CN106650874B - Display, anti-counterfeiting method and device for paper fiber characteristics - Google Patents

Abstract

The invention relates to a display and anti-counterfeiting method and device of paper fiber characteristics. The invention first collects the electronic image of a specific area of the paper; then uses the fiber image template to perform template matching on the collected electronic image, and selects the feature point corresponding to the electronic image with the fiber image template similarity greater than a set threshold as the paper fiber image feature point , and screen the obtained feature points according to the set priority; finally, encode the feature information of the screened feature points to form a two-dimensional code map corresponding to the paper fiber characteristics, and the feature display area of the code map uses the The feature point display symbol corresponding to the template information used by the feature point is used to display the fiber texture direction and fiber thickness information at the feature point position. The present invention can acquire accurate and significant paper features, and display the paper features through a two-dimensional code, so that the user can conveniently directly judge the paper features according to the obtained two-dimensional code map.

Description

Display, anti-counterfeiting method and device for paper fiber characteristics

technical field

The invention relates to a display and anti-counterfeiting method and device of paper fiber characteristics, belonging to the technical field of image recognition and coding.

Background technique

With the rapid development of Internet finance recently, various forms of online payment, online transactions, online banking and e-commerce business applications have emerged. It has become very important to build an identity authentication mechanism for the security and mutual trust of both parties. From the current operating mechanism and technical route From the point of view, there are problems such as difficulty in identity authentication, low security, frequent transaction processes, high management costs, and security loopholes in electronic transactions. In this technical area, there is a lot of room for improvement. For example, digital watermarks or digitally signed checks, electronic invoices, electronic contracts, special bills and other transaction media all have security problems of identity authentication and potential security problems of being copied and counterfeited. However, due to the risk of no physical evidence for electronic certificate information such as QR code information, electronic signature scan codes, electronic labels, and software key sharing on the Internet, there are still security loopholes in the identity confirmation of the transaction parties. This security loophole can be solved by identifying paper, but important information on paper materials such as official documents, materials, bills and certificates is usually prevented from counterfeiting by stamping ordinary paper. The easier it is, the more difficult it becomes to prevent counterfeiting, which in turn makes it less secure.

SUMMARY OF THE INVENTION

The purpose of the present invention is to provide a method for displaying paper fiber characteristics to solve the problem that the current paper fiber characteristics cannot be visually displayed; the present invention also provides an anti-counterfeiting method based on paper fiber characteristics to realize the identification of paper.

The present invention provides a method for displaying the characteristics of paper fibers in order to solve the above-mentioned technical problems. The displaying method includes the following steps:

1) Collect an electronic image of a specific area of the paper;

2) using the fiber image template to perform template matching on the collected electronic images, and selecting the feature points corresponding to the electronic images whose similarity with the fiber image template is greater than the set threshold as the paper fiber image feature points;

3) Screen the obtained feature points according to the set priority;

4) encode the feature information of the feature points after the screening, form a two-dimensional code map corresponding to the paper fiber feature, the feature display area of the code map adopts the feature point display symbol corresponding to the template information used by the feature point, the described The feature point display symbol of is used to display the fiber texture direction and fiber thickness information at the feature point position.

Further, the two-dimensional coding map also includes at least one of an angular locator, a collection position symbol, a center calibration locator, an edge calibration area and a feature one-dimensional coding area, and the angular locator is used to locate the paper feature. A corner of the two-dimensional code map; the collection position symbol is used to indicate the collection position of the paper fiber image; the center calibration locator is used for the position calibration of the coding block of the paper feature two-dimensional code map; the edge calibration The area is used to correct the position of the encoding block in the two-dimensional encoding map of the paper feature; the one-dimensional encoding area of the characteristic is used to display the one-dimensional encoding of the characteristic.

Further, the feature one-dimensional encoding is used to encode the feature information of the screened feature points, and the encoded information of each feature point at least includes the template number, abscissa and ordinate used by the feature point.

Further, the method also includes coarse-graining and normalizing the feature information on the scale: dividing the collected electronic image into blocks, and forming a new scale with a numerical value in blocks from top to bottom and from left to right Coordinate system; the center of the connected domain in the binary image of the feature points obtained in step 2) falls into the coordinates of the image block region as the new feature point position coordinates, and the obtained normalized feature point coordinates (X _τ , Y _τ ) are:

X _τ =a _τ -x ₀ , Y _τ =b _τ -y ₀ ,τ=1,...E

Where (x ₀ , y ₀ ) is the centroid of all block feature points; (a _τ , b _τ ) is the block coordinate of the τth block, and E is the number of remaining connected domains after sorting and filtering.

Further, the priority in step 3) adopts the following rules:

a. The priority of feature points with large template width is greater than that of feature points with small template width;

b. The priority of the feature point with a large connected domain area is greater than that of the feature point with a small connected domain area;

c. If the area of the connected domain is the same, the priority is determined from top to bottom and left to right according to the order of appearance.

