CN119155406A

CN119155406A - Method and system for realizing high-quality L-shaped lattice point amplitude modulation screening digital watermark

Info

Publication number: CN119155406A
Application number: CN202411211461.3A
Authority: CN
Inventors: 李广鑫; 周杨; 肖健伟; 张一帆; 李卓贤; 王逸; 王豪; 唐静宇; 席伟星
Original assignee: Xidian University
Current assignee: Xidian University
Priority date: 2024-08-30
Filing date: 2024-08-30
Publication date: 2024-12-17

Abstract

A method for realizing high-quality L-shaped halftone amplitude modulation screening digital watermark includes inputting carrier image color space, outputting monochromatic channel, splitting halftone units to form several sub-halftone and calculating central coordinates of all sub-halftone, halftone selection and separation to obtain L-shaped halftone, halftone coding to obtain halftone units of regulated halftone coding, correction and compensation to obtain final regulated and corrected halftone units, screening to generate final target image, and system including file reading and preprocessing module, halftone unit splitting module, halftone selection and separation module, information coding module, correction and compensation module and amplitude modulation screening module.

Description

Method and system for realizing high-quality L-shaped lattice point amplitude modulation screening digital watermark

Technical Field

The invention relates to the technical field of digital image watermarking, in particular to a method and a system for realizing high-quality L-shaped halftone amplitude modulation screening digital watermarking.

Background

Digital watermarking is a technique of hiding information in digital media by embedding specific hidden information in digital media (e.g. images, audio, video, text, etc.), which is not perceptible to the human visual or auditory system, but can be extracted by a specific decoder if necessary. The core principle of the digital watermarking technology is that hidden information is embedded into carrier media on the premise of not affecting the quality of the carrier media, so that the functions of authentication, copyright protection, tracing and the like of digital content are realized.

In the current background of rapid development of digital information and printing technology, products such as books, drawings, music and advertisements have become an important way for people to acquire information and enjoy culture. However, with the development of network technology and electronic products, access and copying of multimedia products become extremely convenient, which also causes a proliferation of piracy and infringement, and brings about huge losses to enterprises and individuals in related industries. To address these challenges, digital watermarking techniques have evolved. The property protection such as product authentication, copyright traceability and the like is increasingly paid attention to, a powerful means is provided for maintaining markets, and necessary protection and support are provided for creators and enterprises.

Particularly, the anti-printing scanning digital watermarking technology for graphic printing has become one of the important directions in the field of digital watermarking research. However, the print scanning of digital products is subject to limitations such as resolution of the printing device, the print medium used, the type of ink, etc., which may result in distortion or damage. Low resolution printing devices may result in loss of image detail and blurring of edges. Meanwhile, noise, color deviation or image distortion of printed matters can also affect the reliability and stability of the watermark. Certain physical effects in the printing process can also affect the performance of the watermark, and higher requirements are placed on the traditional watermarking technology.

Amplitude modulation watermarking technology based on phase modulation method is an important development branch of graphic printing watermarking technology. Compared with the traditional watermarking technology, the key idea of the technology is to embed the watermark by fine-tuning the halftone dot position of the halftone image, and to use the interaction between the grating and the halftone dot to form an effective pattern for watermark detection. Thus, the use of amplitude modulated watermarking techniques is not only affected by image quality, paper quality and printing device accuracy, but is also limited by the raster design.

The invention patent application with publication number of CN117834791A provides a double-channel different-angle amplitude modulation screening latent image processing method, which improves the hiding effect of the image to a certain extent, but the method is imperfect, and the physical grating is scattered, and the watermark extraction quality is poor. Meanwhile, the simple circular dots and the fixed screening angle lead the watermark area and the non-watermark area to have larger visual difference, so the watermark can be still understood as a visual watermark, and effective protection is difficult to realize.

The difficulty in solving the existing problems and defects is how to ensure the balance between the image quality of the target image and the concealment of the watermark image, thereby achieving the technical requirement of high-quality printing image watermark. At present, no better processing algorithm exists, so that the boundaries of the watermark region and the non-watermark region have less overlarge visual difference, and the improvement of the embedding capacity of the watermark is an important breakthrough point of the technology.

Disclosure of Invention

In order to overcome the defects of the prior art, the invention aims to provide a realization method and a system of high-quality L-shaped mesh point amplitude modulation screening digital watermark, which are characterized in that common mesh points in amplitude modulation screening are split into a plurality of mesh point groups with square mesh point aggregation shapes, then partial mesh points are removed, the rest mesh point groups with L-shaped mesh point aggregation shapes are obtained, namely L-shaped mesh points, and the quality effect of watermark embedding is further improved; the situation mode generated by the combination of the L-shaped dots at the watermark edge is analyzed, proper dot compensation is given, the lap joint and the missing problem generated by the critical areas of the watermark area and the non-watermark area are solved, the concealment of the watermark is effectively improved, and the aim of gradual change smoothing effect under the condition of high visual definition is fulfilled.

