Anti-counterfeiting electronic device, anti-counterfeiting system and anti-counterfeiting method

Abstract

The invention discloses an anti-counterfeiting electronic device, an anti-counterfeiting system and an anti-counterfeiting method, which are used in the field of anti-counterfeiting, and the anti-counterfeiting method comprises the following steps: storing a reflection image set of a rough surface of a registered object; acquiring a reflection image of the surface of the to-be-detected object under current light irradiation; calculating a grey-scale map or a gradient map of at least one specified fixed area of the reflection image of the surface of the to-be-detected object; calculating a reconstructed image of a gray domain or a reconstructed image of a gradient domain of the specified fixed area of the reflection image set under the current light irradiation; calculating a similarity value of AND or AND; and verifying whether the object to be verified is from the registered object according to the similarity value.

Description

Technical Field

[0001] This invention relates to the field of anti-counterfeiting, and in particular to an anti-counterfeiting electronic device, anti-counterfeiting system, and anti-counterfeiting method. Background Technology

[0002] Counterfeit and imitation goods not only cause huge economic losses to legitimate brand owners but also seriously infringe upon the legitimate interests of consumers. Although numerous new anti-counterfeiting technologies have emerged in the market over the past few decades—such as artificially creating special anti-counterfeiting features on products, including special inks, materials, and printing processes—in an attempt to prevent counterfeiting by keeping these special processes secret, it has proven that secrecy is simply insufficient to stop counterfeiters from stealing or imitating.

[0003] Existing anti-counterfeiting technologies rely on consumers' "observation with the naked eye and judgment with the brain" to identify the authenticity of goods. However, ordinary consumers find it difficult to learn and master the wide variety of anti-counterfeiting knowledge for different products, and their visual discrimination ability is also difficult to reach the level of being able to identify counterfeit products that are indistinguishable from genuine ones.

[0004] Because various anti-counterfeiting labels are easily counterfeited, many brands have to use a combination of different anti-counterfeiting measures on a single product (commonly known as "comprehensive anti-counterfeiting") to increase the difficulty of counterfeiting. However, this undoubtedly requires consumers to have the knowledge and skills to identify multiple anti-counterfeiting technologies of the purchased product. Since the vast majority of consumers simply cannot possess such complex anti-counterfeiting knowledge and skills, this "comprehensive anti-counterfeiting" has become an ineffective decoration.

[0005] Furthermore, the widespread availability of image scanning and printing equipment has made it incredibly easy for anyone to copy QR code traceability labels. Most counterfeit products found on the market bear traceability QR codes identical to genuine products, thus causing QR code traceability labels to gradually lose consumer trust.

[0006] Existing RFID chip anti-counterfeiting technology uses a string of numbers to store data, but this is easily cloned and does not have anti-counterfeiting capabilities. RFID chips with encryption or anti-counterfeiting capabilities require a CPU, which makes them expensive to manufacture and difficult to popularize in the field of anti-counterfeiting for ordinary goods.

[0007] Of course, there are also anti-counterfeiting technologies that use "second-line anti-counterfeiting". "Second-line anti-counterfeiting" refers to anti-counterfeiting technology that requires the use of professional experimental equipment to identify authenticity. Although this technology increases the difficulty for counterfeiters, it cannot effectively protect consumers because consumers cannot equip themselves with such professional equipment and the verification experiment is expensive.

[0008] For a long time, the anti-counterfeiting industry and academia have made unremitting efforts to try to solve the problem of the poor effectiveness of the aforementioned anti-counterfeiting technologies. However, due to the high difficulty, complexity and challenge of anti-counterfeiting technology as a global problem, anti-counterfeiting technology has not yet broken through the aforementioned predicament. Summary of the Invention

[0009] The main objective of this invention is to provide an anti-counterfeiting electronic device, anti-counterfeiting system, and anti-counterfeiting method, aiming to improve the accuracy of anti-counterfeiting and the convenience for users to identify items to be verified using anti-counterfeiting technology.

[0010] To achieve the above objectives, the present invention proposes an anti-counterfeiting electronic device for determining the source of objects with rough surfaces, comprising: a storage module, a communication module, and a processing module, wherein: The storage module is used to store a set of reflection images of the rough surfaces of registered objects. The set of reflected images It utilizes an image acquisition device to capture the rough surfaces of multiple registered objects under at least three beams of light. The corresponding reflection image under illumination The resulting set, namely the three-dimensional distribution information of the surface microstructure of the rough surface of the registered object; the communication module is used to receive the surface of the object under test under the current light. Reflected image under illumination The processing module is used to calculate the reflected image. grayscale image of at least one specified fixed region or gradient plot The processing module is further configured to calculate the current light... The set of reflected images under illumination The reconstructed image of the grayscale domain of the specified fixed region Or the reconstruction graph of the gradient domain and calculation and or and The similarity value is used to verify whether the object to be verified originates from the registered object.

[0011] In one embodiment, the at least three beams of light The incident vectors are different, and any three beams of light are not coplanar.

[0012] In one embodiment, the reflected image It also includes label information, and the processing module is further configured to acquire the surface information of the object to be inspected under the current light. Reflected image under illumination The label information is used to determine the preliminary origin of the object to be inspected.

[0013] In one embodiment, the processing module is further configured to obtain the corresponding set of reflection images based on the preliminary source. .

[0014] In one embodiment, the processing module uses the set of reflected images. A linear subspace To represent the current light Reflection signals of rough surfaces in the grayscale or gradient domain under illumination ,in , This represents the parameters that determine or define the linear subspace; The processing module obtains the estimate by solving a convex mathematical programming problem. : Or (1) (2) in Gradient operator The results obtained by the above algorithms (1) and (2) are denoted as follows: and ; The registered item is in the current light Reconstruction image of the specified fixed area of ​​the irradiated lower surface in the grayscale domain ; The registered item is in the current light Reconstruction map of the specified fixed region of the irradiated lower surface in the gradient domain .

[0015] In one embodiment, the processing module calculates similarity using an image feature-based similarity algorithm or a statistical property-based similarity algorithm. and or and The similarity is as follows: and And the threshold for the grayscale range is set to Set the threshold of the gradient domain to ,exist or When the processing module determines that the surface microstructure distribution of the object to be inspected and the registered object are highly similar in the specified fixed area, the object to be inspected is the registered object.

[0016] The anti-counterfeiting electronic device of the present invention stores a set of reflective images of the rough surfaces of registered items. Under the current illumination, a reflection image of the surface of the object to be examined is acquired, and a grayscale image or gradient image of at least one specified fixed region is calculated based on the reflection image to reconstruct the set of reflection images under the current illumination. The system uses at least one grayscale image or gradient image of a specified fixed area, and finally calculates the similarity value between the two to confirm whether the object to be verified is a registered object, which improves the accuracy of anti-counterfeiting and also makes it more convenient for users.

[0017] This invention also provides an anti-counterfeiting system for determining the origin of objects with rough surfaces, comprising: an information extraction subsystem, a server subsystem, and a user terminal. The information extraction subsystem is used to acquire the rough surface of the registered object under at least three beams of light. The corresponding reflection image below That is, the three-dimensional distribution information of the surface microstructure of the rough surface of the registered object; The server subsystem receives the reflected images of the rough surfaces of multiple registered objects. To form a set of reflected images The registered objects from different sources correspond to different sets of reflected images. ; The user terminal acquires the surface of the object to be inspected under the current light. Reflected image under illumination And transmit it to the server subsystem; The server subsystem is also used for: Receive the reflected image And calculate the reflected image grayscale image of at least one specified fixed region or gradient plot ; Calculate the current light The set of reflected images under illumination The reconstructed image of the grayscale domain of the specified fixed region Or the reconstruction graph of the gradient domain ; and calculation and or and The similarity value is used to verify whether the object to be tested originates from the registered object; The user terminal is also used to receive and display the verification result of the source of the object to be verified from the server subsystem.

[0018] In one embodiment, the at least three beams of light The incident vectors are different, and any three beams of light are not coplanar.

[0019] In one embodiment, the reflected image It also includes tag information, the server subsystem acquires the surface of the object to be inspected under the current light. Reflected image under illumination The label information is used to determine the preliminary origin of the object to be inspected.

[0020] In one embodiment, the server subsystem acquires the corresponding set of reflection images based on the initial source. .

[0021] In one embodiment, the server subsystem uses the set of reflected images. A linear subspace To represent the current light Reflection signals of rough surfaces in the grayscale or gradient domain under illumination ,in , This represents the parameters that determine or define the linear subspace; The server subsystem estimates by solving a convex mathematical programming problem. : Or (1) (2) in Gradient operator The results obtained by the above algorithms (1) or (2) are respectively denoted as follows: and ; The registered item is in the current light Reconstruction image of the specified fixed area of ​​the irradiated lower surface in the grayscale domain ; The registered item is in the current light Reconstruction map of the specified fixed region of the irradiated lower surface in the gradient domain .

[0022] In one embodiment, the server subsystem calculates similarity using an image feature-based similarity algorithm or a statistical property-based similarity algorithm. and or and The similarity is denoted as . and ; Set the threshold for the grayscale range to Set the threshold of the gradient domain to ,like or If the surface microstructure distribution of the object to be tested and the registered object are highly similar in the specified fixed area, then the object to be tested is the registered object.

[0023] In one embodiment, the information extraction subsystem includes a conveyor belt, at least three light sources, and an image acquisition device corresponding to each light source. The registered object is placed on the conveyor belt, and only one light beam is emitted from each light source at any given time. Simultaneously, the image acquisition device corresponding to each light source takes a picture to obtain a reflection image of the designated fixed area of ​​the rough surface of the registered object under the illumination of the light beam.

[0024] In one embodiment, the image acquisition device includes: a digital camera, a mobile phone embedded camera, an industrial embedded camera, and a scanner.

[0025] The anti-counterfeiting system of this invention obtains a set of reflective images of the rough surface of the registered object through an information extraction subsystem. The data is stored in the server subsystem. The user terminal acquires the reflection image of the object's surface under the current light. The server subsystem calculates the grayscale image or gradient image of at least one specified fixed region of the reflection image of the object's surface to reconstruct the set of reflection images under the current light illumination. The system uses at least one grayscale image or gradient image of a specified fixed area, and finally calculates the similarity value between the two to confirm whether the object to be verified is a registered object, which improves the accuracy of anti-counterfeiting and also makes it more convenient for users.

[0026] The present invention also provides an anti-counterfeiting method for determining the origin of an object with a rough surface, the method comprising: Stores a collection of reflection images of the rough surfaces of registered objects. The set of reflected images It utilizes an image acquisition device to capture the rough surfaces of multiple registered objects under at least three beams of light. The corresponding reflection image under illumination The resulting set is the three-dimensional distribution information of the surface microstructure of the rough surface of the registered object; Obtain the surface of the object under test in the current light Reflected image under illumination ; Calculate the reflection image of the surface of the object to be tested. grayscale image of at least one specified fixed region or gradient plot ; Calculate the current light The set of reflected images under illumination The reconstructed image of the grayscale domain of the specified fixed region Or the reconstruction graph of the gradient domain ;as well as calculate and or and The similarity value; The similarity value is used to verify whether the object to be verified originates from the registered object.

[0027] In one embodiment, the at least three beams of light The incident vectors are different, and any three beams of light are not coplanar.

[0028] In one embodiment, the reflected image When label information is also included, the method further includes: acquiring the surface of the object to be inspected under the current light. Reflected image under illumination Tag information in; The preliminary origin of the object to be inspected is determined by the label information.

[0029] In one embodiment, the corresponding set of reflection images is obtained based on the initial source. .

[0030] In one embodiment, the "calculation of the current light" The set of reflected images under illumination Reconstruction of the grayscale region of a fixed area Or the reconstruction graph of the gradient domain The steps include: Through the set of reflected images A linear subspace To represent the current light Reflection signals of rough surfaces in the grayscale or gradient domain under illumination ,in , This represents the parameters that determine or define the linear subspace; The estimation is obtained by solving a convex mathematical programming problem. : Or (1) (2) in Gradient operator The results obtained by the above algorithms (1) or (2) are respectively denoted as follows: and ; The registered item is in the current light Reconstruction image of the specified fixed area of ​​the irradiated lower surface in the grayscale domain ; The registered item is in the current light Reconstruction map of the specified fixed region of the irradiated lower surface in the gradient domain .

[0031] In one embodiment, the step "calculate" and or and The similarity test, and the verification of the origin of the object to be tested based on the similarity results, includes: Similarity algorithms can be calculated using either image feature-based similarity algorithms or statistical property-based similarity algorithms. and or and The similarity is denoted as . and ; Set the threshold for the grayscale range to Set the threshold of the gradient domain to ,like or If the surface microstructure distribution of the object to be tested and the registered object are highly similar in the specified fixed area, then the object to be tested is the registered object.

[0032] The anti-counterfeiting method of the present invention stores a set of reflective images of the rough surfaces of registered objects. Under the current illumination, a reflection image of the surface of the object to be examined is acquired, and a grayscale image or gradient image of at least one specified fixed region is calculated based on the reflection image to reconstruct the set of reflection images under the current illumination. The system uses at least one grayscale image or gradient image of a specified fixed area, and finally calculates the similarity value between the two to confirm whether the object to be verified is a registered object, which improves the accuracy of anti-counterfeiting and also makes it more convenient for users. Attached Figure Description

[0033] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0034] Figure 1 This is a diagram of the randomly distributed surface microstructure on a rough paper surface in an example of the present invention; Figure 2 This is a schematic diagram of a cross-section of a rough object surface in one example of the present invention; Figure 3 This is an example of the two-dimensional features of the surface microstructure of the same sheet of paper under illumination from different directions in an example of the present invention.

[0035] Figure 4 This is a schematic diagram of an anti-counterfeiting electronic device according to one embodiment of the present invention; Figure 5 This is a schematic diagram of an anti-counterfeiting system according to one embodiment of the present invention; Figure 6 for Figure 5 A schematic diagram of the information extraction subsystem in the diagram; Figure 7 This is a flowchart of an anti-counterfeiting method according to one embodiment of the present invention; Figure 8 This is a flowchart of an anti-counterfeiting method in another embodiment of the present invention; Figure 9 for Figure 7 Flowchart of step S308; Figure 10 for Figure 7 Flowchart of step S310.

[0036] Explanation of icon numbers: 100. Anti-counterfeiting electronic device; 102. Storage module; 104. Communication module; 108. Processing module; 200. Anti-counterfeiting system; 202. Information extraction subsystem; 204. Server subsystem; 206. User terminal; 2022. Conveyor belt; 2024. Light source; 2026. Image acquisition device.

[0037] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0038] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0039] It should be noted that if the embodiments of the present invention involve directional indicators (such as up, down, left, right, front, back, etc.), the directional indicators are only used to explain the relative positional relationship and movement of the components in a specific posture. If the specific posture changes, the directional indicators will also change accordingly.

[0040] Furthermore, if the embodiments of this invention involve descriptions such as "first" or "second," these descriptions are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined with "first" or "second" may explicitly or implicitly include at least one of those features. Additionally, the use of "and / or" or "and / or" throughout the text includes three parallel solutions. For example, "A and / or B" includes solution A, solution B, or a solution where both A and B are satisfied simultaneously. Furthermore, the technical solutions of the various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. When the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed by this invention.

[0041] The premise of this invention is that the registered items need to have the following common characteristics: 1) A certain part of the registered object itself has a rough surface; or / and 2) A certain part of the packaging material of the registered item has a rough surface; The rough surface mentioned here refers to a type of product or packaging surface that appears flat and similar to each other on a macroscopic scale, but has a random, distinct, and unique three-dimensional convex-concave distribution on a microscopic scale. The unique microstructural information of the local rough surface on the product or its packaging material can be used as the product's identification, i.e., the product's anti-counterfeiting feature.

[0042] The unique microstructure on the rough surface of a product is formed naturally during the manufacturing process. For example, different small areas on the surface of a piece of pottery have completely different microstructure distributions. The formation of these different microstructures is caused by a variety of complex factors, such as the inhomogeneity of the clay raw materials, the inhomogeneity of the temperature during firing, manufacturing defects, and the uncontrollability of the microscopic structure.

[0043] Please refer to Figure 1 , Figure 1 The image shown is a diagram of the randomly distributed surface microstructure on a rough paper surface in an example of the present invention.

[0044] from Figure 1As can be seen, two different areas of paper exhibit different surface microstructures. This is caused by a complex interplay of factors, including the varying thicknesses and shapes of the natural plant fibers used in papermaking, their random distribution, the uneven distribution of papermaking fillers, and the random variations in papermaking processes. The microstructure of a particular rough surface of a product is not only unique but also three-dimensional, making the surface microstructure of each item extremely difficult to replicate or clone. Furthermore, even if a counterfeiter could replicate the surface microstructure of a single product at a high cost, it would be impossible to mass-produce / clone a large number of products with different surface microstructures.

[0045] Please see Figure 2 , Figure 2 The figure shown is a schematic diagram of a cross-section of a rough object surface in an example of the present invention.

[0046] from Figure 2 As can be seen, a rough surface has a surface feature of randomly distributed bumps and depressions. In this embodiment, the rough surface is defined as the surface roughness when the surface bumps and depressions of the object are (see...). Figure 2 The wavelength of H in the image is much larger than the wavelength of the incident light on the object's surface. At that time, that is At this point, the surface is said to be rough.

[0047] Please see Figure 3 , Figure 3 The image shows the two-dimensional features of the surface microstructure of the same sheet of paper under illumination from different directions in an example of the present invention.

[0048] from Figure 3 As can be seen, the same rough surface will exhibit significantly different two-dimensional reflective images under unknown lighting conditions from different directions due to complex shadow effects and reflection characteristics. This visual difference in reflective images caused by uncontrollable lighting conditions makes it extremely difficult to directly compare whether two reflective images originate from the same surface under unknown lighting conditions. Therefore, how to enable ordinary consumers to reliably verify this in any everyday lighting scenario without professional equipment constitutes the core technical problem that this invention needs to solve.

[0049] Therefore, this invention proposes an anti-counterfeiting electronic device, an anti-counterfeiting system, and an anti-counterfeiting method for determining the origin of objects with rough surfaces.

[0050] In this invention, an object with a rough surface can be an object whose components themselves have a rough surface, or an object whose surface is smooth but which is formed by artificially coating or attaching a material with rough features to the smooth surface.

[0051] Please refer to Figure 4 , Figure 4 The diagram shown is a schematic diagram of an anti-counterfeiting electronic device according to an embodiment of the present invention.

[0052] In this embodiment, the anti-counterfeiting electronic device 100 includes: a storage module 102, a communication module 104, and a processing module 108.

[0053] In this embodiment, the storage module 102 is used to store a set of reflection images of the rough surface of the registered object. .

[0054] In this embodiment, the set of reflected images It utilizes an image acquisition device to capture the rough surfaces of multiple registered objects under at least three beams of light. The corresponding reflection image under illumination The resulting set is the three-dimensional distribution information of the surface microstructure of the rough surface of the registered object.

[0055] In this embodiment, at least one fixed area is pre-defined on the rough surface of the registered object.

[0056] In this embodiment, the image acquisition device is a device independent of the anti-counterfeiting electronic device 100, including digital cameras, mobile phone embedded cameras, industrial embedded cameras, scanners, and other devices.

[0057] In this embodiment, the at least three beams of light The incident vectors are different, and any three beams of light are not coplanar.

[0058] In this invention, each beam of light corresponds to a light source. The light source can be a point light source or a surface light source with a very small area that approximates a point light source.

[0059] In this invention, the light source can be visible light or invisible light.

[0060] In this invention, each light source corresponds to an image acquisition device, and the light source emits a beam of light onto a fixed area of ​​the registered object. At that time, the corresponding image acquisition device acquires the reflected image of the fixed area. .

[0061] In other embodiments of the present invention, the same light source and the same image acquisition device can also be used to complete image acquisition. That is, different reflected images can be obtained by moving the light source to adjust the incident vector of the light beam, and it is ensured that the incident vectors of at least three emitted light beams are different, and that the light beams emitted in any three times are not coplanar.

[0062] In summary, in this embodiment, Incident light from different directions Each surface of the registered object is illuminated separately, and each surface is photographed / scanned using an image acquisition device. Here... Incident light vectors in different directions Non-parallel and any three non-coplanar elements have the same or different wavelengths, and can be visible or invisible light; hereinafter referred to as this... The set of incident beams is called the incident beam training set, denoted as . The above The surface reflection images of the registered object obtained under beam illumination are digitized to obtain multiple digital information of the object surface, for each type of illumination. The image obtained under illumination is denoted as The set of these digital information is denoted as the set of reflected images. This is called the training image set of the registered object. Since the surface of the registered object contains rough surfaces with many complex microstructures, the same rough surface will exhibit different reflection signals under different incident light illuminations. The distribution characteristics of these reflection signals are determined by the spatial relationship between the microstructure and macroscopic distribution of the object's surface roughness and the incident light. They are not the same as each other, but are a collection of reflected images. It contains three-dimensional distribution information of the microstructure of the rough surface.

[0063] In this embodiment, the communication module 104 is used to receive the surface of the object to be inspected under the current light. Reflected image under illumination .

[0064] In this embodiment, the surface of the object to be inspected also has at least one fixed area, and the user obtains the image of at least one fixed area under the current light through a user terminal. Reflected image under illumination .

[0065] In this embodiment, the current light Arbitrary light refers to incident light with arbitrary direction, intensity, and wavelength (within the sensing range of the image acquisition device). It can be ordinary ambient light or lamplight, or light with a certain directionality, such as direct sunlight or lamplight.

[0066] In this embodiment, the user terminal can be an electronic device with a camera, such as a mobile phone, handheld computer, laptop computer, camera, etc.

[0067] In this embodiment, the processing module 106 is used to calculate the reflection image. grayscale image of at least one specified fixed region or gradient plot .

[0068] In this embodiment, the processing module 106 processes the reflected image. and a collection of reflected images In A registered image The process involves grayscale conversion, spatial registration, low-pass filtering, and resolution normalization. One or more specific sub-regions (ROIs) are designated within these reflective images as the areas to be analyzed and compared. Gradient maps of the ROIs are calculated, and noise is removed. This ensures the fairness and accuracy of subsequent comparisons between the objects to be examined and the registered objects.

[0069] In this embodiment, the processing module 106 uses the set of reflected images. linear subspace To represent the current light Reflection signals of rough surfaces in the grayscale or gradient domain under illumination ,in , This represents the parameters that determine or define the linear subspace.

[0070] In this embodiment, the linear subspace Its dimension is less than N.

[0071] In this embodiment, the processing module 106 obtains the estimate by solving a convex mathematical programming problem. : Or (1) (2) in Gradient operator The results obtained by the above algorithms (1) and (2) are denoted as follows: and ; The registered item is in the current light Reconstruction image of the specified fixed area of ​​the irradiated lower surface in the grayscale domain ; The registered item is in the current light Reconstruction map of the specified fixed region of the irradiated lower surface in the gradient domain .

[0072] In this embodiment, the processing module 106 is further configured to calculate the current light The set of reflected images under illumination The reconstructed image of the grayscale domain of the specified fixed region Or the reconstruction graph of the gradient domain and calculation and or and The similarity value is used to verify whether the object to be verified originates from the registered object.

[0073] In this embodiment, the processing module 106 calculates similarity using an image feature-based similarity algorithm or a statistical characteristic-based similarity algorithm. and or and The similarity values ​​are respectively and And the threshold for the grayscale range is set to Set the threshold of the gradient domain to ,exist or When the object to be inspected is identified as the registered object, the processing module 106 determines that the surface microstructure distribution of the object to be inspected and the registered object are highly similar in the specified fixed area, that is, the object to be inspected is the registered object.

[0074] In this embodiment, the similarity algorithm includes known algorithms in the field of computer vision used to determine the differences or similarities between two images, including but not limited to similarity measurement algorithms based on image features (such as SIFT, SURF, ORB, etc.) or similarity measurement algorithms based on image statistical characteristics (such as MSE (Mean Squared Error), SSIM (Structural Similarity Index Measure), Cosine Similarity, Pearson SimilarityCoefficient, etc.).

[0075] In other embodiments of the present invention, the storage module 102 stores the set of reflected images using the label information provided by the manufacturer as an index. Among them, the set of reflected images Reflection image in It contains label information, which can be a QR code or barcode, corresponding to the relevant manufacturer. The processing module 106 is also used to acquire the surface of the object to be inspected under the current light. Reflected image under illumination The label information is used to determine the preliminary origin of the object to be inspected.

[0076] In one embodiment, the processing module 106 is further configured to obtain the corresponding set of reflection images based on the preliminary source. .

[0077] Thus, the processing module 106 only needs to calculate the current light. The corresponding set of reflected images under illumination The reconstructed image of the grayscale domain of the specified fixed region Or the reconstruction graph of the gradient domain Therefore, this invention can be perfectly integrated with QR codes / barcodes, making the traceability of QR codes / barcodes highly reliable, while also reducing the workload of the anti-counterfeiting electronic device 100, improving computing efficiency, and shortening verification time.

[0078] The anti-counterfeiting electronic device of the present invention stores a set of reflective images of the rough surfaces of registered items. Under the current illumination, a reflection image of the surface of the object to be examined is acquired, and a grayscale image or gradient image of at least one specified fixed region is calculated based on the reflection image to reconstruct the set of reflection images under the current illumination. The system uses at least one grayscale image or gradient image of a designated fixed area. The similarity value between the two images is then calculated to confirm whether the item being verified is a registered item. This improves the accuracy of anti-counterfeiting measures while also making it more convenient for users. Users do not need to learn or possess any knowledge of product anti-counterfeiting; they can scan and authenticate any product using a regular smartphone.

[0079] Please refer to Figure 5 , Figure 5 The diagram shown is a schematic diagram of an anti-counterfeiting system according to one embodiment of the present invention.

[0080] In this embodiment, the anti-counterfeiting system 200 is used to determine the origin of objects with rough surfaces.

[0081] In this embodiment, the anti-counterfeiting system 200 includes: an information extraction subsystem 202, a server subsystem 204, and a user terminal 206.

[0082] In this embodiment, the information extraction subsystem 202 is used to obtain the rough surface of the registered object under at least three beams of light. The corresponding reflection image under illumination That is, the three-dimensional distribution information of the surface microstructure of the rough surface of the registered object.

[0083] In this embodiment, at least one fixed area is pre-defined on the rough surface of the registered object.

[0084] Please refer to the reference. Figure 6 , Figure 6 As shown Figure 5 A schematic diagram of the information extraction subsystem 202 in the diagram.

[0085] In this embodiment, the information extraction subsystem 202 includes a conveyor belt 2022, at least three light sources 2024, and an image acquisition device 2026 corresponding to the light source 2024. The registered object is placed on the conveyor belt 2022. At any given time, only one light source 2024 emits a light beam, and the image acquisition device 2026 corresponding to the light source takes a picture simultaneously to obtain a reflection image of the designated fixed area of ​​the rough surface of the registered object under the illumination of the light beam.

[0086] In this embodiment, the at least three light sources are not located on the same straight line, and the at least three beams of light The incident vectors are different, and any three beams of light are not coplanar.

[0087] In this embodiment, the light source 2024 is a point light source. It can be understood that the light source 2024 can also be a surface light source with a very small area, approximating a point light source.

[0088] In this embodiment, the light source 2024 can be visible light or invisible light.

[0089] In one embodiment, the image acquisition device 206 includes, but is not limited to, a digital camera, a mobile phone embedded camera, an industrial embedded camera, and a scanner.

[0090] In other embodiments of the present invention, the same light source and the same image acquisition device can also be used to complete image acquisition. That is, different reflected images can be obtained by moving the light source to adjust the incident vector of the light beam, and it is ensured that the incident vectors of at least three emitted light beams are different, and that the light beams emitted in any three times are not coplanar.

[0091] In summary, in this embodiment, Incident light from different directions Each surface of the registered object is illuminated separately, and each surface is photographed / scanned using an image acquisition device. Here... Incident light vectors in different directions Non-parallel and any three non-coplanar elements have the same or different wavelengths, and can be visible or invisible light; hereinafter referred to as this... The set of incident beams is called the incident beam training set, denoted as . The above The surface reflection images of the registered object obtained under beam illumination are digitized to obtain multiple digital information of the object surface, for each type of illumination. The image obtained under illumination is denoted as The set of these digital information is denoted as the set of reflected images. This is called the training image set of the registered object. Since the surface of the registered object contains rough surfaces with many complex microstructures, the same rough surface will exhibit different reflection signals under different incident light illuminations. The distribution characteristics of these reflection signals are determined by the spatial relationship between the microstructure and macroscopic distribution of the object's surface roughness and the incident light. They are not the same as each other, but are a collection of reflected images. It contains three-dimensional distribution information of the microstructure of the rough surface.

[0092] Server subsystem 204 is used to receive the reflected images of the rough surfaces of the plurality of registered objects. To form a set of reflected images The registered objects from different sources correspond to different sets of reflected images. .

[0093] In this embodiment, the user terminal 206 is used to acquire the surface of the object to be inspected under the current light. Reflected image under illumination And transmit it to the server subsystem 204.

[0094] In this embodiment, the user terminal 206 can be an electronic device with a camera, such as a mobile phone, handheld computer, laptop computer, camera, etc.

[0095] In this embodiment, the current light is arbitrary light, incident light with arbitrary direction, intensity, and wavelength (within the sensing range of the image acquisition device). It can be ordinary ambient light or lamp light, or light with a certain directionality, such as direct sunlight or lamp light.

[0096] In this embodiment, the server subsystem 204 is further configured to: Receive the reflected image And calculate the reflection image of the surface of the object under test. grayscale image of at least one specified fixed region or gradient plot ; Calculate the current light The set of reflected images under illumination The reconstructed image of the grayscale domain of the specified fixed region Or the reconstruction graph of the gradient domain ; and calculation and or and The similarity value is used to verify whether the object to be verified originates from the registered object.

[0097] In this embodiment, the server subsystem 204 handles the reflected images. and a collection of reflected images In A registered image The process involves grayscale conversion, spatial registration, low-pass filtering, and resolution normalization. One or more specific sub-regions (ROIs) are designated within these reflective images as the areas to be analyzed and compared. Gradient maps of the ROIs are calculated, and noise is removed. This ensures the fairness and accuracy of subsequent comparisons between the objects to be examined and the registered objects.

[0098] In this embodiment, the server subsystem 204 uses the set of reflected images. A linear subspace To represent the current light Reflection signals of rough surfaces in the grayscale or gradient domain under illumination ,in , This represents the parameters that determine or define the linear subspace.

[0099] In this embodiment, the linear subspace Its dimension is less than N.

[0100] The server subsystem 204 estimates by solving a convex mathematical programming problem. : Or (1) (2) in Gradient operator The results obtained by the above algorithms (1) or (2) are respectively denoted as follows: and ; The registered item is in the current light Reconstruction image of the specified fixed area of ​​the irradiated lower surface in the grayscale domain ; The registered item is in the current light Reconstruction map of the specified fixed region of the irradiated lower surface in the gradient domain .

[0101] In this embodiment, the server subsystem 204 calculates similarity using an image feature-based similarity algorithm or a statistical characteristic-based similarity algorithm. and or and The similarity values ​​are denoted as follows: and .

[0102] Set the threshold for the grayscale range to Set the threshold of the gradient domain to ,like or If the server subsystem 204 considers that the surface microstructure distribution of the object to be inspected and the registered object in the specified fixed area is highly similar, the object to be inspected is the registered object.

[0103] In this embodiment, the similarity algorithm includes known algorithms in the field of computer vision used to determine the differences or similarities between two images, including but not limited to similarity measurement algorithms based on image features (such as SIFT, SURF, ORB, etc.) or similarity measurement algorithms based on image statistical characteristics (such as MSE (Mean Squared Error), SSIM (Structural Similarity Index Measure), Cosine Similarity, Pearson SimilarityCoefficient, etc.).

[0104] Optionally, the server subsystem 204 stores the set of reflected images using the tag information provided by the manufacturer as an index. Among them, the set of reflected images Reflection image in It contains label information, which can be a QR code or barcode, corresponding to the manufacturer. The server subsystem 204 is also used to acquire the surface information of the object to be inspected under the current light. Reflected image under illumination The label information is used to determine the preliminary origin of the object to be inspected.

[0105] Therefore, the server subsystem 204 is also used to obtain the corresponding set of reflection images based on the initial source. Thus, the server subsystem 204 only needs to calculate the current light... The corresponding set of reflected images under illumination The reconstructed image of the grayscale domain of the specified fixed region Or the reconstruction graph of the gradient domain This invention achieves a perfect combination of anti-counterfeiting technology and QR code / barcode, making the traceability of QR code / barcode highly reliable. At the same time, it reduces the workload of the server subsystem 204, improves computing efficiency, and shortens the verification time.

[0106] The user terminal 206 is also used to receive and display the verification result of the source of the object to be verified from the server subsystem 204 via the network.

[0107] In this embodiment, the server subsystem 204 includes Figure 5 The electronic anti-counterfeiting device 100 can be located on a cloud server or a local server. When the server subsystem 204 is located in the cloud, the rough surface of the registered object obtained by the information extraction subsystem 202 is illuminated by at least three beams of light. The corresponding reflection image under illumination It needs to be uploaded to server subsystem 204 via the network. If server subsystem 204 is located locally, the rough surface of the registered object obtained by local information extraction subsystem 202 is within at least three beams of light. The corresponding reflection image under illumination It can be directly transmitted to the server subsystem 204 via a wired interface or network.

[0108] The anti-counterfeiting system of this invention obtains a set of reflective images of the rough surface of the registered object through an information extraction subsystem. The data is stored in the server subsystem. The user terminal acquires the reflection image of the object's surface under the current light. The server subsystem calculates the grayscale image or gradient image of at least one specified fixed region of the reflection image of the object's surface to reconstruct the set of reflection images under the current light illumination. The system uses at least one grayscale image or gradient image of a designated fixed area. The similarity value between the two images is then calculated to confirm whether the item being verified is a registered item. This improves the accuracy of anti-counterfeiting measures while also making it more convenient for users. Users do not need to learn or possess any knowledge of product anti-counterfeiting; they can scan and authenticate any product using a regular smartphone.

[0109] Please see Figure 7 , Figure 7 The diagram shows a flowchart of an anti-counterfeiting method according to one embodiment of the present invention.

[0110] In this embodiment, the anti-counterfeiting method is used to determine the source of an object with a rough surface, and is used in an anti-counterfeiting electronic device or anti-counterfeiting system. The method includes: Step S302: Store the set of reflection images of the rough surfaces of the registered object. The set of reflected images It utilizes an image acquisition device to capture the rough surfaces of multiple registered objects under at least three beams of light. The corresponding reflection image under illumination The resulting set is the three-dimensional distribution information of the surface microstructure of the rough surface of the registered object.

[0111] In this embodiment, the reflected image can be digital or analog.

[0112] In this embodiment, the at least three beams of light The incident vectors are different, and any three beams of light are not coplanar.

[0113] In this embodiment, the image acquisition device includes image acquisition devices such as digital cameras, mobile phone embedded cameras, industrial embedded cameras, or scanners.

[0114] In other embodiments of the present invention, the same light source and the same image acquisition device can also be used to complete image acquisition. That is, different reflected images can be obtained by moving the light source to adjust the incident vector of the light beam, and it is ensured that the incident vectors of at least three emitted light beams are different, and that the light beams emitted in any three times are not coplanar.