The invention also provides a display device for paper fiber characteristics, the identification device includes a collection module, a matching module, a screening module and a coding display module,

The collection module is used to collect electronic images of specific areas of the paper;

The matching module is used to perform template matching on the collected electronic images using the fiber image template, and select the feature points corresponding to the electronic images whose similarity with the fiber image template is greater than the set threshold as the paper fiber image feature points;

The screening module is used for screening the obtained feature points according to the set feature point priority;

The coding display module is used for coding the feature information of the screened feature points to form a two-dimensional coding map corresponding to the paper fiber feature, and the feature display area of the coding map adopts the feature corresponding to the template information used by the feature points. The point display symbol is used to display the fiber texture direction and the thickness information of the fiber at the position of the feature point.

The present invention also provides an anti-counterfeiting method based on the characteristics of paper fibers, and the steps of the anti-counterfeiting method are as follows:

1) Collect an electronic image of a specific area of the paper;

2) Use the fiber image template to perform template matching on the collected electronic images, and select the electronic images whose similarity with the fiber image template is greater than the set threshold as the paper fiber image feature;

3) Screen the obtained feature points according to the priority of the feature points;

4) Code the feature points after the screening, and form a two-dimensional code map corresponding to the paper fiber feature, and the feature display area of the code map adopts the feature point display symbol corresponding to the template information used by the feature point, and the feature point The display symbol is used to display the fiber texture direction and fiber thickness information at the feature point position;

5) The obtained two-dimensional code is printed on the protected paper, and the characteristic display area in the paper characteristic two-dimensional code diagram printed on the paper is compared. If there is the same display symbol structure combination, it is proved that the paper is Paper that has not been scanned and copied.

Further, the two-dimensional coding map also includes at least one of an angular locator, a collection position symbol, a center calibration locator, an edge calibration area and a feature one-dimensional coding area, and the angular locator is used to locate the paper feature. A corner of the two-dimensional code map; the collection position symbol is used to indicate the collection position of the paper fiber image; the center calibration locator is used for the position calibration of the coding block of the paper feature two-dimensional code map; the edge calibration The area is used to correct the position of the encoding block in the two-dimensional encoding map of the paper feature; the one-dimensional encoding area of the characteristic is used to display the one-dimensional encoding of the characteristic.

Further, the method also includes coarse-graining and normalizing the feature information on the scale: dividing the collected electronic image into blocks, and forming a new scale with a numerical value in blocks from top to bottom and from left to right Coordinate system; the center of the connected domain in the binary image of the feature points obtained in step 2) falls into the coordinates of the image block region as the new feature point position coordinates, and the obtained normalized feature point coordinates (X _τ , Y _τ ) are:

X _τ =a _τ -x ₀ , Y _τ =b _τ -y ₀ ,τ=1,...E

Where (x ₀ , y ₀ ) is the centroid of all block feature points; (a _τ , b _τ ) is the block coordinate of the τth block, and E is the number of remaining connected domains after sorting and filtering.

Further, the priority in the step 3) follows the following criteria:

a. The priority of feature points with large template width is greater than that of feature points with small template width;

b. The priority of the feature point with a large connected domain area is greater than that of the feature point with a small connected domain area;

c. If the area of the connected domain is the same, determine the priority from top to bottom and from left to right according to the order of appearance.

The beneficial effects of the present invention are as follows: the present invention first collects the electronic image of a specific area of the paper; then uses the fiber image template to perform template matching on the collected electronic image, and selects the characteristics corresponding to the electronic image whose similarity with the fiber image template is greater than the set threshold The points are used as the feature points of the paper fiber image, and the obtained feature points are screened according to the set feature point priority; finally, the feature information of the screened feature points is encoded to form a two-dimensional coding map corresponding to the paper fiber feature. The feature display area of the coding map adopts the feature point display symbol corresponding to the template information used by the feature point, and the feature point display symbol is used to display the fiber texture direction and fiber thickness information at the feature point position. The present invention can acquire accurate and significant paper characteristics, and visually display the paper characteristics through the two-dimensional code, and the user can directly judge the characteristics of the paper according to the obtained two-dimensional code map.

In addition, the present invention also prints the obtained two-dimensional code map on the protected paper, and compares the feature display area in the two-dimensional code map of the paper characteristics printed on the paper, if it has the same display symbol structure combination, It proves that the paper is the paper that has not been scanned and copied, and realizes the anti-counterfeiting of the paper. This anti-counterfeiting method can be used for the anti-copying watermark technology, single-time effective digital signature, electronic and physical mixed invoices, contracts and bank payment security in Internet finance. and other fields, to improve the credibility of the identity authentication of the current party of the transaction.

Description of drawings

Figure 1 is a fiber image of paper;

Fig. 2 is the basic parameter schematic diagram of standard template;

Fig. 3 is the standard template image of generation and the schematic diagram of numbering;

4 is a schematic diagram of a connected domain screening process of a feature point binary image;

Fig. 5 is a schematic diagram of one-dimensional code encoding of paper characteristics;

Fig. 6 is a schematic diagram of two-dimensional encoding of paper characteristics;

Figure 7-a is a two-dimensional code diagram of the position of the lower left corner of the paper in the embodiment of the present invention;

Figure 7-b is a two-dimensional coding diagram of the position of the lower right corner of the paper in the embodiment of the present invention;

FIG. 8 is a schematic diagram of the anti-counterfeiting release principle of the anti-copying technology;

FIG. 9 is a schematic diagram of the principle applied to hybrid electronic contracts and electronic invoices.

