WO2023071159A1 - 基于内嵌识别码多工艺适用的3d打印制品防伪方法 - Google Patents

基于内嵌识别码多工艺适用的3d打印制品防伪方法 Download PDF

Info

Publication number
WO2023071159A1
WO2023071159A1 PCT/CN2022/094331 CN2022094331W WO2023071159A1 WO 2023071159 A1 WO2023071159 A1 WO 2023071159A1 CN 2022094331 W CN2022094331 W CN 2022094331W WO 2023071159 A1 WO2023071159 A1 WO 2023071159A1
Authority
WO
WIPO (PCT)
Prior art keywords
embedded
identification code
printed
printing
counterfeiting
Prior art date
Application number
PCT/CN2022/094331
Other languages
English (en)
French (fr)
Inventor
陈广学
田婕妮
袁江平
王俪儒
姚丹阳
Original Assignee
华南理工大学
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华南理工大学 filed Critical 华南理工大学
Publication of WO2023071159A1 publication Critical patent/WO2023071159A1/zh

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B29—WORKING OF PLASTICS; WORKING OF SUBSTANCES IN A PLASTIC STATE IN GENERAL
    • B29C—SHAPING OR JOINING OF PLASTICS; SHAPING OF MATERIAL IN A PLASTIC STATE, NOT OTHERWISE PROVIDED FOR; AFTER-TREATMENT OF THE SHAPED PRODUCTS, e.g. REPAIRING
    • B29C64/00—Additive manufacturing, i.e. manufacturing of three-dimensional [3D] objects by additive deposition, additive agglomeration or additive layering, e.g. by 3D printing, stereolithography or selective laser sintering
    • B29C64/10—Processes of additive manufacturing
    • B29C64/188—Processes of additive manufacturing involving additional operations performed on the added layers, e.g. smoothing, grinding or thickness control
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B29—WORKING OF PLASTICS; WORKING OF SUBSTANCES IN A PLASTIC STATE IN GENERAL
    • B29C—SHAPING OR JOINING OF PLASTICS; SHAPING OF MATERIAL IN A PLASTIC STATE, NOT OTHERWISE PROVIDED FOR; AFTER-TREATMENT OF THE SHAPED PRODUCTS, e.g. REPAIRING
    • B29C64/00—Additive manufacturing, i.e. manufacturing of three-dimensional [3D] objects by additive deposition, additive agglomeration or additive layering, e.g. by 3D printing, stereolithography or selective laser sintering
    • B29C64/30—Auxiliary operations or equipment
    • B29C64/386—Data acquisition or data processing for additive manufacturing
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B29—WORKING OF PLASTICS; WORKING OF SUBSTANCES IN A PLASTIC STATE IN GENERAL
    • B29C—SHAPING OR JOINING OF PLASTICS; SHAPING OF MATERIAL IN A PLASTIC STATE, NOT OTHERWISE PROVIDED FOR; AFTER-TREATMENT OF THE SHAPED PRODUCTS, e.g. REPAIRING
    • B29C64/00—Additive manufacturing, i.e. manufacturing of three-dimensional [3D] objects by additive deposition, additive agglomeration or additive layering, e.g. by 3D printing, stereolithography or selective laser sintering
    • B29C64/30—Auxiliary operations or equipment
    • B29C64/386—Data acquisition or data processing for additive manufacturing
    • B29C64/393—Data acquisition or data processing for additive manufacturing for controlling or regulating additive manufacturing processes
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B29—WORKING OF PLASTICS; WORKING OF SUBSTANCES IN A PLASTIC STATE IN GENERAL
    • B29C—SHAPING OR JOINING OF PLASTICS; SHAPING OF MATERIAL IN A PLASTIC STATE, NOT OTHERWISE PROVIDED FOR; AFTER-TREATMENT OF THE SHAPED PRODUCTS, e.g. REPAIRING
    • B29C64/00—Additive manufacturing, i.e. manufacturing of three-dimensional [3D] objects by additive deposition, additive agglomeration or additive layering, e.g. by 3D printing, stereolithography or selective laser sintering
    • B29C64/40—Structures for supporting 3D objects during manufacture and intended to be sacrificed after completion thereof
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B33—ADDITIVE MANUFACTURING TECHNOLOGY
    • B33Y—ADDITIVE MANUFACTURING, i.e. MANUFACTURING OF THREE-DIMENSIONAL [3-D] OBJECTS BY ADDITIVE DEPOSITION, ADDITIVE AGGLOMERATION OR ADDITIVE LAYERING, e.g. BY 3-D PRINTING, STEREOLITHOGRAPHY OR SELECTIVE LASER SINTERING
    • B33Y10/00—Processes of additive manufacturing
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B33—ADDITIVE MANUFACTURING TECHNOLOGY
    • B33Y—ADDITIVE MANUFACTURING, i.e. MANUFACTURING OF THREE-DIMENSIONAL [3-D] OBJECTS BY ADDITIVE DEPOSITION, ADDITIVE AGGLOMERATION OR ADDITIVE LAYERING, e.g. BY 3-D PRINTING, STEREOLITHOGRAPHY OR SELECTIVE LASER SINTERING
    • B33Y50/00—Data acquisition or data processing for additive manufacturing
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B33—ADDITIVE MANUFACTURING TECHNOLOGY
    • B33Y—ADDITIVE MANUFACTURING, i.e. MANUFACTURING OF THREE-DIMENSIONAL [3-D] OBJECTS BY ADDITIVE DEPOSITION, ADDITIVE AGGLOMERATION OR ADDITIVE LAYERING, e.g. BY 3-D PRINTING, STEREOLITHOGRAPHY OR SELECTIVE LASER SINTERING
    • B33Y50/00—Data acquisition or data processing for additive manufacturing
    • B33Y50/02—Data acquisition or data processing for additive manufacturing for controlling or regulating additive manufacturing processes
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B33—ADDITIVE MANUFACTURING TECHNOLOGY
    • B33Y—ADDITIVE MANUFACTURING, i.e. MANUFACTURING OF THREE-DIMENSIONAL [3-D] OBJECTS BY ADDITIVE DEPOSITION, ADDITIVE AGGLOMERATION OR ADDITIVE LAYERING, e.g. BY 3-D PRINTING, STEREOLITHOGRAPHY OR SELECTIVE LASER SINTERING
    • B33Y80/00—Products made by additive manufacturing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/23—Design optimisation, verification or simulation using finite element methods [FEM] or finite difference methods [FDM]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods

Definitions

  • the invention belongs to the technical field of anti-counterfeiting of 3D printed products, relates to 3D printing processes of various substrates, and in particular relates to an anti-counterfeiting method of 3D printed products based on embedded identification codes applicable to multiple processes.
  • 3D printing technology uses layer-by-layer accumulation process to process discrete materials into entities. It was originally used as a high-tech technology in automobiles, aviation, military and other fields. With the emergence of abundant 3D printing materials and the development of new 3D printing processes, people have a deeper understanding of this technology. 3D printing technology has moved from industrial applications to cultural creativity, medicine, construction and other industries, and even embarked on people's table. The popularity of 3D printing technology among the public makes the problem of anti-counterfeiting of 3D printed products urgently to be solved.
  • Material anti-counterfeiting strategy and digital anti-counterfeiting strategy are helpful to solve the embedded anti-counterfeiting problem of 3D printing products.
  • the material anti-counterfeiting strategy is aimed at 3D printing products whose printing raw materials are liquid. By embedding quantum dots, up-conversion nanoparticles and other excitatory response materials in the printing raw material, the printed products themselves have anti-counterfeiting properties, which can be identified by measuring instruments such as spectrometers. A special class of embedded substances.
  • this anti-counterfeiting scheme has great restrictions on the transparency of the printing process and printing materials; secondly, the heat resistance requirements of the labeling agent also limit the selection of the labeling agent to a certain extent.
  • the digital anti-counterfeiting strategy is not limited to the printing method, but uses the inherent digital features of the printed products of the same printer, such as hardware defects, audio signal signatures, vibration marks, etc., to judge the printed products by extracting and classifying these digital features
  • the establishment of a considerable digital feature database requires multi-departmental promotion and cooperation to complete data sharing.
  • embedding digital watermarks or copyright information into 3D printed products can also identify the authenticity of printed products.
  • researchers have proposed a series of solutions for embedding and extracting watermarks into 3D printed products. The robustness is not strong, it can only resist one or several types of attacks, and the extraction efficiency is relatively low.
  • the present invention overcomes the deficiencies of the prior art, and proposes an anti-counterfeiting method for 3D printed products based on an embedded identification code that is applicable to multiple processes, is applicable to multiple printing processes, and solves the problems of weak robustness and poor extraction of 3D digital watermarks.
  • the anti-counterfeiting method for 3D printing products based on the multi-process embedded identification code includes the following steps:
  • the model is created by ANSYS finite element analysis software, the properties of the printing material are defined, the elastic modulus and Poisson's ratio of the material are set, the boundary constraint conditions are set after the model is meshed, the solution is applied to different parts after applying the same load, and the printing is observed.
  • the degree of deformation of the part find the place that has the least impact on the mechanical properties of the printed part and position it as A1; for crafts that do not require mechanical properties, avoid the fine workmanship section and choose the part that has the least impact on the appearance and position it as A1.
  • the modeling software described in step b is Autodesk 3ds Max, and different embedded thicknesses, embedded depths and embedded sizes are selected according to different 3D printing processes.
  • the embedded thickness refers to the parallel distance between the upper surface and the lower surface of the hollow identification code.
  • the embedded thickness refers to the distance between the upper surface and the lower surface of the hollow identification code.
  • step c the ".max" model format established in Autodesk 3ds Max software is converted into ".stl” format, and then converted into a format recognizable by the printer through the printer supporting software, and input into the printer to obtain a print.
  • the identification code is an anti-counterfeiting pattern that can be converted by modeling software so that it has a 3D embedded part with a certain thickness in the Z-axis direction, such as QR code, trademark code, graphic code, anti-counterfeiting code, tracking code.
  • the identification code is fully embedded at one time by the overall embedding method or for small-sized prints with a size less than 50 mm, the identification code is divided into multiple segments by segmented embedding, and then some of the code segments are reversed. Embedded to different depths of the print to increase its anti-counterfeiting properties.
  • the features of the original image are extracted autonomously through a deep learning algorithm, and the image is automatically restored.
  • the deep learning algorithm can independently extract the original image features and automatically restore the image.
  • the deep learning algorithm does not need to change algorithm parameters frequently, and shortens the verification time.
  • a segmented mortise and tenon splicing structure is used, and the identification code is embedded in the splicing part; or during the printing process, the equipment is temporarily stopped after each inner layer is cut, and manually removed After the identification code part, restart the equipment to carry out the sizing process; for the hollow structure of the powder-based parts printed by the binder jetting process, fill the hollow part with powder, and remove the powder through the reserved holes in the post-processing process; for the fused deposition process
  • the printed hollow structure is a water-soluble support frame added inside, and the support frame is dissolved and removed during post-processing.
  • the invention embeds the hollow identification code inside the product, uses the advantages of 3D printing to integrate the anti-counterfeiting and the product, hides the identification code inside the product and is invisible from the outside, this process does not need to add special chemical labeling agents, and also Not limited to 3D printing technology and printing materials, the embedded identification code is captured by industrial CT scanning, and the authenticity information of the product can be obtained after algorithmic processing of the original image.
  • the beneficial effects of the present invention are as follows: :
  • the present invention makes full use of the advantages of 3D printing technology, combines the traditional anti-counterfeit label with the depth of the product, the identification code and the product are integrally formed, and the identification code is hidden inside, avoiding matching the anti-counterfeit label to the product