In order to achieve the above purpose, the technical scheme adopted by the invention is as follows:

a realization method of high-quality L-shaped net point amplitude modulation screening digital watermark comprises the following steps:

Converting the carrier image color space, namely converting the original photo of the input RGB color space into a carrier image of CMYK color space suitable for printing, and outputting a monochromatic channel;

Step 2, splitting the mesh tone unit, namely generating an initial mesh tone unit according to the image resolution and the net adding number of the carrier image obtained in the step 1, further splitting the initial mesh tone unit into a plurality of sub-mesh tone units, splitting the mesh points on the mesh tone unit, distributing the mesh points in the corresponding sub-mesh tone units to form a plurality of sub-mesh points, and calculating the central coordinates of all the sub-mesh points;

step 3, dot cut-off operation, namely, discarding part of the sub-dots obtained in the step 2 to enable the rest dots to form an L-shaped form, so as to obtain L-shaped dots;

step 4, dot coding operation, namely judging the aggregation shape of each dot according to the L-shaped dots obtained in the step 3, calibrating the dots according to whether watermark information is contained or not, and correspondingly adjusting the encoding of the dots according to the watermark information required to be contained so as to obtain an adjusted dot encoded mesh tone unit;

Step 5, correcting and compensating operation, namely based on the mesh tone unit of the adjusted mesh point code obtained in the step 4, adopting a corresponding correction and compensation strategy of 'black point problem' or 'white point problem' according to the lap joint or missing condition of the image edge and the watermark edge to obtain a mesh tone unit after final adjustment and correction;

And 6, amplitude modulation screening operation, namely screening to generate a final target image based on the final adjusted and corrected mesh tone unit obtained in the step 5.

Step 1, skipping conversion carrier image color space by inputting CMYK pictures;

In the step 2, a splitting coefficient M is introduced into the splitting mesh adjusting units, 1 mesh adjusting unit is split into mesh adjusting units of m×m sub-dots, the mesh line pitch p in the x direction is split into M parts, the mesh line pitch q in the y direction is split into M parts, the calculation formula of the coordinates corresponding to the centers of the sub-dots is formula (1.1), and the expression is:

Wherein, P _x is the x coordinate of the dot center mapped to the original image, P _y is the y coordinate of the dot center mapped to the original image, and P and q are the lengths of the dot lateral side projections in the x and y directions respectively.

In the step 3 dot cut-off operation, the number of sub dots in the L-shaped dots is dynamically generated according to the number of target image lines and the number of extracted grating lines;

In the lattice point cut-off operation, a cut-off coefficient M is introduced, and a plurality of M exist in a mesh tone unit under the same splitting coefficient M to perform the cut-off lattice point operation, so that the rest lattice points form different L-shaped lattice points and present different aggregation shapes.

In the step 4 dot coding operation, the actual coding of the L-shaped dots left by cut-off has 4 coding modes, namely coding 00, namely dropping the left bottom of the gradual change dot, coding 01, namely dropping the right bottom of the gradual change dot, coding 10, namely dropping the left top of the gradual change dot, and coding 11, namely dropping the right top of the gradual change dot.

The specific method in the step 5 is as follows:

in the correction and compensation operation, the method is optimized for three aspects, namely, the method aims at color deficiency caused by cutting off a mesh tone unit at the edge of an image, color deficiency caused by lattice point cut-off operation and edge lap joint or deficiency caused by special lattice point shape combination.

Aiming at the mesh unit cut off due to the image edge, recalibrating the range of the mesh unit for screening so as to optimize the mesh unit cut off due to the image edge, wherein the optimization formula is as follows:

assuming that the length of an original image is w, the height is h, the projection length of the transverse side x axis of the mesh adjusting unit is p, the projection length of the transverse side y axis of the mesh adjusting unit is q, and the expanded screening operation processing range is as follows:

and aiming at the color deficiency caused by the lattice point cut-off operation, determining lattice point compensation quantity according to the gray value of the original image and the gray value of the image after the cut-off operation, and uniformly applying the compensation quantity to the sub lattice points of all the L-shaped lattice points, thereby optimizing the color deficiency caused by the lattice point cut-off operation after the lattice point is split. The optimization formula is as follows:

Assuming that the overflow total d' of the sub-dot gray level increment d _v is m, the cut-off coefficient is m, the sub-dot tone quantity in the blank area is t, and the compensation formula is as follows:

Aiming at edge lap joint or deficiency caused by special dot shape combination, the method can generate white dots or black dots with the other L-shaped dot code according to the structural characteristics of the L-shaped dots, the compensation quantity of the corresponding sub-dots obtained according to different dot situation modes is uniformly applied to all sub-dots of the L-shaped dots, and the optimization strategy is as follows:

