CN103903039A - Hidden code holographic anti-counterfeiting film and manufacturing method and recognition system thereof - Google Patents

Abstract

The invention relates to the field of anti-counterfeiting technology, and specifically discloses an invisible coded holographic anti-counterfeiting film, which includes an anti-counterfeiting information layer, and the anti-counterfeiting information layer includes holographic anti-counterfeiting graphics and invisible codes. The invention also discloses a method for making an anti-counterfeiting information layer in the anti-counterfeiting film; and a device for scanning and identifying the stealth coded holographic anti-counterfeiting film of the invention, an intelligent terminal device, a data server and an identification system. The stealth coded holographic anti-counterfeiting film of the present invention can be produced by a mature holographic molding process with low cost and has double anti-counterfeiting effects: its holographic anti-counterfeiting graphics are suitable for distinguishing authenticity by naked eyes; The decoding procedure reveals the information hidden by the invisible code.

Description

Invisible coding holographic anti-counterfeiting film, its manufacturing method and identification system

technical field

The invention relates to the field of anti-counterfeiting technology, in particular to an invisible coding holographic anti-counterfeiting film, a manufacturing method thereof, and an identification system.

Background technique

Holographic anti-counterfeiting trademarks are helpful to combat counterfeit and shoddy products, and have a wide range of applications, such as the packaging of various wine products and other high-end goods. However, the holographic anti-counterfeiting film currently used in the market has experienced years of development, and it is increasingly losing credibility in terms of structure and anti-counterfeiting strength.

In recent years, with the rapid popularization of smart phones, especially the popularity of Apple mobile phones, it has been realized to use the open programmable module of smart phones to read the coded information on the holographic anti-counterfeiting trademark to display more product information.

Combining holographic anti-counterfeiting technology with encoded information that can be decoded by smartphones to form a double anti-counterfeiting technology with invisible coding holographic anti-counterfeiting film will be a direction of future anti-counterfeiting technology.

Contents of the invention

The present invention aims to overcome the above-mentioned and other problems of the prior art, and combines the holographic anti-counterfeiting technology with the smart phone coding technology to provide an invisible coded holographic anti-counterfeiting film and an identification system with double anti-counterfeiting functions.

The first aspect of the present invention provides an invisible coded holographic anti-counterfeiting film, including an anti-counterfeiting information layer, the anti-counterfeiting information layer contains holographic anti-counterfeiting graphics and texts, and the holographic anti-counterfeiting graphics and texts are overlapped, overlapped and/or separated. Invisible coding.

The invisible code may be a discrete code point output by using a programmable interface open to the smart device to write a program, and the hidden information of the invisible code may be displayed by using a decoding program of the smart terminal device.

The hidden information may include text, images, audio files, video files, or a combination thereof.

The holographic anti-counterfeiting text may be a holographic text, a barcode, a two-dimensional code, or a combination thereof.

The second aspect of the present invention provides a method for making an anti-counterfeiting information layer of an invisible coded holographic anti-counterfeiting film, comprising steps: S1 writes a program through a programmable interface open to smart devices, and outputs discrete coding points; S2 converts the discrete Coding points and holographic anti-counterfeiting graphics are drawn together; S3 uses holographic photography or photolithography technology to make photoresist plates with invisible coding holographic anti-counterfeiting graphics; S4 uses electroforming technology to make holographic molded nickel plates; and S5 uses hot pressing Replication technology, embossing the stealth coded holographic anti-counterfeit text on the holographic molded nickel plate onto the information bearing layer to form an anti-counterfeit information layer.

A third aspect of the present invention provides a device for scanning and identifying the above-mentioned stealth coded holographic anti-counterfeiting film, including: a video capture module for scanning the invisible coded holographic anti-counterfeiting film to complete video capture; and a hidden code analysis module for Perform image preprocessing on the image obtained by scanning the invisible coding holographic anti-counterfeiting film by the video capture module, analyze the hidden code in the preprocessed image, encrypt the analyzed hidden code pattern, and send it to a data server through the wireless network for retrieval and match.

The video capture module can further scan the invisible coded holographic anti-counterfeiting film in the following manner to complete video capture: (a) select a video capture device, define various parameters of the video capture device, and set a preview callback function, thereby starting the video capture device , and set a timer; (b) the video capture device fixes the focus once every set time, and after the focus is successful, it takes an image for the hidden code analysis module to analyze, and the video capture device stops taking pictures but continues to capture and focus the video ; (c) If the hidden code analysis module parses successfully, then end; otherwise, return to execute (b).

The hidden code analysis module may further include: an image preprocessing unit, which is used to perform homomorphic filtering, histogram equalization and Canny operator edge detection on the image obtained by scanning the invisible coding holographic anti-counterfeiting film by the video capture module, so that the image The hidden code pattern and noise point distribution are highlighted in the middle; the hidden code pattern analysis unit is used to extract the hidden code pattern from the noise and identify it, including: using the Markov random field model for binary images to scan the video capture module invisible The image obtained by encoding the holographic anti-counterfeiting film contains noise in the hidden code pattern denoising, and the machine learning method of linear multi-class support vector machine is used for model training to obtain a classification function, and the 10 cardinal numbers of the hidden code image are identified by using the classification function. The denoised image is scanned to obtain a sequence of hidden code patterns; the hidden code mode encryption unit is configured to encrypt the analyzed hidden code patterns, including: encrypting the character strings of the hidden code patterns by using a private key encryption algorithm.

