Three-dimensional image identification based on coagulated color particles and manufacturing and identification method thereof

Abstract

This invention discloses a three-dimensional image identifier based on solidified colored microparticles and its manufacturing and recognition method. The image identifier includes a transparent, cured resin substrate with randomly dispersed and solidified colored micro-ink droplets within its three-dimensional space. These particles undergo Brownian motion before solidification in the liquid resin, and their spatial distribution and color combination after instantaneous solidification constitute a difficult-to-replicate three-dimensional physical fingerprint. The substrate surface is covered with a transparent protective film, and the substrate has an integrally molded frame around its perimeter, which can be embedded in a pre-reserved structure of a metal object awaiting verification, providing both wear resistance and tamper-proof functionality. During recognition, the user uses a dedicated smartphone app to vertically photograph the identifier. The app automatically extracts the particle distribution and color features and generates a feature code, which is compared with pre-stored registration feature data on the blockchain, providing immediate feedback on authenticity. For further confirmation, a tilted angle photograph can be taken to obtain particle depth information for secondary verification. This invention combines the physical randomness of Brownian motion of microparticles, convenient vertical imaging recognition with blockchain evidence storage technology, solving the problems of easy wear and replication of traditional planar patterns, while achieving a balance between high security and user-friendliness.

Description

Technical Field

[0001] This invention relates to the field of digital image recognition technology, and more specifically, to a digital image identifier and its manufacturing and recognition method that utilizes the physical non-cloning property of Brownian motion of microscopic particles in a three-dimensional medium, combined with protective design and advanced evidence preservation technology. Background Technology

[0002] In the fields of commodity management, precision equipment manufacturing, and machine vision, image tags with unique characteristics, high durability, and machine-recognizable features have a wide range of applications. They can be used for product anti-counterfeiting and traceability control of goods and devices, as well as for machine vision positioning and precise target alignment of important equipment. Existing tags mostly use fixed printing or coating structures such as holograms, anti-counterfeiting inks, and QR codes. Among them, the random distribution of microscopic particles constructs unique features by arranging monochrome or fluorescent microparticles on the two-dimensional surface of the tag. The feature image is formed by the random landing points of the particles. The position, shape, and color information of the particles are extracted by microscopic imaging to generate a unique digital identity code.

[0003] However, these random particle markers on two-dimensional planes have significant shortcomings: First, the feature particles are completely exposed, making them susceptible to friction, scratches, chemical corrosion, or ultraviolet radiation, leading to the shedding, displacement, or wear of tiny particles and permanent loss of feature information, thus reducing the reliability of subsequent machine recognition. Second, the two-dimensional particle images on the surface are easily captured completely by high-magnification microscopes and scanning equipment, and can be mass-produced using micro-nano processing techniques, making it difficult to guarantee the uniqueness and non-replicability of the markers. Furthermore, traditional markers have limited functionality, mostly limited to anti-counterfeiting scenarios, and cannot simultaneously meet the requirements of target alignment and spatial positioning for equipment. Moreover, image recognition often relies on expensive and complex specialized equipment, resulting in a high barrier to entry and limited applicability.

[0004] Therefore, the industry urgently needs a new 3D image identification technology solution that combines high durability, uniqueness, and easy identification, while also being adaptable to various application scenarios. Summary of the Invention

[0005] The purpose of this invention is to provide a solidified colored microparticle three-dimensional image identifier and its manufacturing and identification method, so as to solve the problems of pattern easy damage and wear, easy high-precision copying and forgery, and the need for special equipment for verification in existing planar image identifier technology.

[0006] To achieve the above objectives, the technical solution of the present invention has been innovatively designed from three core dimensions.

[0007] First, in terms of physical image features, this invention constructs a three-dimensional particle distribution space. The core of the three-dimensional image identifier is a transparent solid substrate made of low-viscosity liquid transparent resin that has been instantly solidified. Within the three-dimensional space of the substrate, a large number of tiny colored ink droplets are randomly suspended and permanently solidified. The distribution of these particles is generated through a unique "dynamic freezing" process: using a piezoelectric on-demand inkjet printhead, colored microparticles are implanted in the form of extremely small droplets into different three-dimensional coordinate positions of the low-viscosity liquid transparent resin layer; then, the entire assembly is placed in a Brownian motion stationary position, where, within a controlled time window, the droplets undergo completely random Brownian motion under the impact of the thermal motion of liquid molecules, their trajectories being unpredictable and unreproducible; at a certain instant during the motion, a high-power ultraviolet curing light source is activated, instantly solidifying the entire low-viscosity liquid transparent resin matrix, permanently sealing the precise three-dimensional spatial coordinates of each ink droplet at that instant, its shape resulting from the motion, and its spatial relationship with other ink droplets within the substrate. The resulting particle spatial distribution is a three-dimensional fingerprint with physically unclonable properties. Any attempt to replicate this would mean precisely reproducing the Brownian motion history of thousands of particles within a complex fluid at a microscopic scale, which is physically nearly impossible.