The black point problem is that a splitting coefficient M and a cut-off coefficient M of a gradual change lattice point are firstly determined, then a gray scale increment R and the number t of sub lattice points needing correction on a boundary line are calculated, a lattice point compensation quantity lambda is calculated according to the gray scale increment R and the number t of the sub lattice points, and finally the t sub lattice points are traversed and the actual threshold value is modified, wherein a calculation formula (1.4) of the lattice point compensation quantity lambda of the black point problem is as follows:

The white point problem is that the calculation formula of the white point problem dot compensation quantity lambda is (1.5), the meaning of each coefficient is the same as that of the black point problem, and the expression is:

wherein M is a splitting coefficient of the gradual change mesh point, M is a taking-off coefficient, R is a gray scale increment on a boundary line, t is the number of sub mesh points to be corrected, and lambda is mesh point compensation quantity.

In the amplitude modulation screening operation in the step 6, amplitude modulation screening is carried out on only the sub-net points with the calibration.

In the amplitude modulation screening operation in the step 6, the shape of the sub-mesh point comprises rectangular mesh points, round mesh points, diamond mesh points, star-shaped mesh points, diamond mesh points, wave line mesh points or cross mesh points.

The invention provides a realization system of high-quality L-shaped net point amplitude modulation screening digital watermark, which comprises:

The file reading and preprocessing module is used for reading the target image and the watermark image and preprocessing the target image and the watermark image;

the system comprises a mesh tone unit splitting module, a mesh tone unit calculating module and a mesh tone unit calculating module, wherein the mesh tone unit splitting module is used for generating an initial mesh tone unit according to the image resolution and the net adding number of an acquired carrier image, further splitting the initial mesh tone unit into a plurality of sub-mesh tone units, splitting the mesh points on the mesh tone unit to enable the mesh points to be distributed in the corresponding sub-mesh tone units, forming a plurality of sub-mesh points, and calculating the center coordinates of all the sub-mesh points;

The net point cut-off module is used for changing the cut-off of the sub net points after the splitting into L-shaped net points, and simultaneously ensuring that the whole net adding line number of the L-shaped watermark net points is the same as that of the net points before the splitting;

An information encoding module for encoding information of the watermark image to the target image;

the correction and compensation module is used for correcting the problems of image edge distortion, image color distortion and overlap joint or missing of watermark edge areas;

and the amplitude modulation screening module is used for generating a halftone target image containing the watermark and obtaining a full-color printed matter of plate making printing.

The file reading and preprocessing module comprises:

the target image reading module is used for selecting and reading target images from the computer file system;

the watermark image reading module is used for selectively reading the watermark image from the computer file system;

A CMYK generation module for converting an original image of an RGB color space into a carrier image of a CMYK color space using an ICC file;

the channel splitting module is used for splitting the carrier image into four monochromatic channels;

and the watermark image binarization module is used for converting the watermark image into a binary image.

The mesh adjustment unit splitting module comprises:

The splitting parameter generating module is used for calculating a splitting coefficient M and sub-dot coordinates p (x, y) according to the printing precision of the target image and the number of extracted grating lines;

The net point splitting module is used for splitting the original net point into M ² sub net points, and the net point group presents a rectangular shape at the moment.

The lattice point cut-off module comprises:

the choice parameter generation module is used for calculating the choice range of the sub-net points and obtaining a choice coefficient m;

and the lattice point cut-off operation module is used for cutting off part of lattice points in the lattice point group to form an L-shaped lattice point group, and the lattice points cut off part of the lattice points are rectangular lattice point groups with the number of the constituent lattice points of m ².

The information encoding module includes:

the parameter input module is used for selecting an embedding direction of the embedded watermark according to the watermark image;

and the dot coding module codes the L-shaped dots in the printing plate according to the horizontal watermark image or the longitudinal watermark image.

The correction and compensation module includes:

The image edge correction module is used for expanding the mesh tone unit and recalibrating the range of the mesh tone unit with the net;

the screen dot color compensation module is used for calculating the gray scale total before and after splitting the screen dots and compensating gray scale values for the sub screen dots;

the watermark edge analysis module is used for analyzing and obtaining the situation mode of the dot combination according to the dot code of each dot and the dot codes nearby the dot code;

And the watermark edge compensation module is used for introducing dot compensation quantity according to different situation modes and correcting overlapping or missing image structures generated between the dots.

The amplitude modulation screening module comprises:

the threshold matrix generation module is used for generating a corresponding dot threshold matrix according to the set shape of the sub-dots;

a screening generation module for generating a halftone target image for printing according to the threshold matrix;

a file storage module for storing the halftone target image to a designated position of a user;

and the platemaking and printing module is used for platemaking and printing high-quality full-color printed matters containing invisible watermarks.