Detailed ways

The specific embodiments of the present invention will be further described below with reference to the accompanying drawings.

An embodiment of a method for displaying paper fiber characteristics of the present invention

The invention first collects the electronic image of a specific area of the paper; then uses the fiber image template to perform template matching on the collected electronic image, and selects the feature point corresponding to the electronic image with the fiber image template similarity greater than a set threshold as the paper fiber image feature point ; Finally, the obtained feature points are screened according to the set feature point priority, so as to identify the fiber characteristics of the paper.

1. Capture an electronic image of a specific area of the paper.

The paper feature in the present invention refers to the image feature of the paper fiber in the electronic image collected by the image sensor from the paper feature area, magnified by an optical microscope lens. The specific area is defined at the intersection of parallel lines at a certain distance from the two edges of the paper (10mm from the two edges can be selected), and the area is about 0.3*0.3mm. The reason for defining the feature extraction object in this way is that the accuracy of manually holding and positioning the paper is about 0.1mm, and the physical magnification of the optical microscope is roughly: 6mm/0.3mm area, that is, about 20 times, of which 6mm is the width of the imaging plate , 0.3mm is the actual size of the microscopic area. Assuming that the resolution of the image sensor is 640*480, the effective area is set to 480*480, that is, the width is 480 sampling points. If fibers with a width greater than 10 pixels in the image can be detected and identified, the actual fiber width that can be detected and identified is 10*0.3mm/480=0.006mm, which means that fibers with a thickness greater than 6 microns can be identified.

The paper fiber image is defined as: the fibers in the paper are projected by light and collected by the image sensor, and the two parallel curves formed by the change of gray level in the obtained grayscale image are the fiber imaging in the paper. The image of the crisscrossed fibers constitutes the fiber image, as shown in Figure 1. This image is taken at the intersection of 10mm from the lower edge of the paper and 10mm from the left edge, with an area of about 0.3mm*0.3mm. Therefore, the feature extraction object set by the present invention is: paper fibers with a thickness of more than 6 microns are imaged on an imaging panel of about 6 mm.

2. Construct the image template

It is very difficult to obtain a fixed and stable paper identity feature from the chaotic fiber images. Using different image processing techniques and feature extraction techniques, different feature description results will be obtained, and it is impossible to form a unique and consistent. Identity feature encoding information. To this end, the present invention uses a self-constructed fiber feature image template, and performs image matching on the captured image from a self-constructed standard fiber template image to extract paper fiber image features.

The physical properties of fibers in paper are closely related to their image features in imaging. The most obvious physical features of local fiber images are: width, length, direction, and changes in light and shade, as shown in Figure 2. Standard feature matching templates such as As shown in Figure 2, where w represents the imaging width of the fiber, which actually reflects the diameter of the fiber. In order to reduce the ambiguity of extracting features through standard template matching, discrete values are specified. Width is used as the standard fiber imaging width, d represents the width of the standard template, and d=6w is set to represent the extension direction of the fibers in the current standard template image unit, and the discrete value is also used, and only the direction of the set number is taken. This implementation In the example, only 12 radian directions are taken, θ∈Φ:

Φ={θ _j |j=1,2,…12}={0 π/12 π/6 π/4 π/3 5π/12 π/2 7π/12 2π/3 3π/4 5π/6 11π/ 12}

The present invention adopts two inverted Gauss cap mathematical models, and synthesizes the fiber template image through the method of rotation and scaling, and the obtained fiber template image model is:

where (x,y)=(x'cosθ+y'sinθ,-x'sinθ+y'cosθ), θ∈Φ (2)

Assuming that there are 3 fiber widths and 12 fiber directions, a total of 36 standard fiber template images are generated. In formula (1), μ is the average gray value of the effective area of the entire sampling image:

where f(i,j) is the grayscale of the sampling point obtained by the image sensor, N is the number of sampling points in the effective sampling width direction, and η in the formula (1) is the contrast of the grayscale of the fiber edge relative to the mean grayscale of the template image, set is: η=20%; in formula (1), c ₁ and c ₂ are the coordinates of the two gray-scale valleys, which are actually the respective mean values of the two Gaussian curves. Formula (1) uses a flipped Gaussian cap to simulate fiber imaging. The specific values _of the _two dark edges, c1 and c2 are determined by the center coordinates of the template image and the fiber width, namely:

c ₁ =x _m -w/2

c ₂ =x _m +w/2

The Gaussian variance σ in formula (1) is: σ=w/7, (x', y') are the coordinates before rotation, (x, y) are the coordinates of the image after rotation, the rotation angle is θ, θ∈Φ, That is, the radian value of the rotation angle is taken according to the radian value of the above-mentioned 12 directions.

When 3 fiber widths and 12 fiber directions are selected, 36 standard template images can be obtained. As shown in Figure 2, the subscripts of characters B ₀ to B ₃₅ represent the template numbers. Template: 12 standard template fibers with numbers 0-11 have the largest fiber width; template fibers with numbers 12-23 have the middle width, and template fibers with numbers 24-35 have the smallest width.

When using the above standard templates for screening, the priority must be set. The priority of the template determines the selection and selection order of features. In order to select the features with good stability, the features extracted by the template with high priority will be selected first. According to The generated standard fiber template images and their numbering rules, the priority ordering method of the templates is: the template with the same width has the same priority; the template with a smaller number has a higher priority than the template with a larger number.

