basis of the L-type dot emphasized by the present invention. The split coefficient M and the cut-off coefficient m should satisfy: M>m>0.

[0100] Step 4, dot coding operation: In the L-type dot watermark algorithm, a single dot can embed 2 bits of hidden information, that is, each dot can exist in four different states: 00, 01, 10, and 11 due to different coding methods. Figure 3 In this embodiment, bit0 is used as the mark bit of the horizontal watermark, and bit1 is used as the mark bit of the vertical watermark. The actual encoding of all L-shaped dots in the target image can be set according to the grayscale of the pixels in the horizontal watermark image and the vertical watermark image.

[0101] Step 5, correction and compensation operations:

[0102] Step 5.1, recalibrate the halftone unit range in the target image, and expand the processing range of the screening operation according to formula (1.2);

[0103] Step 5.2, first traverse the total overflow of the grayscale increment of the sub-grid points in the gradient grid, and calculate the number of sub-grid tones in the blank area based on the split coefficient and the cut-off coefficient, and perform color compensation according to formula (1.3), such as Figure 7 As shown;

[0104] Step 5.3, calculate the center coordinates of all L-type dots in the target image, traverse all L-type dots, due to the structural characteristics of L-type dots, the dot will produce white dots or black dots when encoded with another L-type dot; the corresponding sub-dot compensation amount obtained according to different dot situation modes is evenly applied to all sub-dots of the L-type dot; the optimization strategy is as follows:

[0105] Black dot problem: The reason for the black dot problem is the overlap between adjacent gradient dots. It is precisely because of the overlap between gradient dots that the original intervals between gradient dots are covered, so a larger gradient dot is generated, which affects the overall effect of watermark information hiding. This strategy subtracts a value related to the current overlap amount from the actual threshold of all sub-dots at the mesh adjustment boundary line. The purpose of this strategy is to remove the grayscale increment caused by dot overlap and make the grayscale value at the boundary line as consistent as possible with the nearby area. The calculation formula is formula (1.4). First determine the split coefficient M and the cut-off coefficient m of the gradient dot, then calculate the grayscale increment R on the boundary line and the number of sub-dots t that need to be corrected, and calculate the dot compensation amount λ accordingly, and finally traverse t sub-dots and modify their actual thresholds.

[0106] White dot problem: In order to ensure that there are no white dots with missing structures in the target image after screening, it is necessary to add a part of sub-dots between two gradient dots. The present invention improves the threshold sampling method of compensation dots and adopts a compensation strategy that converts the white dot problem into the black dot problem. First, the sub-dot units in the blank area are marked as sub-dots, and the detailed texture features of the continuous tone image are restored as much as possible in the white dot area. After marking the dots, the white dot problem is converted into a black dot problem with dot overlap. The calculation formula is formula (1.5).

[0107] like Fig. 9 As shown, the coding of the surrounding four dots will produce 9 different modes, namely: black, white, black-black, white-white, one-dimensional black-white, two-dimensional black-white, black-black-white, white-white-black and black-black-white-white. Fig.10 According to the combination of the L-shaped dot code and four nearby codes, the L-shaped dot situation mode is marked.

[0108] Step 5.4, according to the dot situation mode, combined with formula (1.4) and formula (1.5), find the corresponding compensation strategy and compensation value. Traverse all L-shaped dots related to the edge and perform dot compensation operations to optimize the edge overlap or loss caused by special dot shape combinations. The effect is as follows Figure 6 shown.

[0109] Step 6, AM screening operation:

[0110] Step 6.1, according to the shape of the sub-dots, generate a suitable dot threshold matrix, use the conventional AM screening algorithm to perform AM screening on all the marked sub-dots of the target image, and generate a halftone target image, such as Figure 5 shown.

[0111] Step 6.2, storing the halftone target image in a designated folder, inputting it into a printing device, realizing final plate printing, and obtaining a high-quality full-color print with watermark.

[0112] After this, there are two different watermark images in the printed matter. The hidden watermark information can be extracted using common detection methods, such as machine scanning, optical reflection, etc. This embodiment uses grating extraction. The implementation principle is to use the light splitting effect of the grating sheet to detect the hidden information. The extraction effect is as follows: Figure 8 shown.

[0113] The present invention relates to a high-quality L-type dot amplitude modulation screening digital watermarking implementation system, such as Figure 2 The structure diagram of this system is shown in FIG.

[0114] The system includes: a file reading and preprocessing module M201, a screen adjustment unit splitting module M202, a dot selection module M203, an information encoding module M204, a correction and compensation module M205, and an amplitude modulation screening module M206.