  • the outer packaging is easy to wear and risk of being replaced.
  • the present invention is not limited by the printing process, and any printing method that can print a hollow structure or indirectly print a hollow structure can realize this anti-counterfeiting solution without adding Special materials do not need to consider the failure conditions of the added response materials, and at the same time avoid the toxicity and recycling problems of materials, which help to solve the anti-counterfeiting problems caused by the gradual popularity of 3D printing products.
  • the present invention can resist various post-processing attacks including surface wear, dipping, gluing, etc., and the identification hidden inside can be captured through a 5-min industrial CT scan code, the identification code extraction efficiency is high, and the algorithm is automatically processed, saving time and effort.
  • the multi-process anti-counterfeiting method applicable to 3D printed products based on the embedded identification code can not be limited to multiple printing methods, and helps to solve the anti-counterfeiting problems existing in the market where multiple 3D printing technologies coexist.
  • the multi-process anti-counterfeiting method applicable to 3D printing products based on the embedded identification code of the present invention can not only adapt to various printing processes, but also solve the key problems of weak robustness and difficult extraction of 3D digital watermarks, which greatly improves the market Anti-counterfeiting protection measures for 3D printing products flowing on the Internet, escorting the further promotion of 3D printing industrialization.
  • Fig. 1 is a flow chart of the anti-counterfeiting method of the present invention.
  • Figure 2 is a schematic diagram of the operation of industrial CT scanning to capture the hidden internal identification code, where 1 is the radiation source, 2 is the print, 3 is the rotating stage, and 4 is the detector.
  • Fig. 3 is a frame diagram of the post-processing algorithm of the present invention.
  • Figure 1 shows the flow chart of the anti-counterfeiting method applicable to 3D printed products based on the multi-process embedded identification code.
  • the original image of the hollow identification code is obtained through industrial CT, and the authenticity of the printed matter is identified by decoding with a restoration algorithm.
  • the anti-counterfeiting method for 3D printing products based on the multi-process embedded identification code comprises the following steps:
  • the embedded thickness can be 0.1mm.
  • the thickness of a single layer of paper is not enough to be completely captured during scanning, so the embedded thickness needs to be more than 0.5mm.
  • the optional embedded thickness range of different processes is 0.1-1.5mm; the embedded depth is determined by the thickness of the printed product itself, and the optional range is 1-10mm; the embedded size is determined by the printing process and the part size, in which the printing process determines the minimum size of the identification code, and the part size determines the identification code
  • the maximum size range is 3*3-30*30mm, that is, for high-precision processes such as metal 3D printing, the identification code size can be 3*3mm, and the identification code size for FDM printing wood-plastic wire can be 10*10mm , the size of the identification code for desktop paper-based 3D printing can be selected as 30*30mm, and the overall volume of the identification code must be less than 5% of the overall volume of the part, subject to the complete identification code captured by CT scanning.
  • the front of the identification code is parallel to the scanning radiation source, the distance is 70mm, and the rotation angle is 360°.
  • Set the accelerator energy to 6 Mev select the ordinary CT scanning mode, and scan the time for 5-15 minutes to obtain a three-dimensional reconstruction model, and slice the reconstructed model in layers.
  • the slice plane is perpendicular to the embedding direction of the identification code, and the identification A sliced image of the code.
  • the original identification code and the error code similar to the original code are set as the training set to train the convolutional neural network, and the images obtained by CT scanning are used as the test set.
  • the network is used to process sliced images, and the deep learning algorithm is used to independently extract the original image features and automatically Restore the image.
  • the splicing structure selects the dovetail tenon in the mortise and tenon structure, and specifies that the identification code be embedded in the joint of the tenon and tenon structure, and can be hidden after printing is completed through tenon and tenon splicing;
  • the embedded thickness is 0.5mm; the embedded depth is 3mm; the embedded size is 30*30mm;
  • the embedded 3D model is converted from ".max” format to ".stl” format in Autodesk 3ds Max , and then use the Mcor Orange software to convert the ".stl” format into the ".mcor” format recognized by the McorArke full-color 3D printer.
  • the hollow printing scheme is realized by adding 6 water-soluble support frames inside, by reserving holes and water-soluble post-processing;
  • the printer is the Cube Pro printer of 3D system company;
  • the supporting material is Infinity Rinse-Away water-soluble supporting material, and the printing material of the product is PLA wire;
  • the printed part is a cube, and the embedding direction of the pattern code is positioned on the top surface after evaluation by ANSYS software;
  • the 3D model after embedding is converted into ".stl” format, and the ".stl” format is converted by Cubify Sculpt software
  • the model is uploaded to the Cube Pro printer.
  • the parts are dissolved in water, the support is removed, and the model is closed after forming a hollow structure. Then put it on the scanning table of Shimadzu SMX-225 industrial CT. The scanning time is 5 minutes. After the scanning is completed, the original image is obtained, and the image is restored through the neural convolution network to identify the authenticity of the product.
  • the hollow printing scheme is realized by filling the liquid raw material inside, and the opening size of the process hole is recommended to be at least 3mm;
  • the printer is the SprintRay Pro printer of SprintRay;
  • the printing material is Die and Model resin;
  • the printed part is a regular tetrahedron, and the embedding direction of the pattern code is positioned on the front after evaluation by ANSYS software;
  • the embedding thickness is 0.1mm;
  • the embedding depth is 0.5mm ;
  • the embedded size is 4*4mm; Convert the embedded 3D model into ".stl” format, upload the ".stl” format to the SprintRay Pro printer through RayWare software, and pass the process hole after printing
  • the internal liquid is poured out to form a hollow structure and then close the model. Then put it on the scanning table of Shimadzu SMX-225 industrial CT.
  • the scanning time is 5 minutes. After the scanning is completed, the original image is obtained, and the image is restored through the neural convolution