3. Extract paper image features using the constructed standard template.

Use the constructed standard template to perform point-by-point matching within the effective area (M×N) of the captured paper fiber image. Assuming that the template image matrix is represented by B _k , the image grayscale matrix at the corresponding scanning position is represented by X _{(x, y)} , and the subscript (x, y) is the matching center coordinate, then the similarity between the template and the image is:

Compare the calculated similarity between the template and the image with the set threshold, and take the point whose similarity is greater than the set threshold as the primary feature point set.

Considering the standardization of feature extraction, in order to avoid ambiguity, the present invention adopts uniform standard parameters, and sets the threshold value T=0.5; the matching range parameters M=640, N=480. The image at the coordinate position of the feature point in the feature point set and the image template image used by the feature point have similar structural features. Every time a standard template is used for matching scan, a binary image of feature point position will be generated, and 36 standard matching templates will generate 36 binary images of feature points. The feature points in the binary image will form some areas that are internally connected and isolated from each other. These connected areas are numbered, and there are D connected areas. means, a=1,2...,D, the maximum diameter of all connected domains is:

Among them, Δx and Δy represent the maximum width and height of the connected area of the feature points, respectively. The diameter of the connected area is sorted from large to small, only the connected areas with the first order are retained, and the remaining connected areas with small diameters are discarded. The method of smaller connected domain can enhance the stability of feature extraction. In this embodiment, the change of the connected domain of feature points in the binary image after sorting and screening is shown in FIG. 4 .

4. Normalize the extracted feature points.

In order to further enhance the feature stability and reduce the influence of the disturbance of the position of the image acquisition point on the feature stability, the position coordinate information of the connected domain of the above feature points is coarse-grained on the scale. The specific process is as follows:

The entire image acquisition area is divided into blocks, that is, the image acquisition area is divided into 15 rows×15 columns larger blocks, the size of each block area is (N/15)×(M/15), and N and M are The width and height of the original image. And form a new scale coordinate system with serial numbers arranged in blocks from top to bottom and from left to right. Then the feature information extracted by the template can be reset as follows: in the original feature point binary image, the coordinates of the above-mentioned image block region where the center of gravity of the connected domain falls, as the new feature point position coordinates, its position coordinate value is the above-mentioned points. The coordinate value of the new coordinate system formed by the block takes the coordinate value of the block and the template number used as the new feature point information, which is called "block feature point" in the present invention, or "feature point" for short. Obviously, the feature information of the block feature points includes: the position coordinate information in the new coordinate system and the template number information used to generate the binary image.

Let the number of remaining connected domains after sorting and filtering be E. In order to normalize, set E=16, calculate the barycentric coordinates of each connected area, and determine the above-mentioned image blocks that fall into it. If it falls into the block of row a and column b, the block coordinates (a, b) are used to represent the position information of the block feature, and the position coordinates of the block feature points are normalized as:

X _τ =a _τ -x ₀ , Y _τ =b _τ -y ₀ ,τ=1,...E

In the above formula (x ₀ , y ₀ ) is the centroid of all block feature points; block coordinates (a _i , b _i ) are normalized to obtain normalized feature point coordinates (X _τ , Y _τ ).

The paper feature points can be represented by the following methods: only E block feature points are taken from the image collected each time, and all the feature points information of the paper can be expressed as:

δ _τ (k,s,(X _τ ,Y _τ )),k=0,1,…,35,τ=1,2,…,E

Where k is the template number of the template B _k used, s is the area of the connected domain where it is located, and (X _τ , Y _τ ) is the normalized block feature point coordinates.

5. Screening paper feature points

The selection of paper feature points is determined by sorting the priority of the feature points. The priority is represented by a function O. The priority rules of paper feature points are:

A. The priority of feature points with large template width is greater than that of feature points with small template width. For example, the template number used by the feature point δ ₁ is k ₁ , the template width is w _k1 , and the area of the connected domain where it is located is s ₁ ; the template number used by the feature point δ ₂ is k ₂ , the template width is w _k2 , so The area of the connected domain is s ₂ ; if the template width w _k1 >w _k2 , then P(δ ₁ (k ₁ ,s ₁ ))>P(δ ₂ (k ₂ ,s ₂ )).

B. The priority of the connected domain where the feature point is located is larger than the priority of the feature point with a small connected domain area.

C. If the area of the connected domain is the same, the priority is determined from top to bottom and left to right according to the order of appearance.

According to the priority principle of feature points, special cases are handled as follows:

i. If the barycentric coordinates of the connected domain fall into the same block, the feature points with small connected domain area are directly discarded.

ii. If the center of gravity of different connected regions generated by the same template falls into the same block, first determine the area of the connected region, and if the area is the same, compare the template width. The template with the smaller width is discarded; if both are the same, it is determined according to the order of appearance, and the first one is retained.

According to the above-mentioned feature point priority principle and special case processing method, the reserved feature points are reordered so that the feature point number is equal to its priority. If P(δ _τ ) is used to represent the priority of feature point δ _τ , then:

P(δ ₁ )>P(δ ₂ )>...>P(δ _E )

Obviously, the feature points with smaller serial numbers have higher priority.