Compared with the prior art, the invention has the following beneficial effects:

1. the invention provides a realization method and a system of L-shaped halftone amplitude modulation screening digital watermark, which improve the fineness of a target image through a mode of splitting a halftone unit, split a plurality of sub-halftone units through the halftone unit, improve the fineness of the target image embedded with the watermark, and effectively solve the problem of hiding low-line amplitude modulation watermark in a high-resolution image through halftone cut-off into an L-shaped shape, realize the visual effect of the watermark-containing image with low parallax and ensure the definition of the target image.

2. The invention can hide 2bit information units by a single dot in a dot coding mode, realizes that the watermark exists horizontally and longitudinally on a monochrome channel, and effectively improves the embedding capacity of an amplitude modulation watermark algorithm.

3. The invention is particularly optimized aiming at the image edge and the watermark edge, solves the problems of color distortion, dot overlap, dot missing and the like by adopting correction and compensation operation, enhances the watermark hiding effect of the image, and realizes the generation of a high-quality halftone target image.

In summary, the invention realizes a high-quality target image by splitting the common halftone dots in the amplitude modulation screening into L-shaped halftone dots formed by a plurality of sub-halftone dots, ensures the embedding effect of watermark images, solves the defect that the amplitude modulation watermark extraction depends on the limitation of a design grating, achieves the effect of retaining the details of high-resolution images, expands the watermark capacity of the amplitude modulation watermark in a halftone dot coding mode, and further improves the embedding transparency of the watermark and ensures the printing quality of the target image by correcting and compensating the image edge problem and the watermark edge problem. Therefore, the invention has the characteristics of high image definition, good hiding effect, flexible printing resolution and large watermark capacity.

Drawings

Fig. 1 is a flow chart of the method of the present invention.

Fig. 2 is a system configuration diagram of the present invention.

Fig. 3 is a diagram of four codes of the L-shaped dots according to the present invention, in which fig. 3 (a) is a diagram of number 00, fig. 3 (b) is a diagram of number 01, fig. 3 (c) is a diagram of number 10, and fig. 3 (d) is a diagram of number 11.

Fig. 4 is a schematic diagram of different break-up coefficients under the same splitting coefficient of the L-shaped mesh point according to the present invention, wherein fig. 4 (a) is a schematic diagram of splitting coefficient M of 3 and break-up coefficient M of 1, and fig. 4 (b) is a schematic diagram of splitting coefficient M of 3 and break-up coefficient M of 2.

Fig. 5 is a target image generated by the L-shaped dot watermarking algorithm and the conventional amplitude modulation watermarking algorithm, wherein fig. 5 (a) is a conventional algorithm overall effect diagram, fig. 5 (b) is a conventional algorithm partial effect diagram, fig. 5 (c) is an L-shaped dot algorithm overall effect diagram, and fig. 5 (d) is an L-shaped algorithm partial effect diagram.

FIG. 6 is a graph comparing the effects of the edge compensation method of the present invention.

Fig. 7 is a comparison diagram of the effect of the color compensation method according to the present invention, in which fig. 7 (a) is a schematic diagram of a conventional processing target image, fig. 7 (b) is a schematic diagram of an L-type dot algorithm without color compensation, and fig. 7 (c) is a schematic diagram of an L-type dot algorithm after color compensation.

Fig. 8 is two watermark images extracted from a halftone target image generated by an L-dot watermark in the present invention, wherein fig. 8 (a) is a horizontal channel watermark information extraction diagram and fig. 8 (b) is a vertical channel watermark information extraction diagram.

Fig. 9 is a diagram of white and black dots produced by combining L-shaped dots with another dot code in the present invention.

Fig. 10 shows 9 patterns of the present invention produced by combining L-shaped dots with surrounding 4 dot codes.

Detailed Description

The invention will be further described with reference to the drawings and specific examples. However, the specific embodiments described herein are merely illustrative of the invention and are not limiting.

The invention relates to a method and a system for realizing high-quality L-shaped net point amplitude modulation screening digital watermarking, which are shown in figure 1 and are flowcharts of the method.

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

Step 1, converting a carrier image color space;

Step 1.1, the original image is read, if the original image is RGB color space, the original image is needed to be converted into a carrier image of CMYK color space, and if the original image is CMYK color space, the original image is not processed. In this embodiment, an RGB image with a length ω and a height h is read, and the image is color space converted. Wherein RGB profile uses sRGB2014.icc, CMYK profiles uses JapanColor2001coated. Icc;

And 1.2, reading two watermark images, and respectively embedding the two watermark images as a transverse watermark and a longitudinal watermark. Converting the watermark image into a gray image, and binarizing the gray value into 0 or 255, wherein the binarization threshold parameter is 128.

Step 2, splitting a mesh adjusting unit:

Step 2.1, calculating a splitting coefficient M according to the resolution P of the input image and the net twine number P of the target image, and considering that the splitting coefficient M is also related to the net twine number C, wherein the calculating formula of the splitting coefficient M is as follows:

step 2.2, according to the splitting coefficient M, splitting all the mesh tone units in the target image into M multiplied by M sub-mesh tone units, and simultaneously obtaining M multiplied by M sub-mesh points;

and 2.3, calculating the center coordinates of all the sub-dots according to the formula (1.1).

The mesh tone unit splitting operation in the step 2 can be replaced by a mesh point splitting operation.

And step 3, dot cut-off operation, namely selecting M ² sub dots from M ² sub dots to carry out cut-off operation, and taking the rest M ²-m² sub dots forming an L shape. And setting a screening mark to obtain the foundation of the L-shaped net point emphasized by the invention. Wherein the splitting coefficient M and the rounding coefficient M satisfy that M > M >0.

And 4, performing dot coding operation, namely embedding 2bit hidden information into a single dot in an L-shaped dot watermarking algorithm, namely, enabling each dot to have four different states of 00, 01, 10 and 11 according to different coding modes, wherein the detailed description is shown in figure 3. In this embodiment, bit0 is taken as the marking bit of the horizontal watermark, and bit1 is taken as the marking bit of the vertical watermark. The actual coding of all L-shaped dots in the target image can be set according to the pixel gray scale conditions in the horizontal watermark image and the longitudinal watermark image.

Step 5, correction and compensation operation:

step 5.1, recalibrating the range of the mesh adjustment unit in the target image, and expanding the screening operation processing range according to a formula (1.2);

Step 5.2, firstly traversing the overflow total amount of the sub-dot gray scale increment in the gradual change dot, calculating the sub-mesh tone quantity in the blank area according to the splitting coefficient and the cut-off coefficient, and performing color compensation according to a formula (1.3), as shown in fig. 7;

And 5.3, calculating the center coordinates of all the L-shaped dots in the target image, traversing all the L-shaped dots, wherein the dot and the other L-shaped dot code generate white dots or black dots because of the structural characteristics of the L-shaped dots, and uniformly applying the compensation quantity of the corresponding sub-dots obtained according to different dot situation modes to all the sub-dots of the L-shaped dots, wherein the optimization strategy is as follows:

The black dot problem is that the black dot problem is caused by the overlap joint phenomenon between adjacent gradual change net points, and the overlap joint between gradual change net points is caused, so that the interval between gradual change net points is covered originally, and a larger gradual change net point is generated, thereby influencing the whole effect of watermark information hiding. The present strategy is to subtract a value related to the current overlap amount from the actual threshold value of all sub-dots of the mesh-tone boundary line, and the purpose of the present strategy is to remove the gray scale increment caused by the dot overlap, so that the gray scale value on the boundary line is kept as consistent as possible with the nearby area. The calculation formula is formula (1.4). The splitting coefficient M and the cut-off coefficient M of the gradual change lattice points are determined firstly, then the gray scale increment R and the number t of the sub lattice points needing to be corrected on the boundary line are calculated, the lattice point compensation quantity lambda is calculated according to the gray scale increment R and the number t of the sub lattice points, and finally the t sub lattice points are traversed and the actual threshold value is modified.

The white point problem is that in order to ensure that the white point with the structure missing does not exist in the target image after the screen is added, a part of sub-screen points need to be supplemented between the two gradual-change screen points. The sub-halftone cells in the blank area are marked as sub-halftone exists, and detail texture features of the continuous tone image are restored as much as possible in the white area. The white point problem after marking the dots is converted into a black point problem in which dot overlap occurs. The calculation formula is formula (1.5).

As shown in fig. 9, and the surrounding 4 dot encodings will produce 9 different patterns of the case, black, white, black-and-black, white-and-white, one-dimensional black-and-white, two-dimensional black-and-white, black-and-white-and-black-and-white, and black-and-white, respectively, as shown in fig. 10. The L-shaped dot situation pattern is marked according to a combination of L-shaped dot coding and four nearby codes.

And 5.4, according to the situation mode of the mesh point, combining the formula (1.4) and the formula (1.5), and finding the corresponding compensation strategy and compensation value. And traversing all L-shaped lattice points related to the edges, and performing lattice point compensation operation so as to optimize edge overlap or missing caused by special lattice point shape combination. The effect is shown in fig. 6.

Step 6, amplitude modulation screening operation:

And 6.1, generating a proper dot threshold matrix according to the shape of the sub-dots, and performing amplitude modulation screening on all marked sub-dots of the target image by using a conventional amplitude modulation screening algorithm to generate a halftone target image, as shown in fig. 5.

And 6.2, storing the halftone target image into a designated folder, inputting the designated folder into printing equipment, and realizing final plate printing to obtain a high-quality full-color printed matter containing water marks.

Therefore, two different watermark images exist in the printed matter, and the hidden watermark information can be extracted by using common detection means such as machine scanning, optical reflection and the like. The embodiment uses grating extraction, and the implementation principle is that the detection of hidden information is realized by utilizing the light splitting effect of a grating sheet, and the extraction effect is shown in fig. 8.

The invention relates to a realization system of high-quality L-shaped net point amplitude modulation screening digital watermark, which is shown in figure 2 and is a structural diagram of the system.

The system comprises a file reading and preprocessing module M201, a mesh unit splitting module M202, a mesh point choosing and sorting module M203, an information encoding module M204, a correction and compensation module M205 and an amplitude modulation screening module M206.

The file reading and preprocessing module M201 is mainly responsible for reading the target image, the watermark image and preprocessing, and is convenient for the operation of the subsequent modules.

The file reading and preprocessing module M201 specifically comprises a target image reading module M2011 for selecting and reading target images from a computer file system, a watermark image reading module M2012 for selecting and reading watermark images from the computer file system, a CMYK generating module M2013 for converting an original image of an RGB color space into a carrier image of a CMYK color space by using an ICC file, a channel splitting module M2014 for splitting the carrier image into C, M, Y, K single-color channel images, and a watermark image binarization module M2015 for carrying out threshold comparison after the watermark images are grayed so as to obtain a binary gray image with a gray value of only 0 or 255.

The mesh tone unit splitting module M202 is used for generating an initial mesh tone unit according to the acquired image resolution and the added mesh line number of the carrier image, splitting the initial mesh tone unit into a plurality of sub-mesh tone units, splitting the mesh points on the mesh tone unit to enable the mesh points to be distributed in the corresponding sub-mesh tone units to form a plurality of sub-mesh points, and calculating the center coordinates of all the sub-mesh points, so that the fineness of the target image can be improved, and meanwhile, a composition premise is created for the L-shaped mesh points.

The mesh adjustment unit splitting module M202 specifically comprises a splitting parameter generating module M2021 for calculating a generated mesh splitting coefficient N and sub-mesh point coordinates p (x, y) according to the printing precision of the target image and the number of extracted grating lines, and a mesh point splitting module M2022 for splitting an original mesh point into M ² sub-mesh points, wherein the mesh point group presents a rectangular shape.

The net point cut-off module M203 is used for changing the cut-off of the split sub net points into L-shaped net points, and simultaneously ensuring that the whole net-adding line number of the L-shaped watermark net points is the same as that of the net points before splitting;

the lattice point cut-off module M203 specifically includes a cut-off parameter generating module M2031, configured to calculate a value range of the cut-off coefficient M, and may actively select a suitable value of the cut-off coefficient M. It should be appreciated that the cut-off coefficient m is an important parameter in the implementation of the L-dot watermarking algorithm, as shown in fig. 4. The increase of the cut-off coefficient M under the same splitting coefficient can reduce the number of the sub-dots, thereby affecting the expressive ability of the L-shaped dots, and meanwhile, the influence of the too small cut-off coefficient M on the watermark quality should be avoided, so that the watermark quality should be flexibly adjusted according to the splitting coefficient and the application scene, and the dot cut-off operation module M2032 is used for cutting off the selected sub-dots, so that the rest sub-dot shapes form an L-shaped shape, namely the L-shaped dots.

The information encoding module M204 is configured to process the watermark image information to encode the watermark image information into the target image.

The information encoding module M204 specifically includes a parameter input module M2041 for selecting an embedding direction of the watermark image according to the watermark image, and a dot encoding module M2042 for determining whether the watermark image is a horizontal watermark image or a vertical watermark image, so as to encode L-shaped dots in the printing plate.

The correction and compensation module M205 is used for correcting the problems of image edge distortion, image color distortion and overlap or deletion of watermark edge areas;

The correction and compensation module M205 specifically comprises an image edge correction module M2051 for expanding a mesh tone unit and recalibrating the range of the mesh tone unit of the screening so as to eliminate the problem of missing of the image edge caused by missing of sub-mesh points, and a mesh point color compensation module M2052 for uniformly distributing the gray values which are increased in the mesh tone unit to all the sub-mesh points of the gradual change mesh points, so that the distortion of the relative gray of the sub-mesh points is reduced while the absolute gray of the sub-mesh points is kept to be increased, and a more natural color compensation result is obtained. The watermark edge analysis module M2053 analyzes and obtains the situation mode of the combination of the dots according to the dot code of each dot and the dot codes nearby the dot, and the watermark edge compensation module M2054 introduces dot compensation quantity according to different situation modes to correct the overlapping or missing image structure generated between the dots.

The amplitude modulation screening module M206 is used for generating a halftone target image containing the watermark and printing out a full-color printed matter with high quality.

The amplitude modulation screening module M206 specifically includes a threshold matrix generation module M2061 for generating a suitable screen dot threshold matrix according to the sub-screen dot shape of the "L" shape, a screening generation module M2062 for generating a halftone target image for printing according to the threshold matrix, a file storage module M2063 for storing the halftone target image in a computer file system, and a platemaking printing module M2064 for printing a high quality full color print containing an invisible watermark.

The method divides the dots in the amplitude modulation screening watermark into a plurality of dot groups with square dot aggregation shapes formed by sub dots, then discards part of the dots, obtains the dot groups with L-shaped dot aggregation shapes formed by the rest dots, namely the L-shaped dots, improves the definition of the target image, eliminates the problem of line number limitation on the design of the grating, analyzes the situation mode generated by combination among the L-shaped dots at the edge of the watermark, gives proper dot compensation, solves the problem of lap joint and missing generated by critical areas of the watermark area and the non-watermark area, effectively improves the concealment of the watermark, reduces the visual difference between the two areas under the high-quality target image, and can simultaneously accommodate two watermark information by a dot coding method, thereby effectively improving the embedding capacity of a watermark algorithm. The system comprises a file reading and preprocessing module, a mesh adjustment unit splitting module, a mesh point choosing and rejecting module, an information encoding module, a correction and compensation module and an amplitude modulation screening module. The realization method and the system for the high-quality L-shaped halftone amplitude modulation screening digital watermark image realize clearer halftone target images under the condition of the same number of extracted grating lines, reduce watermark traces, improve the embedding capacity of the watermark, and have the characteristics of high image definition, good hiding effect, flexible printing resolution and large watermark capacity.

In order to prove the effectiveness of the invention, the invention designs a corresponding subjective and objective evaluation experiment:

in the analysis of the image definition effect, the L-shaped dot watermarking algorithm is compared with the conventional amplitude modulation screening watermarking algorithm on the disclosed image data set, and the effectiveness of the invention is illustrated from two aspects of subjectivity and objectivity.

Firstly, the invention uses the photo in the DIV2K_train_HR high-resolution reconstruction data set as the data set of the experiment, and uses the conventional amplitude modulation screening watermarking algorithm and the L-shaped screen watermarking algorithm to carry out amplitude modulation screening respectively, and generates a halftone target image based on the same number of extracted grating lines. Taking the data set "0149.Png" as an example, different halftone target images were obtained, as shown in fig. 5. It can be seen that there is a great gap between the amplitude modulation screening watermarking algorithm using conventional dots and the present invention, and the conventional algorithm has difficulty in satisfactory effect on detail processing, especially in the case of obvious loss at the embossment and the tower tip. In contrast, the algorithm of the L-shaped lattice point can still keep more details, and the image definition is obviously improved.

The invention evaluates the conventional amplitude modulation screening watermark algorithm and the L-shaped dot watermark algorithm by taking SSIM as an objective evaluation index. SSIM is a measurement method for measuring the similarity of two image structure information, the index considers the perception difference of a human visual system on an image structure, and meanwhile, the index comprises multiple aspects of brightness, structure and contrast and the like, so that the SSIM evaluation index is superior to PSNR. The invention selects SSIM to objectively evaluate the overall image quality of the halftone target image. The calculation formula is as follows:

Mu _x in the formula (1.9) represents the average value of x, sigma _x represents the standard deviation of x, sigma _xy represents the covariance of x and y, C ₁、C₂ is a normal number constant, and the structural similarity is prevented from being 0.

The halftone target image obtained by computationally comparing the conventional amplitude modulation screening watermarking algorithm and the L-type dot watermarking algorithm are shown in table one, in which the best results for each picture have been marked with bold.

Surface halftone target image SSIM quality evaluation

According to the first table, the conclusion can be drawn that compared with the conventional amplitude modulation screening watermarking algorithm, the halftone target image obtained by the algorithm provided by the invention has better structural similarity, which means that the L-shaped halftone watermarking algorithm can obtain the halftone target image with higher definition. Even in the case of more image details or complex image textures, better image quality can be maintained.

And then objectively evaluating the embedded watermark trace, wherein SSIM is selected to serve as an objective evaluation standard for the watermark trace, at the moment, the calculation area is reduced to a rectangular area containing the watermark, and the calculation amount is reduced while the accuracy of the result is improved. The results are shown in table two, where the best results for each picture have been marked with bold:

Surface two watermark trace area SSIM evaluation

According to the second table, the conclusion can be drawn that compared with the conventional amplitude modulation screening watermarking algorithm, the watermarking trace area obtained by the algorithm provided by the invention has better structural similarity, the watermarking hiding effect is improved, and the balance between the image definition and the watermark invisibility is realized.

The above process includes all processes of file reading and preprocessing, dot splitting, dot picking and dropping, information coding, correction and compensation, amplitude modulation and screening.