Landscapes

  • Engineering & Computer Science (AREA)
  • Materials Engineering (AREA)
  • Physics & Mathematics (AREA)
  • Chemical & Material Sciences (AREA)
  • Manufacturing & Machinery (AREA)
  • Theoretical Computer Science (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Optics & Photonics (AREA)
  • Mechanical Engineering (AREA)
  • General Physics & Mathematics (AREA)
  • Artificial Intelligence (AREA)
  • Software Systems (AREA)
  • Molecular Biology (AREA)
  • Mathematical Physics (AREA)
  • General Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Computing Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Computational Linguistics (AREA)
  • Health & Medical Sciences (AREA)
  • Biophysics (AREA)
  • Biomedical Technology (AREA)
  • Geometry (AREA)
  • Computer Hardware Design (AREA)
  • Medical Informatics (AREA)
  • Computer Vision & Pattern Recognition (AREA)

Abstract

本发明公开了一种基于内嵌识别码多工艺适用的3D打印制品防伪方法，属于3D打印制品防伪技术领域；先确定对3D打印件力学性能影响最小的位置或对外观影响最小的位置为A1；通过建模软件以内嵌的方式将立体的中空识别码嵌入到A1处；转换成打印机可识别的格式，打印得到打印件；对打印件进行CT扫描，得到三维重建的模型，对重建的模型进行分层切片，获取带有识别码的切片图像；使用网络处理切片图像，自主提取原始图像特征并自动还原图像；本发明将中空的识别码内嵌在产品内部，利用3D打印的优势将防伪和产品一体化，可以适应多种打印工艺，可以解决3D数字水印鲁棒性弱、不好提取的问题。

Description

基于内嵌识别码多工艺适用的3D打印制品防伪方法
本申请要求于2021年10月25日提交中国专利局、申请号为202111242179.8、发明名称为“基于内嵌识别码多工艺适用的3D打印制品防伪方法”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本发明属于3D打印制品防伪技术领域，涉及多种基材的3D打印工艺，具体涉及一种基于内嵌识别码多工艺适用的3D打印制品防伪方法。
背景技术
3D打印技术利用逐层堆积的工艺将离散化的材料加工成实体，最初作为高精尖技术应用于汽车、航空、军事等领域。随着丰富的3D打印材料的出现和新的3D打印工艺的开发，人们对这一技术的理解愈发深刻，3D打印技术从工业应用走向了文化创意、医药、建筑等行业，甚至走上了人们的餐桌。3D打印技术在大众间的流行使得3D打印制品的防伪问题亟待解决。
材料防伪策略和数字防伪策略有助于3D打印制品的内嵌防伪问题的解决。材料防伪策略针对打印原料为液态的3D打印制品，通过在打印原浆中嵌入量子点、上转换纳米粒子等激发性响应材料，使得打印制品本身具有防伪性能，通过光谱仪等测定仪器可以识别到这类特殊的内嵌物质。但这种防伪方案对于打印工艺及打印材料的透明度限制较大；其次，标签剂耐热性能要求也在一定程度上限制了标签剂的选择范围。数字防伪策略不局限于打印方式，而是利用同一台打印机打印制品的固有数字特征如硬件缺陷、音频信号签名、振动记号标记等，通过对这些数字特征的提取归类建立模型，来判断打印制品的来源和真伪，但建立一个可观的数字特征数据库需要多部门推动合作，完成数据共享。此外，向3D打印制品内嵌入数字水印或者版权信息也可以鉴定打印制品的真伪，研究人员提出了一系列向3D打印品中嵌入水印和提取水印的方案，但目前面临的普遍问题是水印的鲁棒性不强，只能抵抗某一种或某几种的攻击行为，且提取效率较为低下。
发明内容
本发明克服了现有技术的不足，提出一种基于内嵌识别码多工艺适用的3D打印制品防伪方法，适用多种打印工艺，解决3D数字水印鲁棒性弱、不好提取的问题。
为了达到上述目的，本发明是通过如下技术方案实现的：
基于内嵌识别码多工艺适用的3D打印制品防伪方法，包括以下步骤：
a)确定对3D打印件力学性能影响最小的位置或对外观影响最小的位置为A1；
b)通过建模软件以内嵌的方式将立体的中空识别码嵌入到A1处；
c)转换成打印机可识别的格式，输入打印机得到打印件，使打印件内部形成中空结构的识别码；
d)对打印件进行CT扫描，得到三维重建的模型，对重建的模型进行分层切片，切片平面与识别码内嵌方向垂直，获取带有识别码的切片图像；
e)设定原始识别码和与原始码相近的错误码为训练集训练卷积神经网络，CT扫描得到的图像为测试集，使用网络处理切片图像，自主提取原始图像特征并自动还原图像。
优选的，通过ANSYS有限元分析软件创建模型，定义打印材料属性，设定材料的弹性模量和泊松比，对模型划分网格后设置边界约束条件，对不同部位施加相同载荷后求解，观察打印件的变形程度，寻找对打印件力学性能影响最小的地方定位为A1；对于无力学性能要求的工艺品，避开精细做工的区段而选择对外观影响最小的部位定位为A1。
优选的，步骤b所述的建模软件为Autodesk 3ds Max，并根据不同的3D打印工艺选择不同的内嵌厚度、内嵌深度和内嵌尺寸。
更优的，所述内嵌厚度指中空的识别码上表面和下表面之间的平行距离，对于弯曲处的内嵌识别码，所述内嵌厚度指中空的识别码上表面和下表面之间的平均距离；所述内嵌深度指中空的识别码下表面距零件上表面之间的平均距离；所述内嵌尺寸指的是识别码的长宽。
更优的，步骤c中，将在Autodesk 3ds Max软件中建立的“.max”模型格式转换成“.stl”格式，再通过打印机配套软件转换成打印机可识别的格 式，输入打印机得到打印件。
优选的，所述识别码为能通过建模软件转换使其在Z轴方向上有一定厚度的3D的内嵌置入零件的防伪图案，例如QR码、商标码、图形码、防伪码、追踪码。
优选的，采用整体内嵌的方式一次性完整嵌入识别码或对于尺寸小于50mm的小尺寸打印件，采用分段内嵌的方式，将识别码分割成多段，再将其中部分码段进行翻转，内嵌到打印件的不同深度，以提高其防伪性能。
优选的，通过深度学习算法自主提取原始图像特征，自动还原图像。所述深度学习算法可自主提取原始图像特征，自动还原图像。针对传统图像识别算法中需要图像对比度调整，降噪和二值化等一系列繁琐且人为地设置和操作过程，深度学习算法无需频繁更改算法参数，并缩短了验证时间。
优选的，对于纸基大型拼接件的3D打印，采用分段榫卯拼接结构，在拼接部位嵌入识别码；或在打印过程中，在每个内嵌层切割完毕后暂时停止设备，手动移除识别码部分后再启动设备进行施胶工艺；对于粘合剂喷射工艺打印的粉基零件的中空结构，用粉末填充中空部位，在后处理过程中通过预留的孔洞去除粉末；对于熔融沉积工艺打印的中空结构，是在内部增加水溶性的支撑架，在后处理过程中溶解去除支撑架。
本发明将中空的识别码内嵌在产品内部，利用3D打印的优势将防伪和产品一体化，将识别码隐藏于产品内部而在外观不可见，这一过程无需添加特殊的化学标签剂，也不局限于3D打印工艺和打印材料，通过工业CT扫描捕捉到内嵌的识别码，对原始图像进行算法处理后可以获得产品的真伪信息，本发明相对于现有技术所产生的有益效果为：
(1)本发明充分利用起3D打印技术的优势，将传统意义上的防伪标签和产品深度结合起来，识别码和产品一体成型，且识别码在隐藏在内部，避免了将防伪标签匹配到产品的外包装上容易磨损、被替换的风险。
(2)本发明与目前的3D打印制品内嵌响应材料防伪相比，不受打印工艺的限制，任何能打印中空结构或间接打印成中空结构的打印方式都可以实现这一防伪方案，不用添加特殊的材料，也不用考虑添加的响应材料的失效条件，同时避免了材料的毒性和回收问题，有助于解决3D打印 制品逐渐流行造成的防伪问题。
(3)本发明与目前的3D打印制品内嵌式数字防伪相比，可以抵抗包括表面磨损、浸渍、涂胶等多种后处理攻击，通过5min的工业CT扫描可以捕捉到隐藏在内部的识别码，识别码提取效率高，算法自动处理，省时省力。
所述基于内嵌识别码的多工艺适用3D打印制品的防伪方法能够不局限于多种打印方式，有助于解决如今多种3D打印技术共存的市场上存在的防伪问题。本发明的基于内嵌识别码的多工艺适用3D打印制品的防伪方法，既可以适应多种打印工艺，又能解决3D数字水印鲁棒性弱、不好提取的关键问题，极大地完善了市场上流动的3D打印制品的防伪保护措施，为进一步推动3D打印产业化保驾护航。
说明书附图
为了更清楚地说明本发明实施例或现有技术中的技术方案，下面将对实施例中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动性的前提下，还可以根据这些附图获得其他的附图。
图1是本发明所述防伪方法的流程图。
图2是工业CT扫描捕捉隐藏在内部识别码的操作示意图，其中1为辐射源，2为打印件，3为旋转载物台，4为探测器。
图3是本发明后处理算法的框架图。
具体实施方式
为了使本发明所要解决的技术问题、技术方案及有益效果更加清楚明白，结合实施例和附图，对本发明进行进一步详细说明。应当理解，此处所描述的具体实施例仅仅用以解释本发明，并不用于限定本发明。下面结合实施例及附图详细说明本发明的技术方案，但保护范围不被此限制。
图1给出了基于内嵌识别码的多工艺适用3D打印制品的防伪方法的流程图，其特征在于，真正利用起3D打印的优势，将防伪识别码和产品真正结合统一起来，而不是看成相互匹配的两部分，通过工业CT获取中空识别码的原始图像，经还原算法解码来辨别打印品的真伪。基于内嵌识 别码的多工艺适用3D打印制品的防伪方法包括以下步骤：
首先，对于有力学性能要求的零件，通过ANSYS有限元分析软件创建模型，定义材料属性，设定材料的弹性模量和泊松比，对模型划分网格后设置边界约束条件，对不同部位施加相同载荷后求解观察打印件的变形程度，寻找对打印件力学性能影响最小的地方定位为A1；对于无力学性能要求的工艺品，可避开精细做工的区段而选择对外观影响最小的部位定位为A1。
其次通过Autodesk 3ds Max建模软件以中空内嵌的方式将立体的识别码嵌入到A1处，根据不同的打印工艺选择不同的内嵌厚度，对于打印精度更高的工艺如SLA打印光固化树脂，内嵌厚度可以为0.1mm，对于纸基打印模型，单层纸的厚度不足以在扫描时被完整捕捉到，则需要使内嵌厚度在0.5mm以上，不同工艺可选择的内嵌厚度范围为0.1-1.5mm；内嵌深度由打印品自身厚度决定，可选择范围为1-10mm；内嵌尺寸由打印工艺和零件尺寸共同决定，其中打印工艺决定识别码的最小尺寸，零件尺寸决定识别码的最大尺寸，尺寸范围为3*3-30*30mm，即精度较高的工艺如金属3D打印，其识别码大小可选择3*3mm，FDM打印木塑线材时识别码大小可以选择10*10mm，桌面纸基3D打印的识别码大小可选择30*30mm，识别码整体体积需小于零件整体体积的5％，以可用CT扫描捕捉到完整识别码为准。
然后将在Autodesk 3ds Max软件中建立的“.max”模型格式转换成“.stl”格式，再通过打印机配套软件转换成打印机可识别的格式，输入打印机得到打印件。
然后对打印完成的打印制品进行后处理，使得打印件形成内部中空结构的识别码。
然后将打印件放置在岛津SMX-225工业CT设备的扫描台上，识别码的正面与扫描辐射源平行，距离为70mm，旋转角度为360°，采用设备自带的夹持器固定，扫描时设定加速器能量为6Mev，选择普通CT扫描模式，扫描时间为5-15min，得到三维重建的模型，对重建的模型进行分层切片，切片平面与识别码内嵌方向垂直，获取带有识别码的切片图像。
最后设定原始识别码和与原始码相近的错误码为训练集训练卷积神经网络，CT扫描得到的图像为测试集，使用该网络处理切片图像，采用深度学习算法自主提取原始图像特征并自动还原图像。
实施例1
若打印纸基大尺寸拼接3D模型：所述拼接结构选择榫卯结构中的燕尾榫，指定将识别码嵌入到榫卯结构的连接处，打印完成经榫卯拼合后可实现隐藏；所述内嵌厚度为0.5mm；所述内嵌深度为3mm；所述内嵌尺寸为30*30mm；在Autodesk 3ds Max中将完成内嵌后的3D模型由“.max”格式转换成“.stl”格式，再通过Mcor Orange软件将“.stl”格式转换成McorArke全彩3D打印机可识别的“.mcor”格式，打印完成后用镊子对零件进行去废后处理，使其形成中空结构的识别码。再将其置于岛津SMX-225工业CT的扫描台上，所述扫描时间为5min，扫描完成获取原始图像，通过神经卷积网络还原图像，可以鉴定产品的真伪。
实施例2
若采用熔融沉积工艺打印3D模型：所述中空打印方案通过在内部增加6个水溶性支撑架，通过预留孔洞、水溶后处理来实现；所述打印机为3D system公司的Cube Pro打印机；所述支撑材料为Infinity Rinse-Away水溶性支撑材料，所述产品打印材料为PLA线材；所述打印件为正方体，经ANSYS软件评估后将图案码的内嵌方向定位在顶面；所述内嵌厚度为0.3mm；所述内嵌深度为1mm；所述内嵌尺寸为5*5mm；将完成内嵌后的3D模型转换成“.stl”格式，在通过Cubify Sculpt软件将“.stl”格式的模型上传到Cube Pro打印机，打印完成后对零件进行水溶，去除支撑，形成中空结构后再封闭模型。再将其置于岛津SMX-225工业CT的扫描台上，所述扫描时间为5min，扫描完成获取原始图像，通过神经卷积网络还原图像，可以鉴定产品的真伪。
实施例3
若采用数字光处理工艺打印原料为液态的3D模型：所述中空打印方案通过内部填充液态原料来实现，建议工艺孔开孔尺寸至少为3mm；所述打印机为SprintRay公司的SprintRay Pro打印机；所述打印材料为 Die and Model树脂；所述打印件为正四面体，经ANSYS软件评估后将图案码的内嵌方向定位在正面；所述内嵌厚度为0.1mm；所述内嵌深度为0.5mm；所述内嵌尺寸为4*4mm；将完成内嵌后的3D模型转换成“.stl”格式，在通过RayWare软件将“.stl”格式上传到SprintRay Pro打印机，打印完成后通过工艺孔将内部液体倒出，形成中空结构后再封闭模型。再将其置于岛津SMX-225工业CT的扫描台上，所述扫描时间为5min，扫描完成获取原始图像，通过神经卷积网络还原图像，可以鉴定产品的真伪。
以上内容是结合具体的优选实施方式对本发明所做的进一步详细说明，不能认定本发明的具体实施方式仅限于此，对于本发明所属技术领域的普通技术人员来说，在不脱离本发明的前提下，还可以做出若干简单的推演或替换，都应当视为属于本发明由所提交的权利要求书确定专利保护范围。

Claims (9)

  1. 基于内嵌识别码多工艺适用的3D打印制品防伪方法，其特征在于，包括以下步骤：
    a)确定对3D打印件力学性能影响最小的位置或对外观影响最小的位置为A1；
    b)通过建模软件以内嵌的方式将立体的中空识别码嵌入到A1处；
    c)转换成打印机可识别的格式，输入打印机得到打印件，使打印件内部形成中空结构的识别码；
    d)对打印件进行CT扫描，得到三维重建的模型，对重建的模型进行分层切片，切片平面与识别码内嵌方向垂直，获取带有识别码的切片图像；
    e)设定原始识别码和与原始码相近的错误码为训练集训练卷积神经网络，CT扫描得到的图像为测试集，使用网络处理切片图像，自主提取原始图像特征并自动还原图像。
  2. 根据权利要求1所述的基于内嵌识别码多工艺适用的3D打印制品防伪方法，其特征在于，通过ANSYS有限元分析软件创建模型，定义打印材料属性，设定材料的弹性模量和泊松比，对模型划分网格后设置边界约束条件，对不同部位施加相同载荷后求解，观察打印件的变形程度，寻找对打印件力学性能影响最小的地方定位为A1；对于无力学性能要求的工艺品，避开精细做工的区段而选择对外观影响最小的部位定位为A1。
  3. 根据权利要求1所述的基于内嵌识别码多工艺适用的3D打印制品防伪方法，其特征在于，步骤b所述的建模软件为Autodesk 3ds Max，并根据不同的3D打印工艺选择不同的内嵌厚度、内嵌深度和内嵌尺寸。
  4. 根据权利要求3所述的基于内嵌识别码多工艺适用的3D打印制品防伪方法，其特征在于，所述内嵌厚度指中空的识别码上表面和下表面之间的平行距离，对于弯曲处的内嵌识别码，所述内嵌厚度指中空的识别码上表面和下表面之间的平均距离；所述内嵌深度指中空的识别码下表面距零件上表面之间的平均距离；所述内嵌尺寸指的是识别码的长宽。
  5. 根据权利要求3所述的基于内嵌识别码多工艺适用的3D打印制品 防伪方法，其特征在于，步骤c中，将在Autodesk 3ds Max软件中建立的“.max”模型格式转换成“.stl”格式，再通过打印机配套软件转换成打印机可识别的格式，输入打印机得到打印件。
  6. 根据权利要求1所述的基于内嵌识别码多工艺适用的3D打印制品防伪方法，其特征在于，所述识别码为能通过建模软件转换使其在Z轴方向上有一定厚度的3D的内嵌置入零件的防伪图案。
  7. 根据权利要求1所述的基于内嵌识别码多工艺适用的3D打印制品防伪方法，其特征在于，采用整体内嵌的方式一次性完整嵌入识别码或采用分段内嵌的方式，将识别码分割成多段，再将其中部分码段进行翻转，内嵌到打印件的不同深度，以提高其防伪性能。
  8. 根据权利要求1所述的基于内嵌识别码多工艺适用的3D打印制品防伪方法，其特征在于，通过深度学习算法自主提取原始图像特征，自动还原图像。
  9. 根据权利要求1所述的基于内嵌识别码多工艺适用的3D打印制品防伪方法，其特征在于，对于纸基大型拼接件的3D打印，采用分段榫卯拼接结构，在拼接部位嵌入识别码；或在打印过程中，在每个内嵌层切割完毕后暂时停止设备，手动移除识别码部分后再启动设备进行施胶工艺；对于粘合剂喷射工艺打印的粉基零件的中空结构，用粉末填充中空部位，在后处理过程中通过预留的孔洞去除粉末；对于熔融沉积工艺打印的中空结构，是在内部增加水溶性的支撑架，在后处理过程中溶解去除支撑架。
PCT/CN2022/094331 2021-10-25 2022-05-23 基于内嵌识别码多工艺适用的3d打印制品防伪方法 WO2023071159A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202111242179.8 2021-10-25
CN202111242179.8A CN113954360A (zh) 2021-10-25 2021-10-25 基于内嵌识别码多工艺适用的3d打印制品防伪方法

Publications (1)

Publication Number Publication Date
WO2023071159A1 true WO2023071159A1 (zh) 2023-05-04

Family

ID=79466708

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/094331 WO2023071159A1 (zh) 2021-10-25 2022-05-23 基于内嵌识别码多工艺适用的3d打印制品防伪方法

Country Status (2)