6. Code the paper features of the feature points after the screening to form a two-dimensional code map of the paper. The feature display area of the code map adopts the feature point display symbols corresponding to the template information used by the feature points, and the feature point display symbols are used. It is used to display the fiber texture direction and fiber thickness information at the position of the feature point.

1) The screened paper features are encoded with a one-dimensional code.

The one-dimensional coding of paper features is to encode the feature information of the feature points, and to determine the coding sequence according to the serial number or priority of the feature points. The encoding method is shown in Figure 5, and the one-dimensional encoding content includes: the version number of the encoding standard, the collection location identifier, the number of feature points, and the feature point encoding (including the template number, abscissa, ordinate, It consists of reserved bits and check code), time code, global unique identification number of paper identification equipment, user-set identification code, user-set cipher text code and check code. This feature one-dimensional encoding scheme can encode 16 feature points, and the maximum number of encoded bits is: 8+2+1+4+1+16×(6+4+4+1+1)+24+32+ 16+24+1=370 bits.

The meaning of the 2-digit collection position code after the version number in Figure 5 is defined as follows: 00 represents the lower left corner of the paper; 01 represents the lower right corner; 10 represents the lower left corner of the back of the paper; 11 represents the lower right corner of the back of the paper. The 4-bit abscissa and 4-bit ordinate in the feature point encoding represent the block coordinates of the normalized feature points, and the highest bit of the 4 bits represents the sign of the coordinates: 1 represents negative coordinates, 0 represents positive coordinates . The one-dimensional code of paper features can be directly transmitted to the terminal through the interface of the device, and can also be encoded into the two-dimensional code map of paper features, displayed by the paper recognition device or printed on the paper, so as to facilitate the direct extraction of mobile phones, computers and other devices.

2). On the basis of one-dimensional encoding, two-dimensional encoding is performed on the image paper.

The two-dimensional code of paper features is shown in Figure 6. The content of the two-dimensional code includes coding block, corner locator, collection position symbol, center calibration locator, edge calibration area, feature one-dimensional coding area, feature display area and feature points Symbols are displayed, and the above-mentioned contents included are described in detail below.

The coding block is the smallest coding basic unit printed on the paper or displayed on the device. It consists of a black rectangular block or a white rectangular block of the same size. The black block represents the code "1", and the white block of the same size represents Code "0".

The corner locator is composed of 5 × 5 coding blocks, and the shape is a mouth shape with a dot in the middle. It is used to locate a corner of the two-dimensional coding map of the paper feature. For example, the other three corners outside the symbol in the lower left corner in Figure 6 The position symbols are all corner locators.

The collection position symbol is composed of 5 × 5 coding blocks, and the shape is shown in the lower left corner of Figure 6, which is used to indicate the collection position of the paper fiber image. Figure 6 represents the paper feature extracted in the lower left corner of the paper. For the image collected in the lower right corner, the lower left corner symbol needs to be flipped 180 degrees left and right and placed in the lower right corner position, and the original lower left corner position is changed to a corner locator.

The center calibration locator is composed of 3×3 coding blocks, the shape is like a mouth shape, and the position is in the center of the feature display area of the feature 2D coding map. The center position of the area is also the centroid position of each paper feature point.

On the four connecting positions of the central coding block of the diagonal locator or the central coding block of the acquisition position symbol, the side calibration area is composed of equidistant, black and white coding blocks, which are used for calibrating features. The encoding block location in the 2D encoding map.

The area on the left and right sides of the feature one-dimensional coding area or the upper and lower parallel positions of the edge calibration area is the paper feature one-dimensional coding area. It is mainly used to encode and display the one-dimensional code of paper features in this area. The coding sequence of the one-dimensional coding of the paper features is carried out in sequence according to the numerical labels shown in FIG. 6 . That is, starting from the position of the coding block numbered 1, and rotating counterclockwise along the characteristic one-dimensional coding area for two circles, until the position numbered 370 ends, a total of 370 bits can be encoded. According to the design scheme of feature one-dimensional encoding, the entire encoding position can be occupied.

The feature display area is the feature display area in addition to the corner locator or the collection position symbol and the feature one-dimensional code area. The remaining middle area of the paper feature two-dimensional code map is the feature display area. The feature display area is used to display some key and high-priority features. point information, and the feature point information is expressed by the feature point display symbol. The purpose is to assist in fine-tuning the position of the paper when using the paper recognition device for paper image acquisition and feature recognition, so as to obtain accurate and significant paper features. Another function That is, when the user performs paper verification, it is convenient for the user to directly judge the characteristics of the paper with the naked eye, so as to compare the two-dimensional code map of the paper characteristics displayed by the paper identification device with the printed two-dimensional code map, and observe whether they have the same characteristics. Determine whether the paper has been scanned and copied.

The feature point display symbol is a simplified representation of the template image used by the feature point, which corresponds to the template information used by the feature point, and can express the fiber texture direction, fiber thickness and other information at the position of the feature point. Table 1 shows the correspondence between display symbols and template numbers in the embodiment.

Table 1

Through the above coding method, the obtained two-dimensional code map of paper characteristics is shown in Figure 7-a and Figure 7-b, wherein Figure 7-a is the characteristic two-dimensional code map formed by collecting and identifying the lower left corner of the paper; Figure 7-b It is a QR code map of paper characteristics collected and recognized in the lower right corner of the paper.

An embodiment of a display device for paper fiber characteristics of the present invention

The identification device in this embodiment includes a collection module, a matching module, a screening module, and a coding display module. The collection module is used to collect electronic images of a specific area of the paper; the matching module is used to perform template matching on the collected electronic images using a fiber image template. , select the feature points corresponding to the electronic image whose similarity of the fiber image template is greater than the set threshold as the feature points of the paper fiber image; the screening module is used to screen the obtained feature points according to the set priority of the feature points; the coding display module uses The feature information of the screened feature points is encoded to form a two-dimensional code map corresponding to the paper fiber feature. The feature display area of the code map uses the feature point display symbols corresponding to the template information used by the feature points, and the feature points display The symbol is used to display the fiber texture direction and fiber thickness information at the feature point position. The specific implementation means of each module has been described in detail in the embodiment of the identification method, and will not be repeated here. Through the display method and device of the present invention, the same feature description results can be obtained, unique and consistent identity feature encoding information can be formed, and the identification of paper fiber features can be realized.

The present invention can acquire accurate and significant paper characteristics, and visually display the paper characteristics through the two-dimensional code, and the user can directly judge the characteristics of the paper according to the obtained two-dimensional code map.

An embodiment of an anti-counterfeiting method based on paper fiber characteristics of the present invention

The anti-counterfeiting method in this embodiment is implemented on the basis of the method for displaying the characteristics of paper fibers. The obtained two-dimensional code map of the paper is directly printed on ordinary paper, which can not only detect the copied digital watermark and digital signature, but also be used for network transactions. Physical retention of the parties. Print the obtained two-dimensional code on the protected paper, and compare the characteristic display area in the two-dimensional code map of paper characteristics printed on the paper. Scanned and copied paper.

The following is an example of applying the anti-counterfeiting method to scan-resistant copying and hybrid electronic contracts, wherein application example 1 is applied to scan-resistant copying, and application example 2 is applied to hybrid electronic contracts.

Application example 1

In recent years, digital watermarking technology and encrypted two-dimensional code technology have been widely used. However, these two technologies still have insurmountable technical shortcomings in practical applications: that is, they cannot resist the infringement of copyright by high-resolution scanning and copying technology. However, if the paper anti-counterfeiting method of the present invention is applied, the above problems can be solved. The following is an application example 1 of the paper anti-counterfeiting method of the present invention.

As shown in Figure 8 is an anti-counterfeiting publishing method: a. First, the paper identification device collects and identifies the paper at the lower left corner or the lower right corner of the blank paper at a distance of 10mm from the edge of the paper, and obtains a two-dimensional code map of paper characteristics; b . The cipher text information of the issuing party is processed by digital watermarking technology or two-dimensional code technology to form a digital watermark or ordinary two-dimensional code, and print it on the paper to be protected together with the two-dimensional code image of the paper. The meaning of the dotted line in the figure is that the extracted one-dimensional code information of paper characteristics and the ciphertext information of the publisher can also be mixed to form a digital watermark or ordinary two-dimensional code printed on the protected paper.

Determine if the paper is originally published or scanned or copied by:

(a) Use the paper identification device to display the feature two-dimensional code map of the current paper, and compare the feature display area in the paper feature two-dimensional code map printed on the paper. If it has the same display symbol structure combination, it can prove that the paper Paper that has not been scanned and copied.

(b) Extract digital watermark information or ordinary two-dimensional code information with mobile phones and other scanning devices, and decode the ciphertext of the issuing party with the one-dimensional encoding information of paper characteristics obtained from the two-dimensional encoding map of paper characteristics, so as to determine that the content of the paper comes from the issued square.

Application example 2

In recent years, Internet finance, e-commerce, and online transactions have developed rapidly. However, the problem of identity authentication in transactions has not been well resolved. For example, fully electronic or virtualized information such as electronic seals or electronic signatures cannot provide secure physical evidence, which brings unprecedented difficulties to the legal arbitration that occurs after the transaction. How to balance the security and reliability of electronic certificates and the exchange of physical evidence is the most important problem. In the present invention, the electronic information of the paper feature information mixed contract is exchanged with each other, and the above-mentioned problems are well solved by adopting the method of printing out the physical evidences and keeping them. Fig. 9 is the application method of hybrid electronic contract or invoice, and the operation steps are described as follows:

(1) Party A takes an ordinary blank paper, and uses the paper identification device to obtain the two-dimensional code map of the paper characteristics of the paper. Party A edits the contract on the computer, inserts the digital seal or electronic signature of Party A, and inserts the two-dimensional code map of paper characteristics into the edited contract.

(2) For the first time, print the contract containing Party A's digital seal or electronic signature and two-dimensional code map of paper characteristics on the blank paper.

(3) Scan the contract printed in step (2) with a scanner to obtain a scanned electronic version.

(4) Party A transmits the scanned electronic version of the contract to Party B through the Internet.

(5) Party B receives the scanned electronic version of the contract from Party A through the Internet.

(6) Party B takes an ordinary blank paper and obtains the two-dimensional code map of the paper characteristics of the paper by the paper identification device.