In the above embodiments, it may be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When used in whole or in part, is implemented in the form of a computer program product comprising one or more computer instructions. When loaded or executed on a computer, produces a flow or function in accordance with embodiments of the present invention, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a computer network, or other programmable apparatus. 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, computer, server, or data center to another website, computer, server, or data center by a wired (e.g., coaxial cable, fiber optic, digital Subscriber Line (DSL), or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that contains an integration of one or more available media. The usable medium may be a magnetic medium (e.g., floppy disk, hard disk, tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid state disk Solid STATE DISK (SSD)), etc.

The foregoing is merely illustrative of specific embodiments of the present invention, and the scope of the invention is not limited thereto, but any modifications, equivalents, improvements and alternatives falling within the spirit and principles of the present invention will be apparent to those skilled in the art within the scope of the present invention.

Claims

1. The method for realizing the high-quality L-shaped net point amplitude modulation screening digital watermark is characterized by comprising the following steps of:

2. The method for realizing high-quality L-shaped halftone amplitude modulation screening digital watermarking according to claim 1, wherein step 1 skips converting carrier image color space by inputting CMYK pictures;

In the step 2, a splitting coefficient M is introduced into the splitting mesh adjusting units, 1 mesh adjusting unit is split into mesh adjusting units of m×m sub-dots, the mesh line pitch p in the x direction is split into M parts, the mesh line pitch q in the g direction is split into M parts, the calculation formula of the coordinates corresponding to the centers of the sub-dots is formula (1.1), and the expression is:

3. The method for realizing high-quality L-shaped halftone amplitude modulation screening digital watermarking according to claim 1, wherein in the step 3, the halftone number in the L-shaped halftone is dynamically generated according to the target image line number and the extracted raster line number;

4. The method for realizing high-quality L-shaped halftone amplitude modulation screening digital watermark according to claim 1, wherein in the step 4 halftone coding operation, the actual coding of the L-shaped halftone left by the tradeoff has 4 coding modes, namely, coding 00 is truncated left of the graded halftone, coding 01 is truncated right of the graded halftone, coding 10 is truncated left of the graded halftone, and coding 11 is truncated right of the graded halftone.

5. The method for realizing high-quality L-shaped halftone amplitude modulation screening digital watermarking according to claim 1, wherein the specific method in step 5 is as follows:

In the correction and compensation operation, the method is optimized for three aspects, namely, the method aims at color deficiency caused by cutting off a mesh tone unit at the edge of an image, color deficiency caused by lattice point cut-off operation and edge lap joint or deficiency caused by special lattice point shape combination;

Aiming at the color deficiency caused by the lattice point cut-off operation, determining lattice point compensation quantity according to the gray value of an original image and the gray value of the image after the cut-off operation, and uniformly applying the compensation quantity to sub-lattice points of all L-shaped lattice points so as to optimize the color deficiency caused by the lattice point cut-off operation after the lattice point is split, wherein the optimization formula is as follows:

The black dot problem is that a splitting coefficient M and a cut-off coefficient M of a gradual change dot are firstly determined, then the gray scale increment on a boundary line and the number t of sub dots needing correction are calculated, dot compensation quantity lambda is calculated according to the gray scale increment, and finally t sub dots are traversed and the actual threshold value is modified, wherein a calculation formula (1.4) of the dot compensation quantity lambda of the black dot problem is as follows:

6. The method for realizing high-quality L-shaped halftone amplitude modulation screening digital watermarking according to claim 1, wherein in the step 6, amplitude modulation screening operation is performed only on the sub halftone with calibration;

the shape of the sub-lattice points comprises rectangular lattice points, round lattice points, diamond lattice points, star lattice points, diamond lattice points, wave line lattice points or cross lattice points.

7. A realization system of high-quality L-shaped net point amplitude modulation screening digital watermark is characterized by comprising:

8. The system for implementing high-quality L-dot amplitude modulation screening digital watermarking according to claim 7, wherein the file reading and preprocessing module includes:

9. The system for implementing high-quality L-dot amplitude modulation screening digital watermarking according to claim 7, wherein the halftone unit splitting module includes:

the splitting parameter generating module is used for calculating and generating a dot splitting coefficient M and sub dot coordinates p (x, y) according to the printing precision of the target image and the number of extracted grating lines;

The net point splitting module is used for splitting the original net point into M ² sub net points, and the net point group presents a rectangular shape at the moment;

The lattice point cut-off module comprises:

The lattice point cut-off operation module is used for cutting off part of lattice points in the lattice point group to form an L-shaped lattice point group, wherein the lattice points of which the cut-off part are rectangular lattice point groups with the number of the composition lattice points of m ²;

the information encoding module includes:

10. The system for implementing high-quality L-dot amplitude modulation screening digital watermarking according to claim 7, wherein the correction and compensation module includes:

the watermark edge compensation module is used for introducing dot compensation quantity according to different situation modes and correcting overlapping or missing image structures generated between dots;

The amplitude modulation screening module comprises: