WO2018068519A1 - 一种防伪纹理识别方法 - Google Patents

一种防伪纹理识别方法 Download PDF

Info

Publication number
WO2018068519A1
WO2018068519A1 PCT/CN2017/087653 CN2017087653W WO2018068519A1 WO 2018068519 A1 WO2018068519 A1 WO 2018068519A1 CN 2017087653 W CN2017087653 W CN 2017087653W WO 2018068519 A1 WO2018068519 A1 WO 2018068519A1
Authority
WO
WIPO (PCT)
Prior art keywords
counterfeiting
identifier
overlapping points
identification
texture structure
Prior art date
Application number
PCT/CN2017/087653
Other languages
English (en)
French (fr)
Inventor
季长平
Original Assignee
无锡新光印标识科技有限公司
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 无锡新光印标识科技有限公司 filed Critical 无锡新光印标识科技有限公司
Publication of WO2018068519A1 publication Critical patent/WO2018068519A1/zh

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
    • G06K19/00—Record carriers for use with machines and with at least a part designed to carry digital markings
    • G06K19/06—Record carriers for use with machines and with at least a part designed to carry digital markings characterised by the kind of the digital marking, e.g. shape, nature, code
    • G06K19/06009—Record carriers for use with machines and with at least a part designed to carry digital markings characterised by the kind of the digital marking, e.g. shape, nature, code with optically detectable marking
    • G06K19/06037—Record carriers for use with machines and with at least a part designed to carry digital markings characterised by the kind of the digital marking, e.g. shape, nature, code with optically detectable marking multi-dimensional coding
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/80—Recognising image objects characterised by unique random patterns