The device may also include a display execution module for receiving retrieved and matched item information returned by the data server through the wireless network, and displaying the item information on a display screen.

The fourth aspect of the present invention provides an intelligent terminal device for scanning and identifying the invisible coded holographic anti-counterfeiting film of the present invention, the intelligent terminal device includes the above-mentioned device for scanning and identifying the invisible coded holographic anti-counterfeiting film of the present invention, the The intelligent terminal equipment is provided with a video capture device, and the intelligent terminal equipment has a wireless Internet access function.

The fifth aspect of the present invention provides a data server for retrieving and matching the invisible coded holographic anti-counterfeiting film of the present invention, the data server includes: a database module for storing service classification codes and items corresponding to the service classification codes Information; a data service module, used to receive an encrypted hidden code pattern sent by a certain device or equipment through a wireless network, and decrypt the hidden code pattern after error checking through a cyclic redundancy check code, and retrieve the data in the database module The service classification code matches the code adapted to the hidden code mode from the service classification code and obtains the item information corresponding to the code, and feeds back the retrieved and matched result information to the device or equipment through the wireless network.

The sixth aspect of the present invention provides a system for identifying the invisible coding holographic anti-counterfeiting film of the present invention, the system includes the above-mentioned intelligent terminal device for scanning and identifying the invisible coding holographic anti-counterfeiting film and the above-mentioned invisible coding holographic anti-counterfeiting film. A data server for retrieval and matching; wherein, the smart terminal device is used to take an image of the invisible coded holographic anti-counterfeiting film, analyze the hidden code pattern in the image, encrypt the hidden code pattern obtained by the analysis, and send it through the wireless network To the data server for retrieval and matching; the data server is used to receive the encrypted hidden code pattern sent by the intelligent terminal device through the wireless network, retrieve the service classification code from the database module, and match the service classification code from the service classification code. The hidden code mode adapts the code and obtains the item information corresponding to the code, and feeds the retrieved and matched item information back to the device or equipment through the wireless network.

Preferably, the wireless network of the present invention may be a GPRS or 3G network.

The invisible coding holographic anti-counterfeiting film of the present invention has double anti-counterfeiting effects, is beautiful and is suitable for public anti-counterfeiting; its hidden information can be displayed through a decoding program of a handheld device (such as a smart phone). It can be produced by mature holographic molding process, and the production cost is low. By attaching randomly generated hidden codes (hash code points containing hidden information) to the holographic anti-counterfeiting graphics, the needs of anti-counterfeiting and traceability can be solved at the same time, and the identification process is simple.

Description of drawings

Fig. 1A to Fig. 1D schematically show different forms of the anti-counterfeiting information layer in the stealth coded holographic anti-counterfeiting film of the present invention.

Fig. 2A and Fig. 2B respectively schematically show the cross-sectional structure of the invisible coded holographic anti-counterfeiting film according to the present invention.

Fig. 3 is a flow chart of the manufacturing process of the anti-counterfeiting information layer of the invisible coded holographic anti-counterfeiting film of the present invention.

Fig. 4 schematically shows a system for identifying the stealth coded holographic anti-counterfeiting film of the present invention.

Reference signs: base film layer 1, information carrier layer 2, anti-counterfeiting information layer 21, high-reflection coating layer 3, hot-melt pressure-sensitive adhesive layer 41, adhesive layer 42, release layer 5, holographic anti-counterfeiting graphic 211, invisible Code 212

Detailed ways

The present invention will be described in further detail below in conjunction with the accompanying drawings and specific embodiments.

The present invention combines holographic anti-counterfeiting technology with smart phone coding technology, upgrades the existing holographic anti-counterfeiting film, and adds discrete code points containing invisible codes to holographic anti-counterfeiting graphics to provide the invisible coded holographic anti-counterfeiting film of the present invention. The anti-counterfeit film includes an anti-counterfeit information layer 21 . The information layer 21 is a grating pattern formed by thermal embossing of a holographic molded nickel plate, which includes a general holographic anti-counterfeit pattern 211 and an invisible code 212 .

Referring first to FIG. 1A to FIG. 1D , four different arrangements of anti-counterfeiting information layers are schematically shown respectively. It can be seen from the figure that the holographic anti-counterfeit text 211 can be a holographic text, a barcode, a two-dimensional code, etc., and can also be a combination of these types. The anti-counterfeiting information layer 21 includes a holographic anti-counterfeiting graphic 211 and an invisible code 212, and the invisible coding 212 and the holographic anti-counterfeiting graphic 211 can be overlapped, overlapped and/or separated.

The holographic anti-counterfeiting graphic 211 can diffract out a holographic anti-counterfeiting image with rich colors, dynamic changes, multi-channel and multi-dimensional light-changing effects, and can distinguish authenticity with dazzling diffracted light changes with the naked eye, and is suitable for public anti-counterfeiting. The invisible code 212 is hidden in the anti-counterfeit information layer 21, which can be, for example, discrete code points, so that when it overlaps with the holographic anti-counterfeit text 211, it will not interfere with the holographic anti-counterfeit text 211 and will not affect the appearance of the product.

The discrete code point can be programmed using a programmable interface open to smart devices, the coded information output, and its hidden information can be displayed through the decoding program of the smart device (such as a handheld smart device such as a smart phone), so as to achieve double anti-counterfeiting effect. These hidden information can be, for example, text, images, voice files, video files, etc., or a combination thereof.

With reference to Fig. 2A and Fig. 2B, the sectional structure of two specific forms of invisible coding holographic anti-counterfeiting film of the present invention is shown schematically, wherein Fig. 2A shows the invisible coding holographic anti-counterfeiting film in bronzing film form; Fig. 2B shows Stealth coded holographic security film in the form of a composite film (or full transfer film).

When the stealth coded holographic anti-counterfeiting film of the present invention is bronzing film (Fig. 2A), it includes base film layer 1, release layer 5, information carrier layer 2, anti-counterfeiting information layer 21, coating layer 3 and thermal The pressure sensitive adhesive layer 41 is melted.

The base film layer 1 is a coated carrier, and a plastic film can be used as the base film layer. Other coatings are evenly attached to the base film layer 1 . The release layer 5 is attached to the base film layer 1, and its raw material is usually wax with a melting point of 80-90°C. The release layer 5 plays a role of dissociation in the use of bronzing film.

The information carrier layer 2 is used to carry holographic information, and the information carrier layer 2 is formed by transferring the coating liquid to the base film layer 1 through the coating equipment and drying it. The anti-counterfeit information layer 21 can be transferred onto the information carrier layer 2 by hot embossing technology.

The high reflection coating layer 3 is attached on the anti-counterfeiting information layer 21, which is usually a metal or medium with high reflection performance, and can be formed by vacuum evaporation. The presence of the coating layer 3 can improve the brightness of the holographic film, making it convenient to observe the holographic anti-counterfeit image directly by using reflected light. The hot-melt pressure-sensitive adhesive layer 4 is the outermost layer of the invisible coding holographic anti-counterfeiting bronzing film, and its function is to adhere to the surface of the substrate when the product is in use.

When the stealth coded holographic anti-counterfeiting film of the present invention is a composite film (or full transfer film) (Figure 2B), it includes a base film layer 1, an information carrier layer 2, an information layer 21, and a high-reflection coating layer 3 from bottom to top. And back adhesive layer 42. As can be seen from Figure 2B, the stealth coded holographic anti-counterfeiting composite film (or full transfer film) of the present invention is similar to the bronzing film shown in Figure 2A, but does not include a release layer, and when it is a composite film, the base film layer 1 is It is not separated during use; and when it is a full transfer film, the base film layer 1 is separated during the composite transfer process and can be recycled and reused.

The stealth coded holographic anti-counterfeiting film of the present invention can generally be produced by a mature holographic molding process, and the cost is relatively low. The feature of the present invention lies in the manufacture of the anti-counterfeiting information layer. Referring to FIG. 3 , it shows the process flow for making the anti-counterfeiting information layer in the invisible coding holographic anti-counterfeiting film. As can be seen from the figure, the production method mainly includes the following steps:

First, in step S1, a program is written through the open programmable interface of the smart phone to output invisible codes, such as discrete code points. Then in step S2, the invisible coding (discrete coding points) and the holographic anti-counterfeiting graphics are drawn together; and in step S3, a photoresist plate with invisible coding holographic anti-counterfeiting graphics is produced by using holographic photography or photolithography technology; Then in step S4, use electroforming technology to make a holographic molded nickel plate; then use hot-press printing technology to print the invisible coded holographic anti-counterfeit image on the holographic molded nickel plate to the information bearing layer to form an anti-counterfeit information layer .

Those skilled in the art will understand that, depending on the specific form of the anti-counterfeiting film, appropriate operations may be added before, during and after the above preparation process steps. For example, in the case of bronzing film, the information bearing layer 2 needs to be coated on the base film layer 1 before embossing and copying the holographic anti-counterfeiting images in batches. After the anti-counterfeit information layer 21 is formed, utilize vacuum coating technology to plate a highly reflective metal film or dielectric film on the anti-counterfeit information layer 21 to form a high-reflective coating layer 3; coat a hot-melt pressure-sensitive adhesive layer 41 or a back glue layer 42 to High reflection coating layer 3. Finally, the anti-counterfeiting film can be cut to be applied to the product.

The products produced in this way can diffract out holographic anti-counterfeiting images with rich and beautiful colors, dynamic changes, multi-channel and multi-dimensional light-changing effects, which can not only distinguish the authenticity with dazzling diffractive light changes with the naked eye, but are also suitable for public anti-counterfeiting; the information layer also hides Encoding, the hidden information can be displayed through the decoding program of the handheld device (smart phone), such as text, image, voice file or video file, etc. It has double anti-counterfeiting effects and can be applied to anti-counterfeiting trademarks and other fields.

In the specific use of bronzing film of the present invention, because bronzing film is subjected to high temperature and certain pressure when bronzing, is subjected to the effect of temperature and pressure, release layer 5 begins melting to be transformed into liquid at first as the part with the lowest melting point in each layer, Under the action of the peeling force, this layer is first destroyed, so that the information carrier layer 2 carrying the information layer 21 is separated from the base film layer 5 . As the temperature and pressure increase, the hot-melt pressure-sensitive adhesive layer 4 begins to soften and become viscous. When the adhesive force of the adhesive layer is greater than the bonding force of the melted wax layer, the bronzing material is transferred to the substrate, and the base film Layer 1 is peeled off, enabling the complete foil stamping process.

In the specific use of the composite film (or full transfer film) of the present invention, the adhesiveness of the back adhesive layer 4 is used to compound and mature the substrate with holographic information and another substrate using composite equipment. Thereby transferring the holographic security film with invisible coding to the substrate. As mentioned above, when the holographic anti-counterfeiting film of the present invention is a composite film, the base film layer 1 will not be separated during the compounding process; and when the holographic anti-counterfeiting film of the present invention is a full transfer film, the base film layer will separated and can be recycled.

Referring to FIG. 4 , it shows a system according to the present invention, using a hand-held device (such as a smart phone) to identify the stealth coded holographic anti-counterfeiting film of the present invention to decode and reveal hidden information. The system mainly includes two parts: an intelligent terminal device (such as a smart phone), which includes a device for scanning and identifying the invisible coding holographic anti-counterfeiting film of the present invention, and a device for retrieving and identifying the invisible coding holographic anti-counterfeiting film of the present invention. matching data server.

Described below respectively.

The embodiment of the present invention also provides a device for scanning and identifying the above-mentioned stealth coded holographic anti-counterfeiting film, the device includes a video capture module and a hidden code analysis module, wherein the video capture module is used to scan the invisible coded holographic anti-counterfeiting film to complete video capture; The hidden code analysis module is used to perform image preprocessing on the image obtained by scanning the invisible coding holographic anti-counterfeiting film by the video capture module, and analyze the hidden code in the preprocessed image, encrypt the analyzed hidden code mode and send it through the wireless network to a data server for retrieval and matching. The wireless network can adopt General Packet Radio Service (GPRS) network or 3G network.

The video capture module can be applied to smart operating systems such as Android, iOS, and Windows Phone (WP for short), covering smartphones, laptops, tablets and other devices with cameras on the market. Its main function is to establish a complete A video capture platform that guides users to use the camera to scan materials of interest (including QR codes, 1D barcodes or laser anti-counterfeiting labels, etc.). The video capture module can scan the invisible coding holographic anti-counterfeiting film in the following ways to complete video capture:

(a) Select a video capture device (such as a camera of a smart terminal device), define various parameters of the camera such as whether to preview, image format, preview frame number, preview size, etc., and then set a preview callback function, receive and analyze in the callback function Each frame, start the preview process, thus start the video capture device, and set the timer;

(b) The camera fixes the focus once every set time. The set time is set by the above timer. For example, it can be set to focus every 1 second. After the focus is successful, take an image, and then start another thread to The image pointer points to the hidden code analysis module, which is analyzed by the hidden code analysis module; at the same time, the camera thread (main thread) sets the status word (for example, setting the status word to 1 or other values except 0) so that the camera stops taking pictures, but the camera continues to video Capture and focus tasks, waiting for processing by the hidden code analysis module;

(c) When the hidden code analysis module thread ends, the result of analysis success or failure may be obtained. If the hidden code analysis module parses successfully, that is, successfully obtained the hidden code mode, then end the work of the video capture module; if the hidden code mode cannot be obtained, remind the user to try again, and return to execution (b). Regardless of the above circumstances, when the user returns to the video capture interface, the main thread clears the status word (make the status word 0), and restarts the process of focusing and taking pictures at intervals of 1 second.

The hidden code analysis module preprocesses the image scanned by the video capture module, obtains the custom random code word distributed in the image, recognizes and analyzes the custom random code word pattern, uses the key to encrypt the custom code word, and calculates Generate fixed-length ciphertext and send the ciphertext to the data server in the cloud through the wireless network. The cloud data server decrypts the received ciphertext, quickly retrieves the service classification code in the server, matches the applicable code and feeds back the corresponding information, and then uses the wireless network (GPRS or 3G) to send the result information to the user's smart terminal device. Users can identify the authenticity and traceability of commodities based on the corresponding information.

Specifically, the hidden code analysis module includes an image preprocessing unit, a hidden code mode analysis unit and a hidden code mode encryption unit. After focusing, the image obtained by taking a photo is easily affected by light and shadow changes, weather, and the photographer, and the image needs to be preprocessed by the image preprocessing unit. According to the characteristics of the image, this embodiment successively performs homomorphic filtering, histogram equalization, and Canny operator edge detection on the image, so that hidden code patterns and noise distribution are highlighted in the image. The three image processing processes of homomorphic filtering, histogram equalization and Canny operator edge detection are further explained below:

(1) Homomorphic filtering:

The dynamic range of the image obtained by taking pictures is very large, but the gray scale of the interested part is very dark, and the image details cannot be identified. The general gray scale linear transformation method does not work well. The homomorphic filtering of images belongs to the category of image frequency domain processing. Its function is to adjust the gray scale range of the image. By eliminating the problem of uneven illumination on the image, the image details in dark areas are enhanced without losing image details in bright areas.

The image f(x, y) of a natural scene can be represented by the product of the illumination function f _i (x, y) and the reflection function f _r (x, y). f _i (x, y) describes the lighting of the scene and has nothing to do with the scene; f _r (x, y) contains the details of the scene and has nothing to do with the lighting.

f(x, y) = f _i (x, y)*f _r (x, y)

0<f _i (x, y)<∞;0<f _r (x, y)<1 (1)

Since the two are multiplied together, it cannot be transformed into the frequency domain and processed separately, so the following processing is done:

Take the logarithm of formula (1)

lnf (x, y) = lnf _i (x, y) + lnf _r (x, y) (2)

Make it an additive relationship in the space domain, and take the Fourier transform of formula (2)

F _ln (u, v) = F _{i, ln} (u, v) + F _{r, ln} (u, v) (3)

In the formula, F _{i, ln} : the illumination function changes slowly in space, and its spectral characteristics are concentrated in the low frequency band; F _{r, ln} : the spectrum of the reflection function is concentrated in the high frequency band (the image has more details and edges), and the reflection function The scene described by the function reflects the details of the image, and its frequency is in the high-frequency region.

If the image is not evenly illuminated, the average brightness of the various parts of the image will fluctuate. The image detail structure corresponding to the dark area is more difficult to distinguish, and this inhomogeneity needs to be eliminated. The grayscale range of the illumination function can be compressed, that is, the components of the illumination function can be weakened in the frequency domain, and the spectral components of the reflection function can be enhanced at the same time, so that the contrast of the reflection function reflecting the contrast of the image can be increased. As a result, image detail in dark areas of the image is increased, and image detail in bright areas is kept as large as possible. Multiplied by the transfer function H(u, v) (homomorphic filter), the low frequency band is compressed, while the high frequency band is expanded.

G _ln (u, v) = F _ln (u, v)*H (u, v)

= G _{i, ln} (u, v) + G _{r, ln} (u, v) (4)

Then find the inverse Fourier transform, and get the expression in the corresponding space domain

g(x, y) = exp{F ^-1 {G _ln (u, v)}} (5)

According to different image characteristics and needs, different H(u, v) is selected, and after the image with poor contrast and indistinct details is processed with a homomorphic filter, the brightness of the image is relatively uniform, and the details are enhanced.

(2) Histogram equalization:

After homomorphic filtering, the image is converted into a grayscale image. Since the grayscale values of many images are non-uniformly distributed, it is very common for images whose grayscale values are concentrated in a small interval. Histogram equalization is a method of enhancing image contrast by redistributing the gray values evenly. The image after histogram equalization is very beneficial to the selection of binarization threshold.

This embodiment uses image scale transformation: the pixels in the gray interval [a, b] are mapped to the interval [z ₁ , z _k ]. In general, the gray interval [a, b] of the original image is often a subspace of the space [z ₁ , z _k ], at this time, the pixel point z in the original interval is mapped to the function of the pixel point z′ in the new interval Expressed as

${\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}{\mathrm{<span\; class="notranslate">\; b</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; a</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; z</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 6</span>}<span\; class="notranslate">\; )</span>$

The above mapping relationship actually extends the interval [a, b] to the interval [z ₁ , z _k ], making the black image blacker and the white image whiter.

(3) Edge detection:

The position of hidden code and noise in the image needs to be extracted, that is, the boundary of the area where the gray level changes sharply in the image. In fact, this is a typical edge detection problem. The essence of edge detection is to use a certain algorithm to extract the boundary line between the object and the background in the image. The Canny operator uses the neighborhood of the pixel to be processed as the basis for grayscale analysis to realize the extraction of image edges and Good processing effect has been obtained. Its basic steps are:

1. Filtering. Edge detection is mainly based on derivative calculations, but is affected by noise. Filters also cause a loss of edge strength while reducing noise.

2. Enhancement. The enhancement algorithm highlights the points with significant gray changes in the neighborhood. This is typically done by computing the gradient magnitude.

3. Detection. In some images, the larger gradient amplitude is not the edge point. The simplest edge detection is gradient magnitude thresholding.

4. Positioning. Precisely determine the position of the edge.

The specific algorithm steps of the Canny operator to find the edge point are as follows:

1. Smooth the image with a Gaussian filter.

2. Calculate the magnitude and direction of the gradient using the first-order partial derivative finite difference.

3. Perform non-maximum suppression on the gradient amplitude.

4. Detect and connect edges with a double-threshold algorithm.

The preprocessed image can highlight the hidden code pattern and the distribution of noise points, and the ultimate purpose of this embodiment is to extract and identify the hidden code pattern from the noise points. The hidden code word in this embodiment contains 10 bases, and every 30 bits constitutes a hidden code pattern, and the number of all hidden code patterns can be calculated to be ¹⁰³⁰ . In order to obtain a hidden code pattern containing noise, it is first necessary to denoise the hidden code image, then use the classification function obtained by machine learning to identify the above 10 bases, and finally scan the image to determine whether it contains the corresponding hidden code pattern.

Therefore, in this embodiment, the hidden code pattern is extracted from the noise by the hidden code pattern analysis unit and identified, including: using the Markov random field model for the binary image to scan the video capture module for the invisible coding holographic anti-counterfeiting film The hidden code pattern containing noise in the obtained image is denoised, the machine learning method of linear multi-class support vector machine is used for model training to obtain the classification function, and the classification function is used to identify the 10 bases of the hidden code image, and the denoised image is processed Scan for a sequence of cryptographic patterns. In addition, the hidden code mode obtained through the analysis is encrypted by the hidden code mode encryption unit, for example, a character string in the hidden code mode is encrypted using a private key encryption algorithm. The processing process of hidden code image denoising and hidden code cardinality recognition is further explained as follows:

(1) Hidden code image denoising:

The image denoising here is aimed at hidden code patterns containing noise. In consideration of efficiency, this embodiment adopts a Markov random field model for binary images. Assuming that the amount of noise is less than that of the real hidden code pattern, it can be considered that any point _xi in the noise-free hidden code pattern has a strong correlation with the corresponding point y _i in the noise-containing hidden code pattern, and also It can be considered that the adjacent points x _i and x _j in the noise-free hidden code mode have a strong correlation. We define such pairs of points with strong correlation as a group of neighborhoods, so we have two types of neighborhoods: ( _xi , y _i ) and ( _xi , x _j ). Put these two kinds of prior knowledge into the energy function, for ( _xi , y _i ), the energy function takes _-ηxi y _i , η is a positive constant, for ( _xi , x _j ), the energy function Take -γx _i x _j , where γ is a positive constant. In addition, for the hidden code mode of this embodiment, an energy function needs to be added to facilitate the appearance of a single point, so the complete form of the energy function is:

$\mathrm{<span\; class="notranslate">\; E.</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; the\; y</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; h</span>}\underset{\mathrm{<span\; class="notranslate">\; i</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; \&gamma;</span>}\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; j</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; \&eta;</span>}\underset{\mathrm{<span\; class="notranslate">\; i</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 7</span>}<span\; class="notranslate">\; )</span>$

The energy function defines the joint probability distribution of all point vectors x and y:

$\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; the\; y</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; Z</span>}}\mathrm{<span\; class="notranslate">\; exp</span>}<span\; class="notranslate">\; \{</span><span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; E.</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; the\; y</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \}</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 8</span>}<span\; class="notranslate">\; )</span>$

In formula (8), Z is the normalization factor.

Since y is the observed point, formula (8) is actually equivalent to defining the conditional probability p(x|y), and the purpose of image denoising is to find the maximum possible probability value. The algorithm used in this system is graph-cuts, which can ensure global convergence and has high efficiency.

(2) Hidden code cardinal number recognition

In this embodiment, the hidden code pattern is composed of 10 basic image patterns. In order to effectively extract the pattern hidden in the image, it is necessary to perform model training in advance to obtain the discriminant function, and then detect the sequence of basic image patterns on the captured image. and encoding pattern sequences.

The machine learning algorithm used in model training in this embodiment is a multi-class support vector machine (Support Vector Machine, SVM). Support vector machine is a new machine learning method proposed by V.Vipnik et al. based on Statistical Learning Theory (SLT). It shows many unique advantages in solving nonlinear and high-dimensional pattern recognition problems. Good results have been achieved in recognition, function approximation and probability density estimation. The support vector machine is essentially a feed-forward neural network. According to the structural risk minimization criterion, the generalization ability of the classifier should be improved as much as possible on the premise of minimizing the classification error of the training samples. From the implementation point of view, the core idea of training support vector machine is equivalent to solving a quadratic programming problem with linear constraints, so as to construct a hyperplane as the decision plane, so that the distance between the two types of patterns in the feature space is the largest, and can guarantee The obtained solution is the global optimal solution.

The SVM method is proposed from the Optimal Hyperplane in the case of linear separability. The optimal classification surface requires that the classification line can not only separate the two types of samples without errors, but also maximize the distance between the two types.

Let the ^linearly separable sample set be (x _i , y _i ), i=1, 2, . The general form of the linear discriminant function in the d-dimensional space is: g(x)= ^ωT x+b, and the classification surface equation is:

ω ^T x + v = 0 (9)

Normalize the discriminant function so that all samples of the two categories satisfy |g(x)|≥1, that is, make g(x)|=1 of the sample closest to the classification surface, so that the classification interval is equal to 2/| |ω||, so the maximum interval is equivalent to making ||ω|| (or ||ω|| ² ) the minimum; and requiring the classification line to correctly classify all samples is to require it to satisfy:

y _i [ω ^T x+b]-1≥0, (i=1, 2, ..., n) (10)

Therefore, the classification surface that satisfies the above conditions and minimizes ||ω|| ² is the optimal classification surface. The training samples on the hyperplane that is closest to the classification surface and parallel to the optimal classification surface in these two types of samples are those samples that make the equal sign in formula (9) true, and they are called support vectors (Support Vectors). According to the above discussion, the optimal classification surface problem can be expressed as the following constrained optimization problem, that is, under the constraints of formula (10), find the function:

$\mathrm{<span\; class="notranslate">\; \&Phi;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; \&omega;</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; 2</span>}}{<span\; class="notranslate">\; |</span><span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; \&omega;</span>}<span\; class="notranslate">\; |</span><span\; class="notranslate">\; |</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 11</span>}<span\; class="notranslate">\; )</span>$

minimum value. This is a quadratic programming problem, and the optimal solution ω ^* , b ^* can be obtained. Then the optimal classification function is obtained as:

f(x)＝sgn{ω ^*T x+b ^* } (12)

Among them: sgn(.) is a symbol function, for a given unknown sample x, only need to calculate the formula (12) to determine the category to which x belongs.

To sum up, this embodiment uses a linear multi-class support vector machine to obtain a classification function after the training phase is completed. Scan the image area to obtain the hidden code pattern sequence, use the key to encrypt the string corresponding to the hidden code pattern (for example, use a private key encryption algorithm to encrypt), and send the ciphertext to the data service module to match and search for the corresponding pattern Item information.

In addition, the device also includes a display execution module for receiving retrieved and matched item information returned by the data server through the wireless network, and displaying the item information on a display screen.

Those skilled in the art should know that the device can be programmed to form an application program, such as an application program suitable for intelligent terminal equipment. After the application program is installed on the intelligent terminal equipment, the video capture device combined with the intelligent terminal equipment can realize scanning and identification The function of the invisible coded holographic anti-counterfeiting film mentioned above can also be superimposed on the video to display the authenticity of the product and other information fed back by the data server after the recognition is successful.

The embodiment of the present invention also provides an intelligent terminal device that scans and recognizes the above-mentioned invisible coded holographic anti-counterfeiting film. The application program is installed in the smart terminal device), the smart terminal device is equipped with a video capture device and has a wireless Internet access function, and the wireless Internet access method can adopt GPRS or 3G network. The smart terminal device may be a mobile smart device such as a smart phone, a notebook computer or a PDA, and generally has a video capture device such as a camera.

The embodiment of the present invention also provides a data server for retrieving and matching the invisible coded holographic anti-counterfeiting film, the data server includes a database module and a data service module, wherein the database module is used to store the service classification code and the corresponding The item information; the data service module is used to receive the encrypted hidden code mode sent by a device or equipment through the wireless network, and decrypt the hidden code mode after error checking through the Cyclic Redundancy Check code (Cyclic Redundancy Check), and retrieve the database module The service classification code in the service classification code matches the code adapted to the hidden code mode from the service classification code and obtains the item information corresponding to the code, and finally feeds back the searched and matched result information to the device or device for scanning and identification through the wireless network. With the increase of products, the records in the database will increase to an order of magnitude that seriously affects the retrieval speed. In this embodiment, a secondary index structure is used to improve query performance.

The embodiment of the present invention also provides a system for identifying the above invisible coded holographic anti-counterfeiting film, the system includes the above intelligent terminal device for scanning and identifying the invisible coded holographic anti-counterfeiting film and the above mentioned stealth coded holographic anti-counterfeiting film for retrieving and matching data server. Among them, the intelligent terminal device takes the image of the invisible coded holographic anti-counterfeiting film, analyzes the hidden code pattern in the image, encrypts the analyzed hidden code pattern and sends it to the data server through the wireless network for retrieval and matching; the data server receives the intelligence code through the wireless network. The encrypted hidden code pattern sent by the terminal device retrieves the service classification code from the database module, matches the code adapted to the hidden code mode from the service classification code and obtains the item information corresponding to the code, and retrieves and matches the code through the wireless network. Feedback of the item information to the device or equipment. Similarly, the wireless network can be GPRS or 3G network.

In specific applications, a random hidden code is added to the QR code, one-dimensional bar code or laser anti-counterfeiting label, and the two-dimensional code, one-dimensional bar code or laser anti-counterfeiting label is scanned by the user's smart phone, laptop, tablet computer and other smart terminal devices with cameras. Scan the anti-counterfeit label and take a sample, and the obtained picture is recognized on the smart terminal device. The smart terminal device analyzes the hidden code pattern corresponding to the item, and transmits the code of the hidden code mode to the data server through the wireless network, and obtains the item after string matching The corresponding information of the item is fed back from the data server to the user's smart terminal device through the wireless network again, thereby completing a recognition process.

The system can be downloaded in the form of APP (Application), and iOS, Android and WP versions are available. When in use, scan the video of the carrier containing the hidden code pattern, and the system will complete the process of auto-focusing, taking pictures, pattern analysis, and feeding back product information. For example: add microcode anti-counterfeiting technology to the holographic laser label, scan the label with a smart terminal device, and the APP decodes the scanned image to form a series of encrypted interval codes, which are then transmitted to the cloud data server through the wireless network for data matching , and return the corresponding data to the user's smart terminal device after the matching is successful.

The specific embodiments of the present invention described above do not constitute a limitation to the protection scope of the present invention. Any other corresponding changes and modifications made according to the technical concept of the present invention shall be included in the protection scope of the claims of the present invention.

Claims (13)

1. A stealth coded holographic anti-counterfeiting film, characterized in that it comprises an anti-counterfeiting information layer, the anti-counterfeiting information layer contains holographic anti-counterfeiting graphics and texts, and the invisible holographic anti-counterfeiting graphics and texts are overlapped, overlapped and/or separated. coding. 2. The invisible coding holographic anti-counterfeiting film according to claim 1, characterized in that, the invisible coding is a discrete coding point output by using a programmable interface open to smart devices, and can be programmed by using smart terminal devices The decoding procedure reveals the hidden information of the invisible code. 3. The stealth coded holographic anti-counterfeiting film according to claim 2, wherein the hidden information includes text, image, voice file, video file, or a combination thereof. 4. The stealth coded holographic anti-counterfeiting film according to claim 1, wherein the holographic anti-counterfeiting text is a holographic text, a barcode, a two-dimensional code, or a combination thereof. 5. A method for making an anti-counterfeiting information layer of an invisible coded holographic anti-counterfeiting film, characterized in that it comprises steps: S1 writes programs through the programmable interface open to smart devices, and outputs discrete code points; S2 draws the discrete coding points together with the holographic anti-counterfeiting graphics; S3 uses holographic photography or photolithography technology to make a photoresist plate with invisible coded holographic anti-counterfeiting graphics; S4 uses electroforming technology to make holographic molded nickel plates; and S5 uses hot-press replication technology to emboss the invisible coded holographic anti-counterfeit text on the holographic molded nickel plate onto the information bearing layer to form an anti-counterfeit information layer. 6. A device for scanning and identifying the invisible coding holographic anti-counterfeiting film according to any one of claims 1-4, characterized in that the device comprises: A video capture module, configured to scan the invisible coded holographic anti-counterfeiting film to complete video capture; The hidden code analysis module is used to perform image preprocessing on the image obtained by scanning the invisible coding holographic anti-counterfeiting film by the video capture module, analyze the hidden code in the preprocessed image, encrypt the analyzed hidden code mode and send it through the wireless network to a data server for retrieval and matching. 7. The device according to claim 6, wherein the video capture module further scans the invisible coded holographic anti-counterfeiting film in the following manner to complete video capture: (a) Select the video capture device, define the parameters of the video capture device, set the preview callback function to start the video capture device, and set the timer; (b) The video capture device fixes the focus once every set time, and after the focus is successful, takes an image for the hidden code analysis module to analyze, and the video capture device stops taking pictures but continues to capture and focus the video; (c) If the hidden code analysis module parses successfully, end; otherwise, return to execute (b). 8. The device according to claim 6, wherein the hidden code analysis module further comprises: The image preprocessing unit is used to perform homomorphic filtering, histogram equalization and Canny operator edge detection on the image obtained by scanning the invisible coding holographic anti-counterfeiting film by the video capture module, so as to highlight the hidden code pattern and noise distribution in the image; The hidden code pattern analysis unit is used to extract the hidden code pattern from the noise and identify it, including: using the Markov random field model for binary images to scan the video capture module to scan the invisible coding holographic anti-counterfeiting film. Denoising the hidden code pattern of the noise, using the machine learning method of linear multi-class support vector machine to carry out model training to obtain the classification function, using the classification function to identify the 10 cardinal numbers of the hidden code image, and scanning the denoised image to obtain a sequence of hidden patterns; The hidden code mode encryption unit is configured to encrypt the analyzed hidden code mode, including: using a private key encryption algorithm to encrypt the character string of the hidden code mode. 9. The device according to any one of claims 6-8, characterized in that the device further comprises a display execution module for receiving the retrieved and matched items returned by the data server through the wireless network information, and display the item information on a display screen. 10. An intelligent terminal device for scanning and identifying the stealth coded holographic anti-counterfeiting film according to any one of claims 1-4, characterized in that, the intelligent terminal device includes any one of claims 6-9 The smart terminal device is provided with a video capture device, and the smart terminal device has a wireless Internet access function. 11. A data server for retrieving and matching the invisible coded holographic anti-counterfeiting film according to any one of claims 1-4, characterized in that the data server comprises: A database module, configured to store service classification codes and item information corresponding to the service classification codes; The data service module is used to receive the encrypted hidden code pattern sent by a certain device or equipment through the wireless network, decrypt the hidden code pattern after error checking through the cyclic redundancy check code, and retrieve the service classification in the database module code, match the code that is adapted to the hidden code mode from the service classification code and obtain the item information corresponding to the code, and feed back the retrieved and matched result information to the device or equipment through the wireless network. 12. A system for identifying the invisible coded holographic anti-counterfeiting film according to any one of claims 1-4, characterized in that the system includes the scanning and identifying invisible coded holographic anti-counterfeiting film according to claim 10 The intelligent terminal device and the data server for retrieving and matching the invisible coded holographic anti-counterfeiting film according to claim 11; wherein, The smart terminal device is used to take the image of the invisible coding holographic anti-counterfeiting film, analyze the hidden code pattern in the image, encrypt the analyzed hidden code pattern and send it to the data server through the wireless network for retrieval and matching ; The data server is used to receive the encrypted hidden code pattern sent by the intelligent terminal device through the wireless network, retrieve the service classification code from the database module, match the code adapted to the hidden code pattern from the service classification code, and obtain the code For the corresponding item information, the retrieved and matched item information is fed back to the device or equipment through the wireless network. 13. The device according to any one of claims 6-9, the smart terminal device according to claim 10, the data server according to claim 11 or the system according to claim 12, wherein the The wireless network is GPRS or 3G network.

Images