[0115] The file reading and preprocessing module M201 is mainly responsible for reading the target image, watermark image and preprocessing to facilitate the operation of subsequent modules.

[0116] The file reading and preprocessing module M201 specifically includes: a target image reading module M2011, used to select and read a target image from a computer file system; a watermark image reading module M2012, used to select and read a watermark image from a computer file system; a CMYK generation module M2013, used to use an ICC file to convert an original image in an RGB color space into a carrier image in a CMYK color space; a channel splitting module M2014, used to split the carrier image into four monochrome channel images of C, M, Y, and K; a watermark image binarization module M2015, used to grayscale the watermark image and then perform a threshold comparison to make it a binary grayscale image with a grayscale value of only 0 or 255.

[0117] The halftone unit splitting module M202 is used to generate an initial halftone unit according to the image resolution and screen line number of the acquired carrier image; further split the initial halftone unit into several sub-halftone units, and at the same time, split the dots on the halftone unit so that they are distributed in the corresponding sub-halftone units to form several sub-dots, and calculate the center coordinates of all sub-dots; in this way, the fineness of the target image can be improved, and at the same time, the composition premise for the L-type dots is created.

[0118] The halftone unit splitting module M202 specifically includes: a splitting parameter generating module M2021, which is used to calculate the generated screen splitting coefficient N and the sub-grid point coordinates p (x, y) according to the printing accuracy of the target image and the number of extracted raster lines; a dot splitting module M2022, which is used to split the original dot into M 2 sub-dots, and the dot group now appears in a rectangular shape.

[0119] The dot selection module M203 is used to change the sub-dot selection after the split into L-shaped dots, while ensuring that the overall screen number of the L-shaped watermark dots is the same as the dots before the split;

[0120] The dot selection module M203 specifically includes: a selection parameter generation module M2031, which is used to calculate the range of the selection coefficient m and can actively select a suitable value of the selection coefficient m. It should be understood that the selection coefficient m is an important parameter in the implementation of the L-type dot watermark algorithm, such as Figure 4As shown in the figure. Under the same splitting coefficient, the increase of the cut-off coefficient m will reduce the number of sub-grid points, thereby affecting the performance of the L-type grid points. At the same time, the influence of the cut-off coefficient m on the watermark quality should be avoided. Therefore, it should be flexibly adjusted according to the splitting coefficient and application scenario; the grid point selection operation module M2032 is used to discard the selected sub-grid points so that the remaining sub-grid points form an "L" shape, that is, an L-type grid point.

[0121] The information encoding module M204 is used to process the watermark image information and encode it into the target image.

[0122] The information encoding module M204 specifically includes: a parameter input module M2041, used to select the embedding direction of the watermark image according to the watermark image; a dot encoding module M2042, used to determine whether the watermark image is horizontal or vertical, and then encode the L-shaped dots in the printing plate.

[0123] The correction and compensation module M205 is used to correct image edge distortion, image color distortion and overlap or missing problems of watermark edge areas;

[0124] The correction and compensation module M205 specifically includes: an image edge correction module M2051, which is used to expand the halftone unit and recalibrate the range of the halftone unit for adding screens, thereby eliminating the missing problem caused by the missing sub-dots at the edge of the image; a dot color compensation module M2052, which evenly distributes the extra grayscale values ​​in the halftone unit to all sub-dots of the gradient dot, while maintaining the increase in the absolute grayscale of the sub-dots and reducing the distortion of the relative grayscale of the sub-dots, and obtaining a more natural color compensation result. A watermark edge analysis module M2053, which analyzes and obtains the situation mode of the dot combination based on the dot code of each dot and the dot code of its neighboring dot; a watermark edge compensation module M2054, which introduces dot compensation according to different situation modes to correct the overlap or missing image structure between dots.

[0125] The AM screening module M206 is used to generate a half-tone target image with a watermark and print high-quality full-color prints.

[0126] The amplitude modulation screening module M206 specifically includes: a threshold matrix generation module M2061, which is used to generate a suitable dot threshold matrix according to the "L"-shaped sub-dot shape; a screening generation module M2062, which is used to generate a halftone target image for printing according to the threshold matrix; a file storage module M2063, which is used to store the halftone target image in the computer file system; and a platemaking and printing module M2064, which is used to print high-quality full-color prints containing invisible watermarks.