Country Link
CN (1) CN113954360A (zh)
WO (1) WO2023071159A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117644680A (zh) * 2023-11-28 2024-03-05 深圳市深大极光科技股份有限公司 一种可烫印三维全视角显示膜及其制备方法

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113954360A (zh) * 2021-10-25 2022-01-21 华南理工大学 基于内嵌识别码多工艺适用的3d打印制品防伪方法

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2015082678A (ja) * 2013-10-21 2015-04-27 大日本印刷株式会社 ３ｄプリンタ出力用データの著作権保護方法および保護システム
CN107977688A (zh) * 2017-12-31 2018-05-01 西安立东行智能技术有限公司 一种防伪印章印鉴图像人工智能识别软件系统及建立方法
CN108058376A (zh) * 2017-12-12 2018-05-22 运城学院 基于纸基3d打印的内嵌式nfc防伪包装及其制作方法和应用
CN108068479A (zh) * 2017-12-31 2018-05-25 西安立东行智能技术有限公司 一种抗3d打印伪造的匹配人工智能识别的防伪印章系统及印章制作方法
CN108830776A (zh) * 2018-07-31 2018-11-16 浙江财经大学 面向3d打印模型的三维可见实体水印版权防伪标识方法
CN113159015A (zh) * 2021-05-07 2021-07-23 上海趋研信息科技有限公司 一种基于迁移学习的印章识别方法
CN113954360A (zh) * 2021-10-25 2022-01-21 华南理工大学 基于内嵌识别码多工艺适用的3d打印制品防伪方法

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9400910B2 (en) * 2014-02-18 2016-07-26 Adobe Systems Incorporated Method and apparatus for storing and retrieving data embedded into the surface of a 3D printed object
CN107563983B (zh) * 2017-09-28 2020-09-01 上海联影医疗科技有限公司 图像处理方法以及医学成像设备
CN110135454A (zh) * 2019-04-02 2019-08-16 成都真实维度科技有限公司 一种基于3d断层扫描图数据集的深度学习模型训练方法

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2015082678A (ja) * 2013-10-21 2015-04-27 大日本印刷株式会社 ３ｄプリンタ出力用データの著作権保護方法および保護システム
CN108058376A (zh) * 2017-12-12 2018-05-22 运城学院 基于纸基3d打印的内嵌式nfc防伪包装及其制作方法和应用
CN107977688A (zh) * 2017-12-31 2018-05-01 西安立东行智能技术有限公司 一种防伪印章印鉴图像人工智能识别软件系统及建立方法
CN108068479A (zh) * 2017-12-31 2018-05-25 西安立东行智能技术有限公司 一种抗3d打印伪造的匹配人工智能识别的防伪印章系统及印章制作方法
CN108830776A (zh) * 2018-07-31 2018-11-16 浙江财经大学 面向3d打印模型的三维可见实体水印版权防伪标识方法
CN113159015A (zh) * 2021-05-07 2021-07-23 上海趋研信息科技有限公司 一种基于迁移学习的印章识别方法
CN113954360A (zh) * 2021-10-25 2022-01-21 华南理工大学 基于内嵌识别码多工艺适用的3d打印制品防伪方法

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117644680A (zh) * 2023-11-28 2024-03-05 深圳市深大极光科技股份有限公司 一种可烫印三维全视角显示膜及其制备方法

Also Published As

Publication number Publication date
CN113954360A (zh) 2022-01-21

Similar Documents

Publication Publication Date Title
WO2023071159A1 (zh) 基于内嵌识别码多工艺适用的3d打印制品防伪方法
Kurfess et al. Rethinking additive manufacturing and intellectual property protection
US20210170690A1 (en) System and method for embedding security identifiers in additive manufactured parts
CN102609948B (zh) 一种针对复制粘贴篡改的数码照片伪造检测方法
WO2016089838A1 (en) Additive manufactured serialization
CN102473329B (zh) 安全文档尤其是纸币的认证
US11504902B2 (en) Methods and apparatus to identify additively manufactured parts
CN110457996B (zh) 基于vgg-11卷积神经网络的视频运动对象篡改取证方法
Chen et al. Embedded product authentication codes in additive manufactured parts: Imaging and image processing for improved scan ability
AlSawadi et al. Copy-move image forgery detection using local binary pattern and neighborhood clustering
CN104021224A (zh) 基于逐层标签融合深度网络的图像标注方法
JP2017073696A (ja) パターン生成装置、情報埋め込み装置、情報検出装置、方法、媒体、及びプログラム
Gültekin et al. Embedding QR codes on the interior surfaces of FFF fabricated parts
JP5435431B2 (ja) 偽造印影検査方法及び記録媒体
Sun et al. Adaptive watershed segmentation of binary particle image
Delmotte et al. Blind watermarking for 3-d printed objects using surface norm distribution
Usama et al. Embedding information into or onto additively manufactured parts: a review of qr codes, steganography and watermarking methods
CN102279969A (zh) 基于contourlet和商空间的抗打印扫描数字水印方法
CN105894435A (zh) 一种新型防伪标签的处理方法
Chen et al. 3D Print-Scan resilient localized mesh watermarking
KR101080069B1 (ko) 위조인영 판독방법 및 기록매체
Baumann et al. Watermarking for fused deposition modeling by seam placement
CN109685862A (zh) 一种ct切片直接转换成3d打印g代码的方法
Ju et al. An authentication method for copy areas of images
CN111709259A (zh) 一种人工智能防伪查询系统获得大数据的方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22885072

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE