An AI large model-based stereoscopic holographic anti-counterfeiting label manufacturing method

Abstract

The application discloses a kind of based on AI big model's stereoscopic holographic anti-fake label production method, belong to anti-fake technical field.A kind of based on AI big model's stereoscopic holographic anti-fake label production method, including using AI to generate stereoscopic model;AI-generated stereoscopic model is rendered into the multiple plane views of different angles, forms holographic image sequence, carries out noise reduction, calibration, optimization processing to the plane view in holographic image sequence;The plane view in holographic image sequence is bound with photoetching angle, uses photoetching machine to etch out the texture corresponding to relevant plane view on photoresist according to the set photoetching angle;Texture on photoresist is transferred to PET film by UV glue photocuring, forms the holographic master copy that can be copied;Metal is deposited or electroplated on the surface of master copy, and the working copy with microstructure on the surface is formed;The microstructure on the surface of working copy is transferred to the surface of substrate, and the preparation of holographic anti-fake label is completed;Meet the demand of brand party to high security, high identification anti-fake.

Description

Technical Field

[0001] This invention relates to the field of anti-counterfeiting technology, and more specifically, to a method for producing three-dimensional holographic anti-counterfeiting labels based on AI large-scale models. Background Technology

[0002] Currently, holographic anti-counterfeiting technology has been widely applied in the anti-counterfeiting of high-value-added products such as alcoholic beverages, high-end cosmetics, electronic products, and metal packaging.

[0003] Traditional 3D holographic anti-counterfeiting images are created by photographing a fixed 3D model or constructing a large model using 3D software such as 3DS MAX. However, these methods suffer from limitations such as limited 3D effects, regular imaging structures, and ease of counterfeiting. If counterfeiters possess the same 3D model, they can easily forge it, failing to meet brands' demands for high security, high recognizability, and high innovation in anti-counterfeiting. With the rapid development of AI and micro-optics technologies, combining artificial intelligence with holographic imaging has become a crucial direction for industry upgrades. However, a mature, mass-producible technical solution that deeply integrates AI-generated random 3D models with holographic plate-making has yet to emerge.

[0004] In view of this, we propose a method for producing 3D holographic anti-counterfeiting labels based on AI large-scale models. Summary of the Invention

[0005] 1. Technical problems to be solved The purpose of this invention is to provide a method for producing three-dimensional holographic anti-counterfeiting labels based on AI large models, so as to solve the problems mentioned in the background art.

[0006] 2. Technical Solution This invention is achieved through the following technical solution: A method for producing 3D holographic anti-counterfeiting labels based on AI large-scale models includes the following steps: S1. Use AI to generate 3D models; S2. Render the AI-generated 3D model into multiple planar views at different angles to form a holographic image sequence. Perform noise reduction, calibration, and optimization processing on the planar views within the holographic image sequence. S3. Bind the planar views in the holographic image sequence to the photolithography angle, and use a photolithography machine to etch the texture corresponding to the relevant planar views on the photoresist according to the set photolithography angle; S4. The texture on the photoresist is transferred to the PET film by UV curing to form a reproducible holographic master. S5. Deposit or electroplate metal onto the surface of the master plate to form a working plate with a microstructure on the surface. S6. Transfer the microstructure on the surface of the working plate to the surface of the substrate to complete the preparation of the holographic anti-counterfeiting label.

[0007] As an optional solution to the technical solution of this application, step S1 includes: using prompt words to input the description of the solid model, the size range, the face count accuracy, the material type and the structural constraint parameters into the AI.

[0008] As an optional solution to the technical solution of this application, step S1 includes: using a multimodal approach, inputting a reference image and / or a reference model, as well as random deformation coefficients, local modification ratios, and detail addition / reduction thresholds into the AI.

[0009] As an optional solution to the technical solution of this application, step S2 includes: different planar views within the holographic image sequence are obtained by moving / scaling / rotating / changing color / deforming, and the viewing angle changes of the planar views within the holographic image sequence are equal and continuous, with the viewing angle change range being 1-2 degrees.

[0010] As an optional solution to the technical solution of this application, step S6 includes: using a molding process to transfer the microstructure on the surface of the working plate to the surface of the substrate; the substrate is a PET or BOPP transparent base film, the molding temperature is 150-170℃, and the pressure is 1.5-2.0MPa.

[0011] As an optional solution to the technical solution of this application, step S6 includes: printing graphics on the surface of the substrate using a positioning printing method.

[0012] As an optional solution to the technical solution of this application, in step S2, the number of planar views in the holographic image sequence is 16-49.

[0013] 3. Beneficial effects Compared with the prior art, the beneficial effects of the present invention are: 1) This invention uses an AI large model to generate a three-dimensional model based on random sampling of latent space. Even if the input prompts and fixed parameters are exactly the same, the topological structure, texture details, and three-dimensional concavity and convexity parameters of the generated model will have random differences each time it is started. Moreover, these differences are irreversible and cannot be manually reproduced. At the same time, through the built-in similarity verification module, the similarity threshold of the model generated repeatedly with the same parameters is preset to ≤85%, which makes it impossible for counterfeiters to obtain the original 3D model data through reverse engineering. Each anti-counterfeiting label corresponds to a unique AI-generated model, thereby completely breaking the technical bottleneck of "the same master can be copied" in traditional holographic anti-counterfeiting and realizing the absolute uniqueness of "one item, one image", which greatly increases the threshold for counterfeiting.

[0014] 2) This invention uses a positioning printing method to precisely position and merge the printed pattern with the holographic pattern, and can be integrated with RFID and digital traceability technologies to form a triple independent security barrier of physical optical anti-counterfeiting, wireless radio frequency identification, and blockchain / cloud traceability. Among them, the physical layer AI random holographic stereo pattern is extremely difficult to counterfeit, the electronic layer RFID supports long-distance batch verification, and the digital layer traceability code enables full-link information to be traceable. Moreover, the three barriers are independent of each other, and the breach of any one layer will not affect the security of the other two layers, thereby significantly improving the overall reliability and anti-attack capability of the anti-counterfeiting system.

[0015] 3) This invention uses AI modeling to replace traditional physical modeling, saving manpower and material costs in mold making, sculpting, photography and other processes. Moreover, the marginal cost of mass replication by nano-molding after one-time holographic plate making is extremely low. At the same time, this invention is compatible with existing production lines and avoids huge equipment investment. This makes the overall anti-counterfeiting cost of this invention significantly lower than that of pure chip anti-counterfeiting solutions with the same security level. Thus, it achieves a balance between high security and high cost performance, and is suitable for widespread application in mid-to-high-end consumer goods fields such as wine, cosmetics, 3C products, and luxury goods. Attached Figure Description

[0016] Figure 1 This is a logic diagram of a method for producing 3D holographic anti-counterfeiting labels based on a large AI model; Figure 2 It consists of multiple planar views within a holographic image sequence. Detailed Implementation

[0017] The technical solution of the present invention will now be clearly and completely described in conjunction with the accompanying drawings.

[0018] Please see Figure 1 This invention provides a method for producing 3D holographic anti-counterfeiting labels based on AI large-scale models, including the following steps: S1. Use AI to generate 3D models; S2. Render the AI-generated 3D model into multiple planar views at different angles to form a holographic image sequence. Perform noise reduction, calibration, and optimization processing on the planar views within the holographic image sequence. S3. Bind the planar views in the holographic image sequence to the photolithography angle, and use a photolithography machine to etch the texture corresponding to the relevant planar views on the photoresist according to the set photolithography angle; S4. The texture on the photoresist is transferred to the PET film by UV curing to form a reproducible holographic master. S5. Deposit or electroplate metal onto the surface of the master plate to form a working plate with a microstructure on the surface. S6. Transfer the microstructure on the surface of the working plate to the surface of the substrate to complete the preparation of the holographic anti-counterfeiting label.

[0019] As a preferred embodiment of this application, step S1 includes: inputting a description of the solid model, size range, facet count accuracy, material type and structural constraint parameters into the AI ​​using prompt words.

[0020] As a preferred embodiment of this application, step S1 includes: using a multimodal approach, inputting a reference image and / or a reference model, as well as random deformation coefficients, local modification percentages, and detail addition / reduction thresholds into the AI.

[0021] As a preferred embodiment of this application, step S2 includes: acquiring different planar views within the holographic image sequence through movement / scaling / rotation / color change / deformation, wherein the viewpoint changes of the planar views within the holographic image sequence are equal and continuous, with a viewpoint change range of 1-2 degrees. As a preferred embodiment of this application, step S6 includes: using a molding process to transfer the microstructure on the surface of the working plate to the surface of the substrate; the substrate is a PET or BOPP transparent base film, the molding temperature is 150-170℃, and the pressure is 1.5-2.0MPa.

[0022] As a preferred embodiment of this application, step S6 includes: printing graphics on the surface of a substrate using a positioning printing method.

[0023] In a preferred embodiment of this application, in step S2, the number of planar views in the holographic image sequence is 16-49.

[0024] The above solution will be illustrated with specific examples below: Enter the following prompts into the AI: "Generate a high-definition 3D cartoon dog logo with a simple and clear outline, uniform texture, high color saturation, no complex and redundant structure, and allow the cartoon dog to rotate left and right, with a horizontal rotation of 48°, which is compatible with holographic anti-counterfeiting technology and has a strong sense of three-dimensionality." At the same time, the basic parameters of the model (image quality, size, stereoscopic strength) are manually locked.

[0025] After the model is started, the background automatically generates random 3D models without human intervention or fixed rules. Even if the same prompts and fixed parameters are repeatedly entered, the 3D models generated each time will be different, which directly leads to significant differences in the topology, texture details, and convexity parameters of the subsequent stereo models.

[0026] The AI-generated stereo model was rendered into 16–49 consecutive color images from different viewing angles, such as Figure 2As shown, each image is assigned precise angle parameters, and then subjected to noise reduction, sharpening, and angle calibration to form an image sequence adapted for holographic plate making. This image sequence is a series of continuously grading images, with 16-49 frames. The main subject of each frame remains unchanged, and the image changes include, but are not limited to, movement, scaling, rotation, color change, and deformation. The changes are subtle, and the transitions between frames are natural. The angle assigned to each frame is an arithmetic sequence, with an angle difference of 1-2 degrees.

[0027] Multi-angle image sequences are imported into a photolithography machine, and after computer color separation, photolithography engraving, UV panelization, and electroplating processes, a high-precision holographic plate is produced. The surface of the holographic plate forms a special texture, and the continuously changing image can be seen by the naked eye from different angles, forming a stereoscopic vision and ensuring that the image details are fully presented.

[0028] Using PET or BOPP transparent base film as the substrate, the AI ​​three-dimensional micro-nano structure on the holographic master is transferred to the surface of the PET base film through a nano-precision molding process. The molding temperature is 150-170℃ and the pressure is 1.5-2.0MPa, forming an anti-counterfeiting film with dynamic three-dimensional visual effects, ensuring the complete replication of the micro-nano structure.

[0029] The graphics and text are printed using a positioning printing method, which can be combined with laser holography, RFID, and digital traceability technologies to form a triple anti-counterfeiting label with AI stereoscopic vision, physical anti-counterfeiting, and digital traceability functions.

Claims

1. A method for producing 3D holographic anti-counterfeiting labels based on AI large-scale models, characterized in that: Includes the following steps: S1. Use AI to generate 3D models; S2. Render the AI-generated 3D model into multiple planar views at different angles to form a holographic image sequence. Perform noise reduction, calibration, and optimization processing on the planar views within the holographic image sequence. S3. Bind the planar views in the holographic image sequence to the photolithography angle, and use a photolithography machine to etch the texture corresponding to the relevant planar views on the photoresist according to the set photolithography angle; S4. The texture on the photoresist is transferred to the PET film by UV curing to form a reproducible holographic master. S5. Deposit or electroplate metal onto the surface of the master plate to form a working plate with a microstructure on the surface. S6. Transfer the microstructure on the surface of the working plate to the surface of the substrate to complete the preparation of the holographic anti-counterfeiting label.

2. The method for producing a three-dimensional holographic anti-counterfeiting label based on an AI large model according to claim 1, characterized in that: Step S1 includes: using prompts to input the description of the 3D model, size range, face count accuracy, material type, and structural constraint parameters into the AI.

3. The method for producing a three-dimensional holographic anti-counterfeiting label based on an AI large model according to claim 1, characterized in that: Step S1 includes: using a multimodal approach, inputting a reference image and / or reference model, as well as random deformation coefficients, local modification percentages, and detail addition / reduction thresholds into the AI.

4. The method for producing a three-dimensional holographic anti-counterfeiting label based on an AI large model according to claim 1, characterized in that: Step S2 includes: different planar views within the holographic image sequence are obtained by moving / scaling / rotating / changing color / deforming. The viewpoint changes of the planar views within the holographic image sequence are equal and continuous, with a viewpoint change range of 1-2 degrees.

5. The method for producing a three-dimensional holographic anti-counterfeiting label based on an AI large model according to claim 1, characterized in that: Step S6 includes: using a molding process to transfer the microstructure on the surface of the working plate to the surface of the substrate; the substrate is a PET or BOPP transparent base film, the molding temperature is 150-170℃, and the pressure is 1.5-2.0MPa.

6. The method for producing a three-dimensional holographic anti-counterfeiting label based on an AI large model according to claim 1, characterized in that: Step S6 includes: printing graphics on the surface of the substrate using a positioning printing method.

7. The method for producing a three-dimensional holographic anti-counterfeiting label based on an AI large model according to claim 1, characterized in that: In step S2, the number of planar views in the holographic image sequence is 16-49.

Images