Definitions

  • the present invention relates to an anti-counterfeiting method, and more particularly to an anti-counterfeiting texture recognizing method.
  • Texture anti-counterfeiting is an anti-counterfeiting technology in which the marking mark inherent in the packaging material itself is an anti-counterfeit identification mark.
  • the markings in nature are always different. Textures such as fingerprints, wood grain, stone grain, zebra pattern, ice pattern, and cracked ground are random, unique, and impossible to have two identical. Using this principle, anti-counterfeit labels are made from clear-textured packaging materials, and the marking marks of each label material are photographed, numbered, documented, and stored in the anti-counterfeiting database.
  • the diffusion lines of the printed code or the printed image are randomly curved micro-curves, which are random, unique, and non-reproducible. Therefore, the anti-counterfeiting identification is performed by using a diffused pattern (commonly known as a burr) or a concave diffusing pattern (commonly known as a slit) which is convex on the edge of the ink such as a code/text, and the sawtooth side 1 in FIG. 1 has a simple process, and the anti-counterfeiting data is difficult.
  • a diffused pattern commonly known as a burr
  • a concave diffusing pattern commonly known as a slit
  • the technical problem to be solved by the present invention is to provide an anti-counterfeiting texture recognition method, which can not only reduce the requirement for the definition of random texture structure, but also greatly simplify the complexity of subsequent recognition processing, greatly improve the identification precision, and thus effectively prevent being
  • the anti-counterfeiting data after the theft is used for fraud, and the process is simple and easy to implement and promote.
  • the technical solution adopted by the present invention to solve the above technical problem is to provide an anti-counterfeiting texture recognition method, which includes the following steps: a) selecting an anti-counterfeiting information carrier having a random texture structure as a coding mark on a substrate a printed area of the identification/textual identification; b) printing an encoded identification/graphic identifier in the printed area, the encoded identification/text identification at least partially obscuring the random texture structure on the anti-counterfeit information carrier; The overlapping points of the coded logo/texture identifier and the random texture structure are used as anti-counterfeiting feature filing information.
  • the surface of the substrate is provided with fibers, and the fibers are applied to the surface of the substrate through the adhesive layer to form an anti-counterfeit information carrier having a random texture structure.
  • the step c) records the number of stored overlapping points, and the anti-counterfeiting identification first compares the encoded identifier or the graphic identifier, and if the inconsistency directly reports the identification conclusion as a counterfeit product; if the encoded identifier or graphic If the identification is consistent, the number of overlapping points is consistent.
  • step c) selects the number of overlapping points of the partial closed area edge line and the random texture structure for record storage and subsequent comparison.
  • step c) further records location information of the storage overlap point, and the anti-counterfeiting identification ⁇ compares the number of overlapping points and the position information.
  • the random texture structure is an artifact stripe formed by adding irregular shape particles to the production substrate; the step c) selecting a mark area of the two-dimensional code as a partial closed area, and the step c) is selected The number of overlapping points of the marking area edge and the random texture structure and the number of artifact stripes contained in the marking area are recorded and stored and subsequently compared.
  • the step c) selects at least three overlapping points in the relative blank area of the coded identifier/text identifier for recording and storage; the anti-counterfeiting ratio ⁇ , the selected overlapping points are sequentially connected to form a partial closed The region, and compares the number of virtual overlapping points formed by the partial closed region edge and the random texture structure.
  • the random texture structure is an artifact stripe formed by adding irregular shape particles to the production substrate; the step c) is also comparing the artifact strips contained in the partial closed region in the anti-counterfeiting ratio Whether the number is consistent.
  • the step c) pre-stores the coded identifier/text identification photo, the anti-counterfeit identification ⁇ , obtains the physical scene photo, first compares the coded identifier or the graphic identifier, and if the inconsistency directly feedbacks the identification conclusion as a counterfeit Product; If the coded identifier or the graphic identification match is consistent, continue to compare whether the number of overlapping points and position information in the physical scene photo match the pre-stored information.
  • the random texture structure is an artifact stripe formed by adding irregular shape particles to the production substrate; if the number of virtual overlap points is consistent, and the number of virtual overlap points is less than a preset threshold, then Continue to compare whether the number of artifact fringes contained in the partially enclosed area is consistent, if the inconsistent feedback identification conclusion is a counterfeit product.
  • the anti-counterfeiting texture recognition method provided by the present invention can obtain the anti-counterfeiting feature filing information by acquiring the overlapping points of the random texture structure and the coding identifier/texture identifier, which can not only reduce the pairing The requirement of random texture structure definition, and greatly simplify the complexity of subsequent recognition processing, and improve the identification accuracy. Since the texture structure is randomly generated, the overlapping points of the coding or the image cover are also random, thereby greatly increasing the difficulty of forgery, effectively preventing theft-proof anti-counterfeiting data from being used for fraud, and the process is simple and easy to implement and popularize.
  • FIG. 1 is a schematic diagram of a conventional two-dimensional code structure using a diffusion pattern
  • FIG. 2 is a schematic structural diagram of a two-dimensional code of the present invention.
  • FIG. 3 is a schematic diagram of an anti-counterfeiting identification process of the present invention.
  • FIG. 4 is a schematic diagram showing an enlarged structure of a hollow space in the middle of the two-dimensional code in FIG. 2;
  • FIG. 5 is a schematic diagram of a partially enclosed region formed by overlapping points at the middle of the two-dimensional code in FIG.
  • FIG. 2 is a schematic diagram of a two-dimensional code structure of the present invention
  • FIG. 3 is a schematic diagram of an anti-counterfeiting identification process of the present invention.
  • the substrate 2 is printed with a two-dimensional code, and the random texture structure on the substrate 2 is fiber. 3, the two-dimensional code is covered on the fiber 3, and the two-dimensional code and the fiber 3 are partially covered.
  • the anti-counterfeiting texture recognition method provided by the present invention comprises the following steps:
  • Step S1 selecting an anti-counterfeit information carrier having a random texture structure as a printing area on the substrate with a coded logo or a graphic logo;
  • Step S2 printing a coded identifier or a graphic identifier in the printing area, where the encoded identifier or the graphic identifier at least partially covers the random texture structure on the anti-counterfeiting information carrier;
  • Step S3 Obtain an overlap point of the coded identifier and the random texture structure, or an overlap point of the graphic identifier and the random texture structure as the anti-counterfeiting feature filing information.
  • the invention can directly adopt the inherent texture structure of natural materials as an anti-counterfeiting information carrier, such as wood blocks, stones, plant stems and leaves, shells, etc. with clear texture, natural spots, or specially added in the production of substrates.
  • a plastic, glass, ceramic or composite that produces clear, random structural textures of pigments, sundries, irregularly shaped particles or bubbles, resulting in clear artifact stripes. Since the present invention does not require direct comparison of the natural texture structure, generally natural texture structures or artifact stripes can substantially satisfy the requirements of the present invention.
  • the surface of the substrate is provided with fibers, and the fibers are applied to the surface of the substrate through the adhesive layer to form an anti-counterfeit information carrier having a random texture structure, and the fibers are used as the image strips.
  • the anti-counterfeiting texture recognition method records and stores the number of overlapping points in advance, and the anti-counterfeiting identification first compares the coded identifier or the graphic identifier, and if the inconsistency directly reports the identification conclusion as a counterfeit product; If the graphic identifiers match, the number of overlapping points is consistent, which greatly increases the difficulty of forgery.
  • the coded identifier used in the present invention may be a serial number, a one-dimensional barcode, a two-dimensional code, a production date, a product batch number, an anti-counterfeit digital, a promotional integral code, a redemption code, an anti-counterfeiting code, or a product source code.
  • the number of overlapping points in the overall printing area may be very large.
  • the step S3 may select the number and position information of the overlapping points in the partial closed area for recording and storage; for example, selecting the two-dimensional code marking area as the local selected area.
  • the present invention can also record the position information of the storage overlap point, and the anti-counterfeiting identification ⁇ compares the number of overlapping points and the position information.
  • the present invention can select a marked area of a two-dimensional code as a partial closed area, and select a target
  • the number of overlapping points of the sideline and the random texture structure and the number of artifact stripes contained in the marking area are recorded and stored and subsequently compared.
  • the position of the marking area of the two-dimensional code is relatively fixed, which affects the forgery difficulty to a certain extent; while the other relative blank areas of the two-dimensional code are generally not closed areas, including the hollowed out area, and other code identification/graphic identification , also have the above problems.
  • the present invention can select at least three overlapping points in the relatively blank area of the coded logo/graphic identifier for record storage; the anti-counterfeiting ratio ⁇ , the selected overlapping points are sequentially connected, and the artificial structure forms a
  • the partial closed area is then compared with whether the number of virtual overlapping points formed by the partial closed area edge and the random texture structure is consistent; and whether or not to continue to compare the number of artifact stripes contained in the partial closed area can be determined as needed.
  • the anti-counterfeiting texture recognition method provided by the invention, the coded logo or the graphic identifier together with the anti-counterfeiting features thereon can be conveniently and uniformly stored, respectively identified, the process is simple, and the implementation is easy to implement. As the printing process and computer image recognition accuracy continue to improve, the anti-counterfeiting texture recognition method provided by the present invention can finally perform the identification alignment in the following order:
  • n is an integer, n>3, and n-sided, n-1-sided, n-2-sided, respectively, can be constructed in sequence, up to the quadrilateral and the triangle; and each polygon is compared one by one.
  • the number of virtual overlap points formed by the random texture structure and whether the number of random textures contained in the polygon is the same.
  • the present invention can further compare whether the positions of the corresponding virtual overlapping points are the same.
  • the present invention can flexibly combine and use some of the conditions for comprehensive judgment.
  • the following uses an existing smartphone as an example to give a specific embodiment.
  • the invention pre-stores the coded logo or the graphic identification photo, the anti-counterfeit identification ⁇ , obtains the physical scene photo, first compares the coding identifier or the graphic identifier, and if the inconsistency directly feedbacks the identification conclusion as a counterfeit product; if the coding identifier or the graphic identifier If the comparison is consistent, the number and position information of the overlapping points in the physical scene photo are consistent with the pre-stored information. If the number of overlapping points and the position information are completely consistent, it can be directly judged as true. Make further follow-up judgments.
  • the number of overlapping points is less than the pre-stored information; or the number of overlapping points is less than a preset threshold. Thereafter, if the positional information of the partial overlap point coincides with the pre-stored information, in order to prevent misjudgment, further subsequent judgment can be made. Then, the corresponding overlapping points are sequentially connected to form a partial closed area on the pre-stored photo and the physical scene photo, and the number of virtual overlapping points formed by the partial closed area edge line and the random texture structure is consistent, if consistent, Can be judged as true, thereby further improving the identification accuracy; if the inconsistent feedback identification conclusion is a counterfeit product.
  • the artifact strips contained in the partially enclosed area may also be continuously compared. Whether the number is consistent, if the inconsistent feedback identification conclusion is a counterfeit product.
  • the two-dimensional code and the fiber artifact form four overlapping points, namely, an overlapping point A, an overlapping point B, an overlapping point C, and an overlapping point D, and four intersections.
  • the quadrilateral formed by the overlap forms three virtual overlapping points with the fiber artifacts, namely a virtual overlapping point A', a virtual overlapping point B' and a virtual overlapping point C', and the number of artifact stripes contained in the quadrilateral is One.
  • the anti-counterfeiting texture recognition method provided by the present invention can obtain the anti-counterfeiting feature filing information by acquiring the overlapping points of the random texture structure and the coding identifier/texture identifier, thereby not only reducing the requirement for the definition of the random texture structure, but also greatly simplifying the subsequent Identify the complexity of the processing and greatly improve the identification accuracy. Since the texture structure is randomly generated, the overlapping points of the code or the image cover are also random, which increases the difficulty of forgery.
  • a plurality of overlapping points are sequentially connected to form a closed polygon, and then the number of virtual overlapping points formed by the closed polygon and the texture structure is searched for anti-counterfeiting identification, which greatly reduces the difficulty of identification, and has the advantages of fast recognition speed and high precision.
  • the invention not only solves the problem that the non-closed area is difficult to identify, but also adopts a combination of a real overlapping point and a virtual overlapping point, and has the characteristics of being difficult to forge.
  • the partially enclosed area is not only non-existent, its position is randomly generated according to different codes, and its shape is determined according to the number of real overlapping points, and is also random.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • Credit Cards Or The Like (AREA)
  • Inspection Of Paper Currency And Valuable Securities (AREA)

Abstract

本发明公开了一种防伪纹理识别方法，包括如下步骤：a）选用具有随机纹理结构的防伪信息载体作为承印物上编码标识/图文标识的印刷区；b）在所述印刷区印刷形成编码标识/图文标识，所述编码标识/图文标识至少部分遮盖住所述防伪信息载体上的随机纹理结构；c）获取编码标识/图文标识和随机纹理结构的交叠点作为防伪特征备案信息。本发明不但能够降低对随机纹理结构清晰度的要求，而且大大简化后续识别处理的复杂度，提高辨识精度。由于纹理结构是随机生成的，编码或图文遮盖的交叠点也是随机的，从而极大地增加伪造难度，有效防止被盗后的防伪数据用于造假，且工艺简单，易于实施推广。

Description

一种防伪纹理识别方法 技术领域
[0001] 本发明涉及一种防伪方法， 尤其涉及一种防伪纹理识别方法。
背景技术
[0002] 纹理防伪是以包装材料本身固有的斑纹记号为防伪识别标记的一种防伪技术。
自然界的斑纹总是千差万别的， 如指纹、 木纹、 石纹、 斑马纹、 冰纹、 干裂的 地面等纹理都是随机的、 唯一的、 不可能有两个完全一样的。 利用这一原理， 选用纹理清晰的包装材料制成防伪标签， 对每一枚标签材料的斑纹记号进行拍 摄、 编号、 建档、 存入防伪数据库中。
[0003] 此外， 印刷编码或印刷图文的扩散纹线是一条条随机弯曲的微曲线， 具有随机 性、 唯一性、 不可复制性。 因此利用编码 /图文等油墨边沿上凸起来的扩散纹 （ 俗称毛刺） 、 或凹下去的扩散纹 （俗称豁口） 进行防伪识别， 如图 1中的锯齿边 1， 具有工艺简单， 防伪数据难于造假的优点。
技术问题
[0004] 但是不管是天然的纹理结构还是印刷形成的随机锯齿纹， 实际进行防伪应用吋 ， 仍存在如下缺点： 1、 纹理特征可能不够清晰， 缺乏个性化； 2、 编码 /图文扩 散纹本身容易产生变形， 褪色， 最终大大增加了识别难度； 3、 对纹理相似性的 判断处理较为复杂， 难于保证达到较高的辨识度。
问题的解决方案
技术解决方案
[0005] 本发明所要解决的技术问题是提供一种防伪纹理识别方法， 不但能够降低对随 机纹理结构清晰度的要求， 而且大大简化后续识别处理的复杂度， 大大提高辨 识精度， 从而有效防止被盗后的防伪数据用于造假， 且工艺简单， 易于实施推 广。
[0006] 本发明为解决上述技术问题而采用的技术方案是提供一种防伪纹理识别方法， 包括如下步骤： a) 选用具有随机纹理结构的防伪信息载体作为承印物上编码标 识 /图文标识的印刷区； b) 在所述印刷区印刷形成编码标识 /图文标识， 所述编 码标识 /图文标识至少部分遮盖住所述防伪信息载体上的随机纹理结构； C) 获取 编码标识 /图文标识和随机纹理结构的交叠点作为防伪特征备案信息。
[0007] 进一步地， 所述步骤 a) 中承印物的表面设有纤维， 所述纤维通过粘合剂层贴 敷在承印物的表面形成具有随机纹理结构的防伪信息载体。
[0008] 进一步地， 所述步骤 c) 记录存储交叠点的数目， 防伪识别吋首先对编码标识 或图文标识进行比对， 如果不一致直接反馈鉴定结论为假冒产品； 如果编码标 识或图文标识比对一致， 则继续比对交叠点的数目是否一致。
[0009] 进一步地， 所述步骤 c) 选取局部封闭区域边线和随机纹理结构的交叠点的数 目进行记录存储及后续比对。
[0010] 进一步地， 所述步骤 c) 还记录存储交叠点的位置信息， 防伪识别吋同吋比对 交叠点的数目和位置信息是否一致。
[0011] 进一步地， 所述随机纹理结构为生产承印物吋添加不规则形状颗粒形成的伪像 条纹； 所述步骤 c) 选取二维码的标记区作为局部封闭区域， 所述步骤 c) 选取标 记区边线和随机纹理结构的交叠点数目以及标记区内包含的伪像条纹数目进行 记录存储及后续比对。
[0012] 进一步地， 所述步骤 c) 在编码标识 /图文标识的相对空白区域内至少选取三个 交叠点进行记录存储； 防伪比对吋， 将选取的交叠点依次相连形成局部封闭区 域， 并比对局部封闭区域边线和随机纹理结构的所形成的虚拟交叠点的数目是 否一致。
[0013] 进一步地， 所述随机纹理结构为生产承印物吋添加不规则形状颗粒形成的伪像 条纹； 所述步骤 c) 在防伪比对吋， 还比对局部封闭区域内包含的伪像条纹数目 是否一致。
[0014] 进一步地， 所述步骤 c) 预先存储编码标识 /图文标识照片， 防伪识别吋， 获取 实物现场照片， 首先对编码标识或图文标识进行比对， 如果不一致直接反馈鉴 定结论为假冒产品； 如果编码标识或图文标识比对一致， 则继续比对实物现场 照片中的交叠点的数目和位置信息是否与预存信息相符。
[0015] 进一步地， 如果实物现场照片中的部分交叠点的位置信息与预存信息相符， 则 分别在预存照片和实物现场照片上将相应的交叠点依次相连形成局部封闭区域 ， 并比对局部封闭区域边线和随机纹理结构所形成的虚拟交叠点的数目是否一 致， 如果不一致反馈鉴定结论为假冒产品。
[0016] 进一步地， 所述随机纹理结构为生产承印物吋添加不规则形状颗粒形成的伪像 条纹； 如果虚拟交叠点的数目一致， 且虚拟交叠点的数目小于预设阀值， 则继 续比对局部封闭区域内包含的伪像条纹数目是否一致， 如果不一致反馈鉴定结 论为假冒产品。
发明的有益效果
有益效果
[0017] 本发明对比现有技术有如下的有益效果： 本发明提供的防伪纹理识别方法， 通 过获取随机纹理结构和编码标识 /图文标识的交叠点作为防伪特征备案信息， 不 但能够降低对随机纹理结构清晰度的要求， 而且大大简化后续识别处理的复杂 度， 提高辨识精度。 由于纹理结构是随机生成的， 编码或图文遮盖的交叠点也 是随机的， 从而极大地增加伪造难度， 有效防止被盗后的防伪数据用于造假， 且工艺简单， 易于实施推广。
对附图的简要说明
附图说明
[0018] 图 1为现有的采用扩散纹的二维码结构示意图；
[0019] 图 2为本发明的二维码结构示意图；
[0020] 图 3为本发明的防伪识别过程示意图；
[0021] 图 4为图 2中二维码中间镂空处放大结构示意图；
[0022] 图 5为图 2中二维码中间镂空处交叠点形成的局部封闭区域示意图。
具体实施方式
[0023] 下面结合附图和实施例对本发明作进一步的描述。
[0024] 图 2为本发明的二维码结构示意图； 图 3为本发明的防伪识别过程示意图。
[0025] 请参见图 2和图 3， 承印物 2上印刷有二维码， 承印物 2上的随机纹理结构为纤维 3， 二维码覆盖在纤维 3上， 二维码和纤维 3部分遮盖， 本发明提供的防伪纹理识 别方法， 包括如下步骤：
[0026] 步骤 S1 : 选用具有随机纹理结构的防伪信息载体作为承印物上编码标识或图文 标识的印刷区；
[0027] 步骤 S2: 在所述印刷区印刷形成编码标识或图文标识， 所述编码标识或图文标 识至少部分遮盖住所述防伪信息载体上的随机纹理结构；
[0028] 步骤 S3: 获取编码标识和随机纹理结构的交叠点， 或者图文标识和随机纹理结 构的交叠点作为防伪特征备案信息。
[0029] 本发明可以直接采用天然材质的固有纹理结构作为防伪信息载体， 如结构纹理 清晰、 天然斑点明显的木块、 石块、 植物茎叶、 贝壳等； 或者在生产制造承印 物吋特意添加能够产生清晰的随机结构纹理的颜料、 杂物、 不规则形状颗粒或 气泡的塑料、 玻璃、 陶瓷或复合材料， 从而形成清晰的伪像条纹。 由于本发明 不需直接比对天然纹理结构， 因此一般的天然纹理结构或伪像条纹基本都能满 足本发明的要求。 优选地， 所述步骤 S1中承印物的表面设有纤维， 所述纤维通 过粘合剂层贴敷在承印物的表面形成具有随机纹理结构的防伪信息载体， 采用 纤维作为伪像条纹。
[0030] 本发明提供的防伪纹理识别方法， 预先记录存储交叠点的数目， 防伪识别吋首 先对编码标识或图文标识进行比对， 如果不一致直接反馈鉴定结论为假冒产品 ； 如果编码标识或图文标识比对一致， 则继续比对交叠点的数目是否一致， 从 而大大增加伪造难度。
[0031] 本发明使用的编码标识可以为序列号、 一维条码、 二维码、 生产日期、 产品批 号、 防伪数码、 促销积分码、 兑奖码、 防窜货编码或产品溯源码等。 对于二维 码， 整体印刷区内的交叠点数目可能非常多。 为了便于后续的存储处理及比对 ， 所述步骤 S3可以选取局部封闭区域内交叠点的数目和位置信息进行记录存储 ； 比如选取二维码标记区作为局部选定区域。
[0032] 为了提高辨识精度， 本发明还可以记录存储交叠点的位置信息， 防伪识别吋同 吋比对交叠点的数目和位置信息是否一致。
[0033] 对于常用的二维码， 本发明可选取二维码的标记区作为局部封闭区域， 选取标 记区边线和随机纹理结构的交叠点数目以及标记区内包含的伪像条纹数目进行 记录存储及后续比对。 二维码的标记区的位置相对固定， 在一定程度上影响了 伪造难度； 而二维码的其它相对空白区域一般都不是封闭区域， 包括中间的镂 空区域， 编对其他码标识 /图文标识， 也均存在上述问题。 为了解决这一难题， 本发明可以在编码标识 /图文标识的相对空白区域内至少选取三个交叠点进行记 录存储； 防伪比对吋， 将选取的交叠点依次相连， 人为构造形成一个局部封闭 区域， 然后就可比对局部封闭区域边线和随机纹理结构的所形成的虚拟交叠点 的数目是否一致； 并可根据需要决定是否继续比对局部封闭区域内包含的伪像 条纹数目。
[0034] 本发明提供的防伪纹理识别方法， 编码标识或图文标识连同其上的防伪特征可 以方便地统一存储， 分别辨识， 工艺简单， 易于实施推广。 随着印刷工艺和计 算机图像识别精度的不断提高， 本发明提供的防伪纹理识别方法， 最终可以严 格按照如下顺序进行识别比对：
[0035] 1) 获取现场实物照片和预存的编码标识 /图文标识照片； 2) 首先比对编码标 识 /图文标识是否相同； 3) 接着比较编码标识 /图文标识与随机纹理结构的交叠 点的数目是否相同， 每一交叠点的位置是否相对应； 4) 再选取相应交叠点构造 多边形， 并比较虚拟交叠点数目以及多边形包含的随机纹理数目是否相同。 对 于具有 n个交叠点的情况， n为整数， n>3， 可以依次分别构造 n边形、 n-1边形、 n -2边形， 直至四边形和三角形； 并逐一比较每个多边形和随机纹理结构所形成的 虚拟交叠点的数目以及多边形包含的随机纹理数目是否相同。 此外， 本发明还 可进一步比较相应虚拟交叠点的位置是否相同。
[0036] 实际应用中， 在满足防伪要求的情况， 为了提高辨识的准确性， 减低辨识难度 ， 防止误判， 本发明可以灵活组合使用其中某些条件进行综合判断。 下面以现 有智能手机使用为例， 给出一个具体实施例。 本发明预先存储编码标识或图文 标识照片， 防伪识别吋， 获取实物现场照片， 首先对编码标识或图文标识进行 比对， 如果不一致直接反馈鉴定结论为假冒产品； 如果编码标识或图文标识比 对一致， 则继续比对实物现场照片中的交叠点的数目和位置信息是否与预存信 息相符， 如果交叠点的数目和位置信息均完全相符， 可以直接判断为真， 也可 作进一步的后续判断。
[0037] 如果实物现场照片中的部分交叠点不清晰， 导致交叠点数目少于预存信息； 或 者交叠点的数目少于预设阀值。 此吋， 如果所述部分交叠点的位置信息与预存 信息相符， 为了防止误判， 可作进一步的后续判断。 接着分别在预存照片和实 物现场照片上将相应的交叠点依次相连形成局部封闭区域， 并比对局部封闭区 域边线和随机纹理结构所形成的虚拟交叠点的数目是否一致， 如果一致， 仍可 判定为真， 从而进一步提高辨识精度； 如果不一致反馈鉴定结论为假冒产品。
[0038] 对于采用了伪像条纹的承印物； 如果虚拟交叠点的数目一致， 但是虚拟交叠点 的数目小于预设阀值， 则还可继续比对局部封闭区域内包含的伪像条纹数目是 否一致， 如果不一致反馈鉴定结论为假冒产品。 如图 4和图 5所示， 二维码和纤 维伪像形成的交叠点有四个， 分别为交叠点 A、 交叠点 B、 交叠点 C和交叠点 D， 四个交叠点构成的四边形又与纤维伪像形成三个虚拟交叠点， 分别为虚拟交叠 点 A'、 虚拟交叠点 B'和虚拟交叠点 C'， 四边形内包含的伪像条纹数目为一个。
[0039] 本发明提供的防伪纹理识别方法， 通过获取随机纹理结构和编码标识 /图文标 识的交叠点作为防伪特征备案信息， 不但能够降低对随机纹理结构清晰度的要 求， 而且大大简化后续识别处理的复杂度， 大大提高辨识精度。 由于纹理结构 是随机生成的， 编码或图文遮盖的交叠点也是随机的， 从而增加伪造难度。 本 实施例将多个交叠点依次相连构成封闭多边形， 然后寻找封闭多边形与纹理结 构所形成的虚拟交叠点的数目进行防伪识别， 大大降低辨识难度， 具有辨识速 度快、 精度高的优点。 本发明不但解决了非封闭区域造成的难于辨识的问题， 而且采用真实交叠点和虚拟交叠点相结合的处理方式， 具有难于伪造的特点。 此外， 局部封闭区域不但是虚拟不存在的， 其位置也是根据不同编码随机产生 的， 而且其形状又是根据真实交叠点的数目确定， 同样具有随机性。 上述多重 随机因素的综合叠加， 加上虚实结合的处理方式， 最后再组合虚拟交叠点数目 和局部封闭区域内包含的伪像条纹数目进行判断； 理论上具有无限多的可能组 合， 从而极大地增加伪造难度， 有效防止被盗后的防伪数据用于造假。
[0040] 虽然本发明已以较佳实施例揭示如上， 然其并非用以限定本发明， 任何本领域 技术人员， 在不脱离本发明的精神和范围内， 当可作些许的修改和完善， 因此 本发明的保护范围当以权利要求书所界定的为准。

Claims

权利要求书
一种防伪纹理识别方法， 其特征在于， 包括如下步骤：
a) 选用具有随机纹理结构的防伪信息载体作为承印物上编码标识 /图 文标识的印刷区；
b) 在所述印刷区印刷形成编码标识 /图文标识， 所述编码标识 /图文标 识至少部分遮盖住所述防伪信息载体上的随机纹理结构；
c) 获取编码标识 /图文标识和随机纹理结构的交叠点作为防伪特征备 案信息。
如权利要求 1所述的微纹防伪方法， 其特征在于， 所述步骤 a) 中承印 物的表面设有纤维， 所述纤维通过粘合剂层贴敷在承印物的表面形成 具有随机纹理结构的防伪信息载体。
如权利要求 1所述的防伪纹理识别方法， 其特征在于， 所述步骤 c) 记 录存储交叠点的数目， 防伪识别吋首先对编码标识或图文标识进行比 对， 如果不一致直接反馈鉴定结论为假冒产品； 如果编码标识或图文 标识比对一致， 则继续比对交叠点的数目是否一致。
如权利要求 3所述的防伪纹理识别方法， 其特征在于， 所述步骤 c) 选 取局部封闭区域边线和随机纹理结构的交叠点的数目进行记录存储及 后续比对。
如权利要求 3或 4所述的防伪纹理识别方法， 其特征在于， 所述步骤 c ) 还记录存储交叠点的位置信息， 防伪识别吋同吋比对交叠点的数目 和位置信息是否一致。
如权利要求 4所述的防伪纹理识别方法， 其特征在于， 所述随机纹理 结构为生产承印物吋添加不规则形状颗粒形成的伪像条纹； 所述步骤 c) 选取二维码的标记区作为局部封闭区域， 所述步骤 c) 选取标记区 边线和随机纹理结构的交叠点数目以及标记区内包含的伪像条纹数目 进行记录存储及后续比对。
如权利要求 1所述的防伪纹理识别方法， 其特征在于， 所述步骤 c) 在 编码标识 /图文标识的相对空白区域内至少选取三个交叠点进行记录 存储； 防伪比对吋， 将选取的交叠点依次相连形成局部封闭区域， 并 比对局部封闭区域边线和随机纹理结构的所形成的虚拟交叠点的数目 是否一致。
[权利要求 8] 如权利要求 7所述的防伪纹理识别方法， 其特征在于， 所述随机纹理 结构为生产承印物吋添加不规则形状颗粒形成的伪像条纹； 所述步骤 c) 在防伪比对吋， 还比对局部封闭区域内包含的伪像条纹数目是否
[权利要求 9] 如权利要求 1所述的防伪纹理识别方法， 其特征在于， 所述步骤 c) 预 先存储编码标识 /图文标识照片， 防伪识别吋， 获取实物现场照片， 首先对编码标识或图文标识进行比对， 如果不一致直接反馈鉴定结论 为假冒产品； 如果编码标识或图文标识比对一致， 则继续比对实物现 场照片中的交叠点的数目和位置信息是否与预存信息相符。
[权利要求 10] 如权利要求 9所述的防伪纹理识别方法， 其特征在于， 如果实物现场 照片中的部分交叠点的位置信息与预存信息相符， 则分别在预存照片 和实物现场照片上将相应的交叠点依次相连形成局部封闭区域， 并比 对局部封闭区域边线和随机纹理结构所形成的虚拟交叠点的数目是否 一致， 如果不一致反馈鉴定结论为假冒产品。
[权利要求 11] 如权利要求 10所述的防伪纹理识别方法， 其特征在于， 所述随机纹理 结构为生产承印物吋添加不规则形状颗粒形成的伪像条纹； 如果虚拟 交叠点的数目一致， 且虚拟交叠点的数目小于预设阀值， 则继续比对 局部封闭区域内包含的伪像条纹数目是否一致， 如果不一致反馈鉴定 结论为假冒产品。
PCT/CN2017/087653 2016-10-12 2017-06-09 一种防伪纹理识别方法 WO2018068519A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201610890975.5 2016-10-12
CN201610890975.5A CN106446866B (zh) 2016-10-12 2016-10-12 一种防伪纹理识别方法

Publications (1)

Publication Number Publication Date
WO2018068519A1 true WO2018068519A1 (zh) 2018-04-19

Family

ID=58173559

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/087653 WO2018068519A1 (zh) 2016-10-12 2017-06-09 一种防伪纹理识别方法

Country Status (2)

Country Link
CN (1) CN106446866B (zh)
WO (1) WO2018068519A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108985799A (zh) * 2018-07-05 2018-12-11 任伟峰 一种基于随机痕迹特征的产品保真方法

Families Citing this family (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106446866B (zh) * 2016-10-12 2018-07-13 无锡新光印标识科技有限公司 一种防伪纹理识别方法
CN108898205A (zh) * 2017-05-09 2018-11-27 罗伯特·博世有限公司 二进制图形编码的创建、认证方法及系统
CN108875870B (zh) * 2017-05-09 2021-09-28 罗伯特·博世有限公司 二进制图形编码的创建、认证方法及系统
CN109754261B (zh) * 2017-11-01 2024-03-05 杭州沃朴物联科技有限公司 一种基于nfc和多纹理图像识别的防伪系统
CN207993025U (zh) * 2018-01-02 2018-10-19 海南亚元防伪技术研究所（普通合伙） 手检纹理防伪印刷物及其专用承印材料
CN109291674B (zh) * 2018-10-10 2020-09-01 福州大学 一种基于喷墨打印的不可复制防伪标签及其制备方法
CN109840781A (zh) * 2019-02-18 2019-06-04 杭州安芯科技有限公司 基于自然纹路的互联网防伪中定位纹路的方法和系统
CN109919640A (zh) * 2019-04-12 2019-06-21 海南亚元防伪技术研究所（普通合伙） 三维锯齿防伪产品
CN112801254A (zh) * 2021-02-07 2021-05-14 拍拍看(海南)人工智能有限公司 高精准锯齿防伪方法
CN115221998A (zh) * 2021-04-20 2022-10-21 中钞特种防伪科技有限公司 一种编码元件和防伪产品
CN114663118B (zh) * 2022-05-23 2022-09-27 武汉朗修科技有限公司 基于镭射随机组合图像的防伪方法

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN2715245Y (zh) * 2003-09-22 2005-08-03 兆日科技（深圳）有限公司 一种纹理密码标签
CN101192356A (zh) * 2006-12-02 2008-06-04 龚镇章 另类随机纹理防伪方法
CN201425813Y (zh) * 2008-12-17 2010-03-17 郑阿奇 条形码埋藏计算机数据结构纹理产品
CN101789199A (zh) * 2009-01-24 2010-07-28 郑阿奇 手机识别埋有信息印刷品随机纹理防伪产品及方法
CN101944305A (zh) * 2009-07-10 2011-01-12 徐克林 一种印章防伪方法
US20120183180A1 (en) * 2009-09-28 2012-07-19 Shanghai Kos Security Paper Technology Co., Ltd. Anti-counterfeit method for random texture and recognizer thereof
CN106446866A (zh) * 2016-10-12 2017-02-22 无锡新光印标识科技有限公司 一种防伪纹理识别方法

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1349186A (zh) * 2001-11-21 2002-05-15 孙显林 一种随机纹理防伪的方法
JP4265180B2 (ja) * 2002-09-09 2009-05-20 富士ゼロックス株式会社 紙識別照合装置
CN103049720A (zh) * 2009-09-27 2013-04-17 天津黑马行云物联网科技有限公司 一种具有编码结构纹理的防伪识别方法
CN103700000A (zh) * 2013-11-28 2014-04-02 海南亚元防伪技术研究所 免喷码纤维防伪方法
CN103646333A (zh) * 2013-12-25 2014-03-19 北京慧眼智行科技有限公司 一种防伪检测方法、设备和系统

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN2715245Y (zh) * 2003-09-22 2005-08-03 兆日科技（深圳）有限公司 一种纹理密码标签
CN101192356A (zh) * 2006-12-02 2008-06-04 龚镇章 另类随机纹理防伪方法
CN201425813Y (zh) * 2008-12-17 2010-03-17 郑阿奇 条形码埋藏计算机数据结构纹理产品
CN101789199A (zh) * 2009-01-24 2010-07-28 郑阿奇 手机识别埋有信息印刷品随机纹理防伪产品及方法
CN101944305A (zh) * 2009-07-10 2011-01-12 徐克林 一种印章防伪方法
US20120183180A1 (en) * 2009-09-28 2012-07-19 Shanghai Kos Security Paper Technology Co., Ltd. Anti-counterfeit method for random texture and recognizer thereof
CN106446866A (zh) * 2016-10-12 2017-02-22 无锡新光印标识科技有限公司 一种防伪纹理识别方法

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108985799A (zh) * 2018-07-05 2018-12-11 任伟峰 一种基于随机痕迹特征的产品保真方法

Also Published As

Publication number Publication date
CN106446866B (zh) 2018-07-13
CN106446866A (zh) 2017-02-22

Similar Documents

Publication Publication Date Title
WO2018068519A1 (zh) 一种防伪纹理识别方法
WO2018068520A1 (zh) 一种防伪标记识别方法
CN207930162U (zh) 包括识别标记的物品和用于确认物品的真实性的确认系统
JP4979769B2 (ja) 隠れ画像の識別システム及び識別手段
WO2011035738A1 (zh) 一种随机纹理的防伪方法及其识别器
CN104372715A (zh) 凸显油墨发泡特征或油墨皱纹特征之纹理防伪标识物
US20210150296A1 (en) Authentication medium, authentication medium manufacturing method, authentication medium reading method, and authentication medium verification method
CN108647980B (zh) 一种微细扩散印迹的防伪识别方法
CN211455117U (zh) 全息烫印特征识别防伪标识
WO2023005996A1 (zh) 一种可定制的防伪元件和防伪产品
CN110110830A (zh) 植入瓷砖装饰图案中的防伪图像码及其防伪方法
JP2017521277A (ja) データキャリア
TWI772420B (zh) 用於鑑別物件，特別是鑑別防偽元件的方法及防偽元件
CN204066185U (zh) 油墨发泡特征或油墨皱纹特征之纹理防伪标识物
CN109809033A (zh) 一种防伪胶帽及其制作和使用方法
US10369826B2 (en) Security document and method of manufacturing
US20070009721A1 (en) Identification media having a security image
CN111985592A (zh) 可动纹理防伪方法及标识物
CN108472982A (zh) 含有生物粒子图案的安全标签
CN101115536B (zh) 信息的保护
CN211506949U (zh) 一种在背景图中嵌入识别符的商标防伪标签
CN210149912U (zh) 一种防伪胶帽
JP2017156947A (ja) マトリクス型二次元コードおよびその製造方法
CN111292605A (zh) 全息烫印特征识别防伪标识及其制备方法
US6772951B1 (en) Method of forming an image, and to a product having an image formed thereon

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: 17861113

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 17861113

Country of ref document: EP

Kind code of ref document: A1