[0008] Secondly, in terms of structural protection and integration, this invention designs an integrated protection and connection structure. At least one outer surface of the solid substrate is coated with a transparent protective film layer with high hardness and high light transmittance, effectively resisting daily scratches and chemical corrosion, ensuring the long-lasting preservation of the internal colored micro-ink droplet particle pattern. When the substrate is formed in the molding cavity, its perimeter is designed as an integrally formed or firmly connected integral frame. This frame has a mechanical locking buckle structure, which can accurately and firmly embed into the pre-reserved mounting slot of the target object body, becoming an integral part of the object body. During assembly, a structural adhesive filler layer can be used to strengthen the connection. This design not only enhances physical strength but also makes the transfer or replacement of the markings extremely difficult, constructing a second line of defense against physical disassembly.

[0009] Finally, in terms of identification and verification, this invention designs a hierarchical identification strategy that primarily uses vertical imaging and secondarily uses tilted imaging, and deeply integrates it with blockchain evidence storage technology. Since the colored micro-ink droplets in the three-dimensional space of the cured substrate have not only horizontal distribution but also depth distribution, even when observed only from the vertical direction, the relative position, local density, and color composition of the particles on the two-dimensional projection surface already possess extremely high randomness and uniqueness, sufficient to support rapid verification in most scenarios. Therefore, the conventional identification process is designed to be extremely simple: the user launches the verification APP and enables camera permissions, using a smartphone camera to capture an image along the vertical imaging optical axis; the system sequentially performs image preprocessing and particle detection, extracts particle coordinates and color features, selects feature particles and generates feature codes, uploads the serial number and feature code to the verification system, compares with the blockchain, and completes conventional verification judgment, finally outputting a conclusion or performing a secondary tilted verification based on the result. The entire process requires only one shot, is easy to operate, and perfectly suited to consumers' daily usage habits.

[0010] For a few scenarios where the matching score is in the critical doubt range, or where the system determines the item to be of high value and requires high security, a secondary confirmation process can be automatically triggered. In this case, the app guides the user to adjust their smartphone camera to a tilted angle. θ The algorithm captures the optical axis and then takes a supplementary image from a tilted perspective. Utilizing the particle parallax between the vertical and tilted images, the algorithm quickly estimates the relative depth information of the colored micro-droplet particles, using these depth features as supplementary criteria for secondary judgment. This hierarchical mechanism ensures efficiency and convenience in most scenarios while providing additional safety assurance when necessary.

[0011] The verification system is connected to a blockchain network. When each 3D image identifier is registered at the factory, its feature code, extracted through vertical imaging, is encrypted, hashed, and stored on the blockchain's distributed ledger, possessing the characteristics of immutability and permanent traceability. Verification involves calling smart contracts for on-chain comparison, eliminating the risk of centralized database attacks or internal tampering.

[0012] In summary, this invention integrates the physical non-cloning property of three-dimensional particle space, embedded protective structural design, hierarchical and convenient identification based on vertical imaging, and secure blockchain evidence storage to provide an image identification solution that is highly secure, easy to use, and capable of long-term stable operation. It can be widely used in anti-counterfeiting and traceability management of various commodities and devices, as well as in visual positioning and target alignment scenarios for equipment. Attached Figure Description

[0013] Figure 1 This is a schematic diagram illustrating the principle of the image label manufacturing process in an embodiment of the present invention.

[0014] Figure 2This is a structural decomposition diagram of the image identifier prepared in an embodiment of the present invention.

[0015] Figure 3 This is a schematic diagram illustrating the installation of an image identifier embedded in a target object in an embodiment of the present invention.

[0016] Figure 4 This is a schematic diagram illustrating the principle of vertical and tilted image recognition in an embodiment of the present invention.

[0017] Figure 5 This is a flowchart illustrating the process of verifying authenticity via a smartphone app in an embodiment of the present invention.

[0018] In the picture 1. Molding mold cavity 2. Low-viscosity liquid transparent resin 3. Piezoelectric on-demand inkjet printhead; 4. Color micro-ink droplet particles. 5 Brownian motion stationary position; 6 High-power UV curing light source. 7. Solid substrate blank after curing 8. Transparent protective film layer 9. One-piece molded frame; 10. Mechanical locking buckle structure. 11. 3D image identification main body; 12. Target object reserved mounting slot. 13 The object to be inspected 14 Structural adhesive filler layer 15 Smartphone camera 16 Vertical shooting optical axis 17. Tilt Angle θ Shoot the optical axis at 18, launch the verification app, and grant camera permissions. 19. Image Acquisition via Vertical Image Capture 20. Image Preprocessing and Particle Detection 21 Extract particle coordinates and color features. 22 Select feature particles and generate feature codes. 23 Upload the serial number and feature code to the verification system. 24 Compare with the blockchain and complete routine verification. 25. Output conclusions based on the results or perform a second-order tilt verification. Detailed Implementation

[0019] To enable those skilled in the art to better understand the technical solutions of the present invention, a further detailed description will be provided below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the following description is merely illustrative of the principles and functions of the present invention.

[0020] Example 1: Image Label Manufacturing Step 1: Preparation of continuous phase liquid resin and mold forming. A UV-curable (2) low-viscosity liquid transparent resin with low viscosity at room temperature is prepared as the continuous phase medium. The resin formulation is optimized, with excellent flowability before curing, sufficient to allow micron-sized particles to undergo active Brownian motion within it; after curing, it has high transparency, excellent mechanical strength and dimensional stability. The prepared and fully degassed (2) low-viscosity liquid transparent resin is injected into a specially designed (1) molding mold cavity. The (1) molding mold cavity not only defines the length, width and thickness of the final solid substrate, but also has a ring of inwardly protruding cavities of a specific geometry around it for building the (9) integral molding frame. The cross-section of the (9) integral molding frame can be designed as a mortise and tenon structure, a T-shaped step or a (10) mechanical locking buckle structure with barbs to adapt to the embedding and fixing requirements of different objects.

[0021] Step 2: Implantation of 3D Color Ink Droplet Clusters. A sophisticated 3D motion control system is used, with a (3) piezoelectric on-demand inkjet printhead at its end. The (3) piezoelectric on-demand inkjet printhead is immersed below the surface of (2) low-viscosity liquid transparent resin. According to a preset spatial coordinate program, micro-level (4) color micro-droplet particles are sprayed into the resin at different time points, different horizontal positions, and different depth positions. The ink is formulated with solvents and nano-sized pigments that are immiscible with (2) low-viscosity liquid transparent resin, and can provide a variety of colors such as red, green, and blue. By precisely controlling the spraying timing and position, a suspended particle cluster composed of a large number of micro-droplets with a preliminary 3D distribution is formed inside the (2) low-viscosity liquid transparent resin.

[0022] Step 3: Controlled Brownian Motion and Instant Freezing. After the ink droplet ejection is completed, the entire device is transferred to the (5) Brownian motion station, kept stable and vibration-free, and given a very short but precisely measured period of stillness. During this period, all the suspended (4) colored micro-ink droplet particles undergo random Brownian motion driven by the thermal motion of liquid molecules. At a carefully set moment during the Brownian motion process, the curing process is triggered: the (6) high-power ultraviolet curing light source is turned on, and the high-intensity ultraviolet light uniformly irradiates the entire (2) low-viscosity liquid transparent resin layer. The resin is completely cured into a solid state in a very short time, perfectly "freezing" and permanently fixing the dynamic morphology, spatial coordinates and relative relationships of all (4) colored micro-ink droplet particles in the system at that instant, forming the (7) cured solid substrate blank.

[0023] Step 4: (8) Application of a transparent protective film. Clean the surface of the solid substrate blank after demolding (7) and then apply a transparent protective film layer (8) to its main appearance surface. This film layer can be formed by dip coating or spraying a transparent high-hardness UV-cured varnish, which is then cured by light again to form a seamless hard protective layer; or a transparent engineering film with anti-scratch and anti-reflection functions can be precisely attached. This transparent protective film layer (8) can ensure that the internal (4) colored micro-ink droplet particles are protected from physical and chemical damage during long-term use, and the four edges of the film layer are tightly fitted to the (1) molding mold cavity. (9) An integral molding frame is built outside the (1) molding mold cavity, and finally the (11) three-dimensional image logo body is obtained.