[0127] The present invention splits the dots in the amplitude modulation screened watermark into several sub-dots to form a dot group with a square dot cluster shape, then discards some dots, and obtains the remaining dots to form a dot group with an "L"-shaped dot cluster shape, namely, L-shaped dots, which improves the clarity of the target image and eliminates the problem of the number of lines limited in the grating design; analyzes the situation pattern generated by the combination of L-shaped dots on the edge of the watermark, gives appropriate dot compensation, solves the overlap and missing problems generated in the critical area of ​​the watermark area and the non-watermark area, effectively improves the concealment of the watermark, and reduces the visual difference between the two areas under high-quality target images; through the dot coding method, two watermark information can be accommodated at the same time, effectively improving the embedding capacity of the watermark algorithm. The system includes: a file reading and preprocessing module, a halftone unit splitting module, a dot selection module, an information encoding module, a correction and compensation module, and an amplitude modulation screen module. The method and system for realizing high-quality L-dot amplitude modulation screening digital watermark image proposed in the present invention realize a clearer halftone target image under the same number of extracted grating lines, reduce watermark traces, and improve the embedding capacity of watermark, and have the characteristics of high image clarity, good hiding effect, flexible printing resolution and large watermark capacity.

[0128] In order to prove the effectiveness of the present invention, the present invention designs corresponding subjective and objective evaluation experiments:

[0129] In the analysis of the image clarity effect, the L-type dot watermark algorithm is compared with the conventional amplitude modulation screening watermark algorithm on a public image data set to illustrate the effectiveness of the present invention from both subjective and objective aspects.

[0130] First, the present invention uses the photos in the DIV2K_train_HR high-resolution reconstruction dataset as the dataset of this experiment, and uses the conventional AM screening watermarking algorithm and the L-type dot watermarking algorithm for AM screening, and generates halftone target images based on the same number of extracted raster lines. Taking "0149.png" in the dataset as an example, different halftone target images are obtained, such as Figure 5 As shown in the figure, it can be seen that there is a big gap between the AM screening watermarking algorithm using conventional dots and the present invention. The conventional algorithm is difficult to achieve satisfactory results in detail processing, especially in the relief and spire where obvious loss occurs. In contrast, the L-dot algorithm can still retain more details and significantly improve the image clarity.

[0131] The present invention uses SSIM as an objective evaluation index to evaluate the conventional amplitude modulation screening watermarking algorithm and the L-type dot watermarking algorithm. SSIM is a measurement method for measuring the similarity of the structural information of two images. This index takes into account the difference in the perception of the image structure by the human visual system. At the same time, the index includes multiple aspects such as brightness, structure and contrast. Therefore, the SSIM evaluation index is better than PSNR. The present invention selects SSIM to objectively evaluate the overall image quality of the halftone target image. Its calculation formula is:

[0132]

[0133] In formula (1.9), μ x represents the mean value of x, σ x represents the standard deviation of x, σ xy Represents the covariance of x and y. C1 and C2 are positive constants to prevent the structural similarity result from being 0.

[0134] The halftone target images obtained by calculating and comparing the conventional AM screening watermarking algorithm and the L-type dot watermarking algorithm are shown in Table 1, where the best result of each image is marked in bold.

[0135] Table 2. SSIM quality evaluation of halftone target images.

[0136]

[0137]

[0138] According to Table 1, the following conclusions can be drawn: Compared with the conventional AM screening watermarking algorithm, the halftone target image obtained by the algorithm proposed in the present invention has better structural similarity, which means that the L-type dot watermarking algorithm can obtain a halftone target image with higher definition. Even in the case of more image details or complex image textures, better image quality can still be maintained.

[0139] Next, we objectively evaluate the embedded watermark traces. The present invention selects SSIM as the objective evaluation standard for watermark traces. At this time, the calculation area is reduced to the rectangular area containing the watermark, which improves the accuracy of the results and reduces the amount of calculation. The results are shown in Table 2, where the best results for each image are marked in bold:

[0140] Table 2 SSIM evaluation of watermark trace area

[0141]

[0142] According to Table 2, the following conclusions can be drawn: Compared with the conventional AM screening watermarking algorithm, the watermark trace area obtained by the algorithm proposed in the present invention has better structural similarity, improves the watermark hiding effect, and achieves a balance between image clarity and watermark invisibility.

[0143] The above process includes the entire process of file reading and preprocessing, dot splitting, dot selection, information encoding, correction and compensation, and amplitude modulation screening.

[0144] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When the use is implemented in whole or in part in the form of a computer program product, the computer program product includes one or more computer instructions. When the computer program instructions are loaded or executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL) or wireless (e.g., infrared, wireless, microwave, etc.) mode) to another website site, computer, server or data center. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state hard disk SolidState Disk (SSD)), etc.

[0145] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any modifications, equivalent substitutions and improvements made by any technician familiar with the technical field within the technical scope disclosed by the present invention and within the spirit and principle of the present invention should be covered by the protection scope of the present invention.

Claims

1. A method for realizing high-quality L-dot amplitude modulation screening digital watermarking, characterized in that: The following steps are involved: Step 1, converting the carrier image color space: converting the original photo in the input RGB color space into a carrier image in the CMYK color space suitable for printing, and outputting a monochrome channel; Step 2, splitting the halftone unit: generating an initial halftone unit according to the image resolution and screen line number of the carrier image obtained in step 1; further splitting the initial halftone unit into a plurality of sub-halftone units, and at the same time, splitting the dots on the halftone unit so that they are distributed in the corresponding sub-halftone units to form a plurality of sub-halftone dots, and calculating the center coordinates of all the sub-halftone dots; Step 3, dot selection operation: discard some of the sub-dots obtained in step 2, so that the remaining dots form an "L" shape, thereby obtaining L-shaped dots; Step 4, dot coding operation: according to the L-shaped dots obtained in step 3, determine the aggregation shape of each dot, calibrate the dots according to whether they contain watermark information, and adjust the coding of the dots accordingly according to the watermark information required to be included, so as to obtain the halftone unit of the adjusted dot coding; Step 5, correction and compensation operation: Based on the adjusted halftone unit of the dot code obtained in step 4, according to the overlap or missing of the image edge and the watermark edge, the corresponding "black point problem" or "white point problem" correction and compensation strategy is adopted to obtain the final adjusted and corrected halftone unit; Step 6, AM screening operation: Based on the final adjustment and the corrected halftone unit obtained in step 5, screen is added to generate the final target image.

2. The method for realizing high-quality L-type dot amplitude modulation screening digital watermark according to claim 1, characterized in that: The step 1 skips converting the carrier image color space by inputting a CMYK picture; In the step 2 of splitting the halftone adjustment unit, a split coefficient M is introduced, and one halftone adjustment unit is split into halftone adjustment units of M×M sub-grid points. The grid line spacing p in the x direction is divided into M parts, and the grid line spacing q in the g direction is divided into M parts. The calculation formula for the coordinates corresponding to the sub-grid point center is formula (1.1), which is expressed as follows: Among them, P x is the x coordinate of the dot center mapped to the original image, P y It is the y coordinate of the dot center mapped to the original image, and p and q are the lengths of the horizontal side projection of the dot in the x and y directions respectively.

3. The method for realizing high-quality L-dot AM screening digital watermarking according to claim 1, characterized in that: In the dot selection operation of step 3, the number of sub-dots in the L-type dot is dynamically generated according to the number of target image lines and the number of extracted grating lines; In the dot selection operation, a selection coefficient m is introduced. There are multiple m mesh adjustment units under the same split coefficient M for dot selection operation, so that the remaining dots form different L-shaped dots and present different aggregation shapes.

4. The method for realizing high-quality L-dot AM screening digital watermarking according to claim 1, characterized in that: In the dot coding operation of step 4, there are four coding modes for the actual coding of the remaining L-shaped dots: code 00, i.e. the gradient dots on the lower left are discarded; code 01, i.e. the gradient dots on the lower right are discarded; code 10, i.e. the gradient dots on the upper left are discarded; code 11, i.e. the gradient dots on the upper right are discarded.

5. The method for realizing high-quality L-dot AM screening digital watermarking according to claim 1, characterized in that: The specific method of step 5 is: During the correction and compensation operation, three aspects are optimized: color loss caused by cutting off the halftone unit due to the image edge, color loss caused by the dot selection operation, and edge overlap or loss caused by special dot shape combination; For the halftone unit cut off due to the image edge: recalibrate the halftone unit range of the screen to optimize the halftone unit cut off due to the image edge; the optimization formula is as follows: Assuming that the original image length is w, the height is h, the x-axis projection length of the halftone unit is p, and the y-axis projection length of the halftone unit is q, the extended screening operation processing range is: For the color loss caused by the dot selection operation: determine the dot compensation amount according to the gray value of the original image and the gray value of the image after the selection operation, and apply the compensation amount evenly to the sub-dots of all L-type dots, so as to optimize the color loss caused by the selection operation after splitting the dots; the optimization formula is as follows: Assume that the grayscale increment of the subnet point is d v The total overflow amount is d′, the cut-off coefficient is m, the number of subnets in the blank area is t, and the compensation formula is: For edge overlap or missing caused by special dot shape combination: According to the structural characteristics of L-type dot, this L-type dot encoding will produce white dots or black dots; the corresponding sub-dot compensation amount obtained according to different dot situation modes is evenly applied to all sub-dots of L-type dot; the optimization strategy is as follows: Black dot problem: first determine the split coefficient M and the cut-off coefficient m of the gradient dot, then calculate the grayscale increment on the boundary line and the number of sub-dots t that need to be corrected, and calculate the dot compensation amount λ accordingly, and finally traverse t sub-dots and modify their actual thresholds; the calculation formula (1.4) for the dot compensation amount λ for the black dot problem is: White dot problem: The calculation formula of the dot compensation λ for white dot problem is (1.5). The meaning of each coefficient is the same as that of the black dot problem, and the expression is: Among them, M is the splitting coefficient of the gradient dot, m is the cut-off coefficient, R is the grayscale increment on the boundary line, t is the number of sub-dots that need to be corrected, and λ is the dot compensation amount.

6. The method for realizing high-quality L-dot AM screening digital watermarking according to claim 1, characterized in that: In the AM screening operation of step 6, AM screening is performed only on the sub-network points with calibration; The sub-dot shapes include rectangular dots, circular dots, rhombus dots, star dots, diamond dots, wavy line dots or cross dot shapes.

7. A high-quality L-dot AM screening digital watermarking implementation system, characterized in that: include: File reading and preprocessing module, used to read the target image and watermark image and perform preprocessing; A halftone unit splitting module is used to generate an initial halftone unit according to the image resolution and screen line number of the acquired carrier image; further split the initial halftone unit into a plurality of sub-halftone units, and at the same time, split the dots on the halftone unit so that they are distributed in the corresponding sub-halftone units to form a plurality of sub-dots, and calculate the center coordinates of all the sub-dots; The dot selection module is used to change the sub-dot selection after splitting into L-shaped dots, while ensuring that the overall screen count of the L-shaped watermark dots is the same as the dots before splitting; An information encoding module, used for encoding the information of the watermark image into the target image; Correction and compensation module, used to correct image edge distortion, image color distortion and overlap or missing of watermark edge area; AM screening module for generating halftone target images with watermarks to obtain full-color prints for plate making.

8. A high-quality L-dot AM screening digital watermarking implementation system according to claim 7, characterized in that: The file reading and preprocessing module includes: A target image reading module is used to select and read a target image from a computer file system; A watermark image reading module is used to select and read a watermark image from a computer file system; A CMYK generation module, used to convert an original image in RGB color space into a carrier image in CMYK color space using an ICC file; A channel splitting module is used to split the carrier image into four monochrome channels; The watermark image binarization module is used to convert the watermark image into a binary image.

9. A high-quality L-dot AM screening digital watermarking implementation system according to claim 7, characterized in that: The mesh adjustment unit splitting module includes: Splitting parameter generation module: used to calculate and generate dot splitting coefficient M and sub-dot coordinates p (x, y) according to the printing accuracy of the target image and the number of extracted raster lines; Dot splitting module: used to split the original dot into M 2 Sub-dots, at this time the dot group presents a rectangular shape; The network point selection module includes: Selection parameter generation module: used to calculate the selectable range of subnet points and obtain the selection coefficient m; Network point selection operation module: used to select some network points in the network point group to make it an L-shaped network point group, and the number of network points formed by discarding some network points is m 2 A rectangular grid of dots; The information encoding module comprises: Parameter input module: select the embedding direction of watermark according to the watermark image; Dot encoding module: encodes the L-shaped dots in the printing plate according to the horizontal watermark image or the vertical watermark image.

10. A high-quality L-dot AM screening digital watermarking implementation system according to claim 7, characterized in that: The correction and compensation module comprises: Image edge correction module: used to expand the halftone unit and recalibrate the halftone unit range for screening; Dot color compensation module: used to calculate the total grayscale before and after splitting dots, and compensate the grayscale value for the sub-dots; Watermark edge analysis module: According to the dot code of each dot and the dot codes near it, analyze and obtain the situation mode of the dot combination; Watermark edge compensation module: according to different situation modes, introduce dot compensation to correct the overlap or missing image structure between dots; The amplitude modulation screening module comprises: Threshold matrix generation module: generates the corresponding dot threshold matrix according to the set sub-dot shape; Screen generation module: generates a halftone target image for printing based on the threshold matrix; File storage module: used to store the halftone target image to the user's specified location; Plate making and printing module: used for plate making and printing high-quality full-color prints with invisible watermarks.

Images