[0115] In summary, in this embodiment, Incident light from different directions Each surface of the registered object is illuminated separately, and each surface is photographed / scanned using an image acquisition device. Here... Incident light vectors in different directions Non-parallel and any three non-coplanar elements have the same or different wavelengths, and can be visible or invisible light; hereinafter referred to as this... The set of incident beams is called the incident beam training set, denoted as . The above The surface reflection images of the registered object obtained under beam illumination are digitized to obtain multiple digital information of the object surface, for each type of illumination. The image obtained under illumination is denoted as The set of these digital information is denoted as the set of reflected images. This is called the training image set of the registered object. Since the surface of the registered object contains rough surfaces with many complex microstructures, the same rough surface will exhibit different reflection signals under different incident light illuminations. The distribution characteristics of these reflection signals are determined by the spatial relationship between the microstructure and macroscopic distribution of the object's surface roughness and the incident light. They are not the same as each other, but are a collection of reflected images. It contains three-dimensional distribution information of the microstructure of the rough surface.

[0116] Step S304: Receive the surface of the object to be inspected under the current light. Reflected image under illumination .

[0117] In this embodiment, the current light is arbitrary light, that is, incident light with arbitrary direction, intensity, and wavelength (within the sensing range of the image acquisition device). It can be ordinary ambient light or lamp light, or light with a certain directionality, such as direct sunlight or lamp light.

[0118] Step S306: Calculate the reflected image grayscale image of at least one specified fixed region or gradient plot .

[0119] Step S308, calculate the current light The set of reflected images under illumination The reconstructed image of the grayscale domain of the specified fixed region Or the reconstruction graph of the gradient domain .

[0120] Step S310, calculate and or and The similarity value; Step S312: Verify whether the object to be verified originates from the registered object based on the similarity value.

[0121] Please refer to the reference. Figure 8 , Figure 8 The diagram shown is a flowchart of an anti-counterfeiting method in another embodiment.

[0122] In this embodiment, the reflected image When tag information is also included, the method further includes: Step S314, obtain the surface of the object to be inspected under the current light. Reflected image under illumination The tag information in the text.

[0123] In this embodiment, step S302 also requires storing the set of reflected images using the label information provided by the manufacturer as an index. The label information can be a QR code or a barcode, corresponding to the manufacturer.

[0124] In this embodiment, the current light Arbitrary light refers to incident light with arbitrary direction, intensity, and wavelength (within the sensing range of the image acquisition device). It can be ordinary ambient light or lamplight, or light with a certain directionality, such as direct sunlight or lamplight.

[0125] In step S316, the preliminary origin of the object to be inspected is determined by the label information.

[0126] If the label information is not stored, it indicates that the object to be verified is obviously not a registered object.

[0127] If the corresponding label information can be found, the preliminary origin of the object to be inspected can be determined.

[0128] In step S318, the corresponding set of reflection images is obtained based on the preliminary source. .

[0129] Thus, this method only needs to calculate the current light. The corresponding set of reflected images under illumination The reconstructed image of the grayscale domain of the specified fixed region Or the reconstruction graph of the gradient domain Therefore, this invention can be perfectly integrated with QR codes / barcodes, making the traceability of QR codes / barcodes highly reliable, while also improving computational efficiency and shortening verification time.

[0130] It should be understood that the order of the steps is not limited to the numerical order. After steps S314 to S318 in this embodiment are completed, step S306 needs to be executed to complete the complete anti-counterfeiting process.

[0131] Please refer to Figure 9 , Figure 9 As shown Figure 7 Flowchart of step S308.

[0132] In this embodiment, the server subsystem or anti-counterfeiting electronic device detects reflected images. and a collection of reflected images In A registered image The process involves grayscale conversion, spatial registration, low-pass filtering, and resolution normalization. One or more specific sub-regions (ROIs) are designated within these reflective images as the areas to be analyzed and compared. Gradient maps of the ROIs are calculated, and noise is removed. This ensures the fairness and accuracy of subsequent comparisons between the objects to be examined and the registered objects.

[0133] In this embodiment, step S308 includes: Step S3082, through the set of reflected images A linear subspace To represent the current light Reflection signals of rough surfaces in the grayscale or gradient domain under illumination ,in , This represents the parameters that determine or define the linear subspace.

[0134] In this embodiment, the linear subspace Its dimension is less than N.

[0135] Step S3084: Obtain the estimate by solving the convex mathematical programming problem. : Or (1) (2) in Gradient operator The results obtained by the above algorithms (1) or (2) are respectively denoted as follows: and ; The registered item is in the current light Reconstruction image of the specified fixed area of ​​the irradiated lower surface in the grayscale domain ; The registered item is in the current light Reconstruction map of the specified fixed region of the irradiated lower surface in the gradient domain .

[0136] Please refer to Figure 10 , Figure 10 As shown Figure 7 Flowchart of step S310.

[0137] In this embodiment, step S310 includes: Step S3102: Calculate similarity using an image feature-based similarity algorithm or a statistical property-based similarity algorithm. and or and The similarity values ​​are denoted as follows: and ; Step S3104: Set the threshold of the grayscale range to... Set the threshold of the gradient domain to ,like or If the surface microstructure distribution of the object to be tested and the registered object are highly similar in the specified fixed area, then the object to be tested is the registered object.

[0138] In this embodiment, the similarity algorithm includes known algorithms in the field of computer vision used to determine the differences or similarities between two images, including but not limited to similarity measurement algorithms based on image features (such as SIFT, SURF, ORB, etc.) or similarity measurement algorithms based on image statistical characteristics (such as MSE (Mean Squared Error), SSIM (Structural Similarity Index Measure), Cosine Similarity, Pearson SimilarityCoefficient, etc.).

[0139] The anti-counterfeiting method of the present invention stores a set of reflective images of the rough surfaces of registered objects. Under the current illumination, a reflection image of the surface of the object to be examined is acquired, and a grayscale image or gradient image of at least one specified fixed region is calculated based on the reflection image to reconstruct the set of reflection images under the current illumination. The system uses at least one grayscale image or gradient image of a designated fixed area. The similarity value between the two images is then calculated to confirm whether the item being verified is a registered item. This improves the accuracy of anti-counterfeiting measures while also making it more convenient for users. Users do not need to learn or possess any knowledge of product anti-counterfeiting; they can scan and authenticate any product using a regular smartphone.

[0140] Application scenarios of the solution in this invention: Scenario 1: For example, a certain brand of Zisha teapots, brand X, enjoys high brand recognition in the market. However, counterfeit products claiming to be from brand X and remarkably similar in function and appearance to the genuine products have appeared on the market. Even the packaging materials, printed patterns, and anti-counterfeiting labels of the counterfeit products are extremely similar to the genuine products, causing consumers to be repeatedly deceived. As a result, the legitimate brand X suffers from declining sales and a loss of market reputation, while consumers also suffer from the harm caused by counterfeit and substandard products.

[0141] According to the present invention, the brand owner can designate an area on the bottom of the Zisha teapot itself, or on the white cardboard or wooden vessel used as packaging, and register the surface microstructure of this area before the product leaves the factory, and then store these surface microstructures of all products in a server.

[0142] Consumers can use their smartphones to open X's official WeChat mini-program and take a picture of the surface of the purchased product at a designated location, guided by the mini-program. The mini-program will automatically send the photographed microstructure of the product surface to a designated cloud server for comparison. The result of the comparison, whether true or false, will be returned to the user's phone within seconds, providing X's official authoritative verification result.

[0143] In this case, X does not need to add any artificial or additional anti-counterfeiting materials, inks, or markings to the teapot or packaging. The surface of the teapot or packaging already has a natural and unique surface microstructure. This surface microstructure information can serve as a unique identifier for each product. Although counterfeiters can relatively easily imitate the appearance of the product and packaging, they cannot or almost cannot clone the unique and random microstructure on the product or packaging. Even X itself cannot manufacture two products with the same surface microstructure.

[0144] Scenario 2: A skincare brand, Y, is sued in court by a consumer who claims that their skin was damaged after purchasing and using a certain skincare product from Y. The consumer provides not only a purchase invoice and the purchased product bearing the Y trademark as evidence, but also confirms that the QR code on the packaging is correctly scanned. Although brand Y suspects that the consumer purchased an illegal counterfeit product, the counterfeit product is so realistic that the brand cannot provide strong evidence to prove its innocence, resulting in damage to its reputation.

[0145] In this scenario, brand owners can use the technology of this invention to extract all surface microstructure information of products before they leave the factory and register them in the cloud for public query and comparison, so as to protect consumers from counterfeit products; in case of disputes, they can also provide the original surface microstructure information of the products and the comparison results to the adjudicating body as evidence to avoid the predicament of being unable to defend themselves.

[0146] Scenario 3: Many e-commerce companies encounter situations where consumers suspect that the goods they sent out were genuine, but the goods returned by consumers are counterfeit (buying genuine and returning fake). They apply to the e-commerce platform to refuse the return, but the platform refuses the e-commerce company's request on the grounds that the e-commerce company cannot provide evidence that the returned goods are counterfeit. As a result, conscientious e-commerce companies suffer economic losses due to fraud by unscrupulous consumers.

[0147] In this case, by utilizing the technology of this invention, e-commerce companies can extract and record the surface microstructure of the goods to be shipped in the e-commerce platform's database before shipment. When a return occurs, the e-commerce company can use a mobile phone or professional equipment to verify whether the surface microstructure characteristics of the returned goods are consistent with the pre-registered microstructure information of the goods, thus protecting itself from fraud involving the purchase of genuine goods and the return of counterfeit goods. In the event of a dispute, the staff of the e-commerce platform can use a mobile phone or professional equipment to compare the surface microstructure of the returned goods with the surface microstructure of the goods in the database, thereby obtaining a sufficiently evidenced and accurate and fair judgment. This can protect both conscientious e-commerce companies from malicious fraud by unscrupulous consumers who deliberately purchase genuine goods and return counterfeit goods, and law-abiding consumers from fraud by unscrupulous merchants who pass off counterfeit goods as genuine.

[0148] Scenario 4: Each item of a certain brand's products is printed with a unique QR code label as a traceability code. However, criminals copy the QR codes on genuine products in large quantities to create fake traceability labels and affix / print them on counterfeit products, thereby illegally profiting and causing serious damage to the brand's reputation.

[0149] In this case, if the brand owner uses QR code labels affixed to products or packaging, they can choose a paper substrate with a certain degree of surface roughness as the QR code label paper. When making the label, the surface microstructure of the label substrate can be extracted / registered, and the registration information of these label surface microstructures can be stored in a cloud database. If the brand owner chooses to directly print the traceability QR code on the white cardboard packaging box, the surface of the white cardboard as the substrate has a certain degree of roughness. The surface microstructure of the QR code substrate can also be extracted / registered after printing, and the registration information of these label substrate surface microstructures can be stored in a cloud database. When consumers scan product QR codes using WeChat mini-programs or apps, aside from the unchanged QR code recognition and parsing process, the system adds an automatic verification process to check the accuracy of the microstructure on the QR code substrate surface. If the product is counterfeit, although its traceability QR code may match a QR code registered by the brand, the microstructure on the substrate surface of the counterfeit QR code label will not match the microstructure registered by the brand, resulting in a "counterfeit product" scan result. By combining the product traceability QR code with the microstructure on the label substrate surface, brand owners only need to add a process to extract / register the label substrate microstructure and upload it to the cloud server during the product production process. Consumers, on the other hand, do not need to change their original scanning habits or acquire any other anti-counterfeiting knowledge to perfectly meet their product anti-counterfeiting needs.

[0150] Therefore, in general, the solution using the present invention requires: 1. Identify a rough surface area on a product, packaging material, or label. This designated fixed area can be marked with a printed mark to facilitate obtaining an image of the designated fixed area of ​​the registered item and the positioning of the user's terminal device when the user scans the item to be verified. 2. Use an image acquisition device to acquire N images of a specified location on the surface of a product or packaging. These N images all contain information about the surface microstructure of the specified fixed area; when each image is taken, the incident light directions illuminating the specified fixed area are not parallel to each other, that is: at least three beams of light have different incident vectors, and the three beams of light are not coplanar; 3. Store these N images on a local or remote server; 4. Users (consumers or any related parties in the commodity circulation process) use the built-in camera of their mobile phones or other image acquisition devices to take pictures or scan a specified location on the commodity to be verified, and obtain one or more images of a specified fixed area of ​​the commodity to be verified. The images contain microstructure information of the surface of the specified fixed area; and upload the images to be verified to a server storing N images using a network / data cable. 5. On the server side, after receiving the image to be verified uploaded by the user, the server executes the anti-counterfeiting method of the present invention and returns the verification result of the source of the object to be verified to the user.

[0151] The above description is merely an exemplary embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural transformations made using the contents of the present invention specification and drawings under the technical concept of the present invention, or direct / indirect applications in other related technical fields, are included within the patent protection scope of the present invention.

Claims

1. An anti-counterfeiting electronic device for determining the origin of an object with a rough surface, characterized in that, include: The storage module is used to store a set of reflection images of the rough surfaces of registered objects. The set of reflected images It utilizes an image acquisition device to capture the rough surfaces of multiple registered objects under at least three beams of light. The corresponding reflection image under illumination The resulting set is the three-dimensional distribution information of the surface microstructure of the rough surface of the registered object; The communication module is used to receive the surface of the object to be inspected under the current light. Reflected image under illumination ; Processing module, used to calculate the reflected image grayscale image of at least one specified fixed region or gradient plot The processing module is further configured to calculate the current light. The set of reflected images under illumination The reconstructed image of the grayscale domain of the specified fixed region Or the reconstruction graph of the gradient domain and calculation and or and The similarity value is used to verify whether the object to be verified originates from the registered object.

2. The anti-counterfeiting electronic device as described in claim 1, characterized in that, The at least three beams of light The incident vectors are different, and any three beams of light are not coplanar.

3. The anti-counterfeiting electronic device as described in claim 1, characterized in that, The reflected image It also includes label information, and the processing module is further configured to acquire the surface information of the object to be inspected under the current light. Reflected image under illumination The label information is used to determine the preliminary origin of the object to be inspected.

4. The anti-counterfeiting electronic device as described in claim 3, characterized in that, The processing module is also used to obtain the corresponding set of reflection images based on the initial source. .

5. The anti-counterfeiting electronic device as described in any one of claims 1 to 4, characterized in that, The processing module uses the set of reflected images. A linear subspace To represent the current light Reflection signals of rough surfaces in the grayscale or gradient domain under illumination ,in , This represents the parameters that determine or define the linear subspace; The processing module obtains the estimate by solving a convex mathematical programming problem. : Or (1) （2） in Describing the gradient operator The results obtained by the above algorithms (1) and (2) are denoted as follows: and ; The registered item is in the current light Reconstruction image of the specified fixed area of ​​the irradiated lower surface in the grayscale domain ; The registered item is in the current light Reconstruction map of the specified fixed region of the irradiated lower surface in the gradient domain .

6. The anti-counterfeiting electronic device as described in claim 1, characterized in that, The processing module calculates similarity using image feature-based similarity algorithms or statistical property-based similarity algorithms. and or and The similarity is as follows: and And the threshold for the grayscale range is set to Set the threshold of the gradient domain to ,exist or When the processing module determines that the surface microstructure distribution of the object to be inspected and the registered object are highly similar in the specified fixed area, the object to be inspected is the registered object.

7. An anti-counterfeiting system for determining the origin of an object with a rough surface, characterized in that, include: An information extraction subsystem is used to obtain the roughness of the registered object under at least three beams of light. The corresponding reflection image under illumination That is, the three-dimensional distribution information of the surface microstructure of the rough surface of the registered object; The server subsystem receives the reflected images of the rough surfaces of multiple registered objects. To form a set of reflected images The registered objects from different sources correspond to different sets of reflected images. ; The user terminal acquires the surface of the object to be inspected under the current light. Reflected image under illumination And transmit it to the server subsystem; The server subsystem is also used for: Receive the reflected image And calculate the reflected image grayscale image of at least one specified fixed region or gradient plot ; Calculate the current light The set of reflected images under illumination The reconstructed image of the grayscale domain of the specified fixed region Or the reconstruction graph of the gradient domain ; and calculation and or and The similarity value is used to verify whether the object to be tested originates from the registered object; The user terminal is also used to receive and display the verification result of the source of the object to be verified from the server subsystem.

8. The anti-counterfeiting system as described in claim 7, characterized in that, The at least three beams of light The incident vectors are different, and any three beams of light are not coplanar.

9. The anti-counterfeiting system as described in any one of claims 7 to 8, characterized in that, The reflected image It also includes tag information, the server subsystem acquires the surface of the object to be inspected under the current light. Reflected image under illumination The label information is used to determine the preliminary origin of the object to be inspected.

10. The anti-counterfeiting system as described in claim 9, characterized in that, The server subsystem obtains the corresponding set of reflection images based on the initial source. .

11. The anti-counterfeiting system as described in claim 7, characterized in that, The server subsystem uses the set of reflected images A linear subspace To represent the current light Reflection signals of rough surfaces in the grayscale or gradient domain under illumination ,in , This represents the parameters that determine or define the linear subspace; The server subsystem estimates by solving a convex mathematical programming problem. : Or (1) （2） in Describing the gradient operator The results obtained by the above algorithms (1) or (2) are respectively denoted as follows: and ; The registered item is in the current light Reconstruction image of the specified fixed area of ​​the irradiated lower surface in the grayscale domain ; The registered item is in the current light Reconstruction map of the specified fixed region of the irradiated lower surface in the gradient domain .

12. The anti-counterfeiting system as described in claim 7, characterized in that, The server subsystem calculates similarity using image feature-based similarity algorithms or statistical property-based similarity algorithms. and or and The similarity is denoted as . and ; Set the threshold for the grayscale range to Set the threshold of the gradient domain to ,like or If the surface microstructure distribution of the object to be tested and the registered object are highly similar in the specified fixed area, then the object to be tested is the registered object.

13. The anti-counterfeiting system as described in claim 7, characterized in that, The information extraction subsystem includes a conveyor belt, at least three light sources, and image acquisition devices corresponding to the light sources. The registered object is placed on the conveyor belt, and only one light beam is emitted from the light source at any given time. The image acquisition device corresponding to the light source takes a picture simultaneously to obtain a reflection image of the designated fixed area of ​​the rough surface of the registered object under the illumination of the light beam.

14. The anti-counterfeiting system as described in claim 13, characterized in that, The image acquisition device includes: a digital camera, a mobile phone embedded camera, an industrial embedded camera, and a scanner.

15. An anti-counterfeiting method for determining the origin of an object with a rough surface, characterized in that, The method includes: Stores a collection of reflection images of the rough surfaces of registered objects. The set of reflected images It utilizes an image acquisition device to capture the rough surfaces of multiple registered objects under at least three beams of light. The corresponding reflection image under illumination The resulting set is the three-dimensional distribution information of the surface microstructure of the rough surface of the registered object; Obtain the surface of the object under test in the current light Reflected image under illumination ; Calculate the reflected image grayscale image of at least one specified fixed region or gradient plot ; Calculate the current light The set of reflected images under illumination The reconstructed image of the grayscale domain of the specified fixed region Or the reconstruction graph of the gradient domain ; and calculation and or and The similarity value; The similarity value is used to verify whether the object to be verified originates from the registered object.

16. The anti-counterfeiting method as described in claim 15, characterized in that, The at least three beams of light The incident vectors are different, and any three beams of light are not coplanar.

17. The anti-counterfeiting method according to any one of claims 15-16, characterized in that, The reflected image When tag information is also included, the method further includes: The surface of the object to be inspected is obtained under the current light. Reflected image under illumination Tag information in; The preliminary origin of the object to be inspected is determined by the label information.

18. The anti-counterfeiting method as described in claim 17, characterized in that, Also includes: The corresponding set of reflection images is obtained based on the initial source. .

19. The anti-counterfeiting method as described in claim 15, characterized in that, The "calculation of the current light" The set of reflected images under illumination Reconstruction of grayscale region of fixed area Or the reconstruction graph of the gradient domain The steps include: Through the set of reflected images A linear subspace To represent the current light Reflection signals of rough surfaces in the grayscale or gradient domain under illumination ,in , This represents the parameters that determine or define the linear subspace; The estimation is obtained by solving a convex mathematical programming problem. : Or (1) （2） in Describing the gradient operator The results obtained by the above algorithms (1) or (2) are respectively denoted as follows: and ; The registered item is in the current light Reconstruction image of the specified fixed area of ​​the irradiated lower surface in the grayscale domain ; The registered item is in the current light Reconstruction map of the specified fixed region of the irradiated lower surface in the gradient domain .

20. The anti-counterfeiting method as described in claim 15, characterized in that, The steps and or and The similarity test, and the verification of the origin of the object to be tested based on the similarity results, includes: Similarity algorithms can be calculated using either image feature-based similarity algorithms or statistical property-based similarity algorithms. and or and The similarity is denoted as . and ; Set the threshold for the grayscale range to Set the threshold of the gradient domain to ,like or If the surface microstructure distribution of the object to be tested and the registered object are highly similar in the specified fixed area, then the object to be tested is the registered object.

Images