[0024] Example 2: Embedding and Installing Image Icons like Figure 3 As shown, (13) the surface of the object to be inspected has a precision-machined (12) pre-installed mounting groove, the outline of which precisely matches the (9) integrally molded frame of the (11) three-dimensional image identifier. During installation, a small amount of (14) structural adhesive filler is applied to the outer contact surface of the (9) integrally molded frame, and then the entire (11) three-dimensional image identifier is mechanically pressed into the groove, and the (10) mechanical locking buckle structure forms a firm interference or snap fit with the (12) pre-installed mounting groove. After the (14) structural adhesive filler has cured, the (11) three-dimensional image identifier and the (13) object to be inspected become an inseparable whole. Any unauthorized attempt to disassemble will result in structural damage to the identifier or object, further increasing the difficulty of forgery and transfer.

[0025] Example 3: Image Marker Capture and Recognition When consumers or inspectors verify the authenticity of the (11) three-dimensional image marker embedded in (13) the object to be inspected, the user first executes (18) to start the verification APP and enable camera permissions, and points the (15) smartphone camera at the (11) three-dimensional image marker. The APP interface displays a shooting guide box. The user adjusts the angle of the phone so that the optical axis of the (15) smartphone camera is perpendicular to the marker surface, that is, along (16) the vertical shooting optical axis to complete (19) vertical shooting of the image marker to obtain an image, and obtain a high-resolution macro image. Since the substrate is transparent, the (15) smartphone camera can clearly capture the (4) colored micro-ink droplet particles solidified inside the substrate. This vertical image records the precise coordinates and color information of all particles on the two-dimensional projection surface.

[0026] The built-in algorithm of the APP then analyzes and processes the image: First, it completes the image preprocessing and particle detection by color threshold segmentation and spot detection algorithm (20), and automatically identifies all (4) colored micro-ink droplet particles in the image; then it performs (21) to extract particle coordinates and color features, and obtains the centroid coordinates and color classification of each particle; then it performs (22) to select feature particles and generate feature codes according to preset rules, such as using regional grid sampling method, particle spacing threshold screening or color distribution ratio selection, to ensure that the selected particle group is representative and stable, and then the coordinate and color data are standardized and sorted and generated by hash operation to generate fixed-length feature codes.

[0027] Subsequently, (23) the serial number and feature code are uploaded to the verification system. The APP sends the feature code and (11) the unique serial number of the 3D image identifying the subject to the cloud verification server through a secure encrypted channel. The verification server calls the blockchain smart contract, executes (24) blockchain comparison and completes the routine verification judgment, and performs a cryptographic comparison between the on-site feature code and the registration hash value corresponding to the serial number on the blockchain. The registration hash value is generated and stored on the chain in the same process when the product is manufactured, and has the characteristics of being tamper-proof and permanently traceable.

[0028] Finally, execute (25) to output the conclusion based on the result or perform a second tilt verification: if the comparison is consistent, output the conclusion that it is genuine; if it is inconsistent, output a suspicious warning; if the matching score is in the doubtful range or is a high-security requirement scenario, the APP guides the user to switch the (15) smartphone camera to the (17) tilt angle. θ The system captures a tilted image by supplementing the optical axis, performs stereo matching on the dual-view images, calculates the relative depth information of the (4) color micro-ink droplet particles, combines the depth features to complete the secondary verification and output the final judgment result.

[0029] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A three-dimensional image identifier, characterized in that, include: A transparent solid substrate is formed by instantaneous curing of (2) low-viscosity liquid transparent resin; a plurality of (4) colored micro-ink droplet particles are dispersed and solidified in the three-dimensional space inside the solid substrate, the spatial position of the plurality of (4) colored micro-ink droplet particles and the relative spatial relationship between the particles are a random distribution pattern with physical non-cloning function characteristics, which is instantaneously solidified and fixed in the Brownian motion state in the liquid medium; a (8) transparent protective film layer covering at least one surface of the solid substrate; a (9) integrally formed frame surrounding and integrally formed around the solid substrate, used to embed and fix the image mark in the (12) target object reserved mounting groove of the (13) target object body to be inspected.

2. The three-dimensional image identifier according to claim 1, characterized in that, The solid substrate is made of a transparent resin material that is cured by ultraviolet light or by heat, and the (8) transparent protective film layer is a transparent coating or a transparent film with high hardness and scratch resistance.

3. The three-dimensional image identifier according to claim 1, characterized in that, The (9) integrally molded frame is integrally cured in the (1) molding mold cavity with the same resin material as the solid substrate, or the material is metal or plastic and is fixed to the substrate by secondary injection molding or gluing. The (9) integrally molded frame has a (10) mechanical locking buckle structure for mechanically locking with external objects.

4. The three-dimensional image identifier according to claim 1, characterized in that, The (4) colored micro-ink droplet particles are composed of particles of colored nano-pigments that are immiscible with the (2) low-viscosity liquid transparent resin. They contain particles of at least two different colors, and the color of the particles serves as an auxiliary feature for particle differentiation and position matching.

5. A method for manufacturing a three-dimensional image identifier as described in any one of claims 1-4, characterized in that, Includes the following steps: a) Prepare a fast-curing (2) low-viscosity liquid transparent resin as a continuous phase and inject it into the (1) molding mold cavity, wherein the inner shape of the (1) molding mold cavity corresponds to the overall shape of the solid substrate with the (9) integral molding frame; b) Spray at least one (4) color micro-droplet particles in the form of micro-droplets into the (2) low-viscosity liquid transparent resin at different time points and different spatial positions through the (3) piezoelectric on-demand inkjet printhead to form a suspended three-dimensional micro-droplet particle group; c) Place the whole system in the (5) Brownian motion stationary position, and in the static state of the system, allow the (4) color micro-droplet particle group to undergo Brownian motion in the liquid resin for a period of time. d) At a certain moment during the Brownian motion, turn on the (6) high-power ultraviolet curing light source to instantly cure the (2) low-viscosity liquid transparent resin, permanently freezing the motion posture, spatial position and random distribution pattern of the (4) colored micro-ink droplet particles, forming the (7) solid substrate blank after curing. In addition, the Brownian motion duration and curing timing in steps c) and d) are precisely controlled by the system program to ensure that each manufactured mark has unique and unreplicable particle spatial distribution characteristics; e) Apply the (8) transparent protective film layer to at least one surface of the (7) solid substrate blank after curing to obtain the (11) three-dimensional image mark body.

6. A method for authenticity identification using a three-dimensional image identifier as described in any one of claims 1-4, characterized in that, The following steps are executed sequentially: (18) Start the verification APP and enable camera permissions; (19) Capture the image identifier to obtain the image; (20) Image preprocessing and particle detection; (21) Extract particle coordinates and color features; (22) Select feature particles and generate feature codes; (23) Upload the serial number and feature code to the verification system; (24) Compare the blockchain and complete the routine verification judgment; (25) Output the conclusion according to the result or perform secondary tilt verification.

7. The method according to claim 6, characterized in that, When (24) the result of blockchain comparison and routine verification is in the doubtful range, or when the system determines that it is a necessary situation such as a high-security requirement scenario, a secondary confirmation process is automatically triggered, including: controlling (15) the smartphone camera along (17) tilt angle. θ A tilted view image is captured to supplement the optical axis; a stereo matching is performed on the vertical view image and the tilted view image to identify the correspondence of the same (4) colored micro-ink droplet particles in the two images, and the relative depth information of at least some particles is calculated based on the parallax principle; the relative depth information is used as an auxiliary supplementary feature, the feature code is updated and the blockchain comparison is performed again, or the depth sorting or layer identification is sent to the verification system together with the vertical feature code for comprehensive judgment.

8. The method according to claim 6, characterized in that, The preset rules for selecting feature particles and generating feature codes in (22) include: sampling according to coordinate grid partitions, filtering according to particle spacing thresholds, or selecting according to color distribution ratios; the feature codes are generated by hashing the coordinate sequence and color values ​​of the feature particles; the registration feature codes stored on the blockchain are hash values ​​generated and stored on the blockchain using the same vertical shooting method and feature extraction method when the (11) three-dimensional image identification subject leaves the factory; if there is secondary confirmation registration information at the same time, it also includes the depth reference features obtained by shooting the optical axis at a certain (17) tilt angle when leaving the factory.

Images