(7) Party B inserts the characteristic two-dimensional code map of the paper into the electronic version of the contract sent by Party A, and edits Party B's own digital seal or electronic signature together into a new electronic version of the contract.

(8) Party B prints the newly synthesized electronic contract on the blank paper obtained in step (6) and keeps it as physical evidence.

(9) Party B transmits the newly synthesized electronic version of the contract to Party A through the Internet.

(10) Party A receives the new synthetic electronic version of the contract from Party B through the Internet.

(11) Party A prints the new synthetic electronic version of the contract on the paper printed in step (2) (the second printing) and keeps it as a physical evidence.

Through steps (1) to (11), it can be seen that both parties have not exchanged physical objects, but have obtained the feature code information of the other party's retained paper, and each has a physical certificate and a scanned electronic version to corroborate. Moreover, due to the method of scanning the contract content, the tampering of the contract content is effectively avoided. In this way, the security of electronic credentials, the reliability of identity authentication, and the exchange of physical evidence have all been solved at the same time.

Obviously, the above application method has the following advantages:

(a) Since the content transmitted through the network is a printed electronic scan, it is difficult to tamper with the content of the contract.

(b) The paper feature code guarantees the single validity of the contract, making the copied or scanned contract invalid. This is ideal for single contract applications such as electronic invoices and checks.

(c) Since the electronic seal or digital signature is not separated from the contract content, its security can be guaranteed.

(d) The arbitrator is not required to retain any physical evidence of the parties to the contract, and the physical evidence shall be retained by both parties. It can greatly reduce the responsibility and pressure of managers. Suitable for electronic invoice applications.

(e) The arbitration method is simple: the arbitrator requires both parties to show the physical certificate of the transaction. Even if one party refuses to show the certificate, it only needs to determine the authenticity of the physical certificate of the presenter (the method of determination is the same as that of application example 1), and can extract the electronic signatures or electronic seals of both parties. information, it can be judged that the party who presents the certificate wins the case.

The invention enlarges a specific area of ordinary paper through an optical microlens, collects an electronic image by a sensor, extracts paper fiber characteristics by matching a pre-designed standard template, encodes the extracted characteristics in a certain way, and forms a one-dimensional code of paper characteristics and a 2D coding map. This characteristic two-dimensional code map is directly printed on ordinary paper, and the transaction parties retain it as physical evidence. Because this coding map has the ability to store paper identification information, display the structural characteristics of the main fiber of the paper, mark the collection position of the fiber image, store the identification number of the paper identification device and the user ciphertext, etc., it can solve the reliability problem of the identity authentication of the transaction parties. . The invention is mainly applied to the security fields such as watermark technology against copy scanning attack, mixed electronic invoice, electronic contract and bank payment in Internet finance.

Claims (10)

1. a kind of display methods of paper fibre feature, which is characterized in that the steps included are as follows for the display methods:

1) electronic image of the specific region of paper is acquired；

2) the fibre image template for constructing multiple local features, using construction fibre image template to the electronic image of acquisition into Characteristic point corresponding greater than the electronic image of given threshold with local feature fibre image template similarity is chosen in the point-by-point matching of row As paper fibre image characteristic point；One standard fibers image template of every use, which carries out matching scanning, can all generate a feature Point bianry image, wherein characteristic point will form several internal connections connected domain isolated from one another in bianry image；Wherein, described Multiple local feature fibre image templates indicate a variety of different fiber imaging widths and a variety of different fiber extension directions；

3) obtained characteristic point is screened according to the priority of setting；

4) characteristic information of the characteristic point after screening is encoded, forms two-dimensional encoded figure corresponding with paper fibre feature, The feature viewing area of the code pattern shows symbol, the feature using characteristic point corresponding with the used Template Information of characteristic point Point display symbol is used to show the thickness information in fiber pattern direction at characteristic point position, fiber.

2. the display methods of paper fibre feature according to claim 1, which is characterized in that the two-dimensional encoded figure is also extremely It less include one in angle finger URL, acquisition position symbol, the quasi- finger URL of central campus, side calibration areas and feature one-dimensional coding area, The angle finger URL is used for one jiao of the two-dimensional encoded figure of locating paper sheets feature；The acquisition position symbol is for indicating paper The acquisition position of fibre image；The quasi- finger URL of the central campus is used for the encoded block position school of the two-dimensional encoded figure of paper features It is quasi-；The side calibration areas is used to correct the encoded block position in the two-dimensional encoded figure of paper features；The one-dimensional volume of the feature Code area is for showing feature one-dimensional coding.

3. the display methods of paper fibre feature according to claim 2, which is characterized in that the feature one-dimensional coding It is encoded for the characteristic information to the characteristic point screened, the encoded information of each characteristic point, which includes at least characteristic point, to be made Template number, abscissa and ordinate.

4. the display methods of paper fibre feature according to claim 1, which is characterized in that this method further includes to feature Information carries out coarse and standardization processing on scale: electronic image collected being carried out piecemeal, and is formed from top to bottom The new scale coordinate system of the serial number numerical value of piecemeal arrangement from left to right；It will be connected in the characteristic point bianry image obtained in step 2) The center in domain falls into image block area coordinate as new characteristic point position coordinate, obtained standardization characteristic point coordinate (X_τ, Y_τ) are as follows:

X_τ=a_τ-x₀,Y_τ=b_τ-y₀, τ=1 ... E

Wherein (x₀,y₀) be all blocking characteristic points centroid；(a_τ,b_τ) be the τ piecemeal piecemeal coordinate, E be sequence screen Remaining connected domain number afterwards.

5. the display methods of paper fibre feature according to claim 1, which is characterized in that the priority in step 3) is adopted With following rule:

A. the priority of the big characteristic point of template width is greater than the small characteristic point priority of template width；

B. the big priority of connected domain area locating for characteristic point is greater than the small characteristic point priority of connected domain area；

If c. connected domain area is identical, priority from top to bottom is determined according to the precedence of appearance from left to right.

6. a kind of display device of paper fibre feature, which is characterized in that the display device include acquisition module, matching module, Screening module and coding display module,

The acquisition module is used to acquire the electronic image of the specific region of paper；

The matching module is used to construct the fibre image template of multiple local features, utilizes the fibre image template pair of construction The electronic image of acquisition is matched point by point, chooses the electronics for being greater than given threshold with local feature fibre image template similarity The corresponding characteristic point of image is as paper fibre image characteristic point；It is every to carry out matching scanning using a standard fibers image template A characteristic point bianry image will be generated, wherein it is isolated from one another to will form several internal connections in bianry image for characteristic point Connected domain；Wherein, multiple local feature fibre image templates indicate a variety of different fiber imaging widths and it is a variety of not Same fiber extension direction；

The screening module is for screening obtained characteristic point according to the characteristic point priority of setting；

The coding display module is formed special with paper fibre for encoding to the characteristic information of the characteristic point after screening Corresponding two-dimensional encoded figure is levied, the feature viewing area of the code pattern uses characteristic point corresponding with the used Template Information of characteristic point Show symbol, the characteristic point shows that symbol is used to show the thickness letter in fiber pattern direction at characteristic point position, fiber Breath.

7. a kind of method for anti-counterfeit based on paper fibre feature, which is characterized in that the step of method for anti-counterfeit is as follows:

1) electronic image of the specific region of paper is acquired；

2) the fibre image template for constructing multiple local features, using construction multiple local features fibre image template to adopting The electronic image of collection is matched point by point, chooses the electronic chart for being greater than given threshold with local feature fibre image template similarity As being used as paper fibre image characteristic point；One standard fibers image template of every use, which carries out matching scanning, can all generate a spy Sign point bianry image, wherein characteristic point will form several internal connections connected domain isolated from one another in bianry image；Wherein, institute The multiple local feature fibre image templates stated indicate a variety of different fiber imaging widths and a variety of different fiber extension sides To；

3) obtained characteristic point is screened according to the priority of characteristic point；

4) characteristic point after screening is encoded, forms two-dimensional encoded figure corresponding with paper fibre feature, the code pattern Feature viewing area shows symbol using characteristic point corresponding with the used Template Information of characteristic point, and the characteristic point shows symbol For showing the thickness information in fiber pattern direction at characteristic point position, fiber；

5) by obtain it is two-dimensional encoded print on shielded paper, compare and print on paper features two dimension on paper and compile Feature viewing area in code figure proves that the paper is without over-scanning and answering if display symbolic construction having the same combines Paper after print.

8. the method for anti-counterfeit according to claim 7 based on paper fibre feature, which is characterized in that the two-dimensional encoded figure Also include at least one in angle finger URL, acquisition position symbol, the quasi- finger URL of central campus, side calibration areas and feature one-dimensional coding area , the angle finger URL is used for one jiao of the two-dimensional encoded figure of locating paper sheets feature；The acquisition position symbol is for indicating The acquisition position of paper fibre image；The quasi- finger URL of the central campus is used for the encoded block position of the two-dimensional encoded figure of paper features Set calibration；The side calibration areas is used to correct the encoded block position in the two-dimensional encoded figure of paper features；The feature one Dimension code area is for showing feature one-dimensional coding.

9. the method for anti-counterfeit according to claim 7 based on paper fibre feature, which is characterized in that this method further includes pair Characteristic information carries out coarse and standardization processing on scale: electronic image collected being carried out piecemeal, and is formed from upper The new scale coordinate system of the serial number numerical value of piecemeal arrangement from left to right under；It will be in the characteristic point bianry image that obtained in step 2) The center of connected domain falls into image block area coordinate as new characteristic point position coordinate, obtained standardization characteristic point coordinate (X_τ, Y_τ) are as follows:

X_τ=a_τ-x₀,Y_τ=b_τ-y₀, τ=1 ... E

Wherein (x₀,y₀) be all blocking characteristic points centroid；(a_τ,b_τ) be the τ piecemeal piecemeal coordinate, E be sequence screen Remaining connected domain number afterwards.

10. the method for anti-counterfeit according to claim 7 based on paper fibre feature, which is characterized in that in the step 3) Priority follow following criterion:

A. the priority of the big characteristic point of template width is greater than the small characteristic point priority of template width；

B. the big priority of connected domain area locating for characteristic point is greater than the small characteristic point priority of connected domain area；

If c. connected domain area is identical, priority from top to bottom is determined according to the precedence of appearance from left to right.