WO2016206173A1 - 一种基于混沌图形标签的防伪系统及方法 - Google Patents

一种基于混沌图形标签的防伪系统及方法 Download PDF

Info

Publication number
WO2016206173A1
WO2016206173A1 PCT/CN2015/086300 CN2015086300W WO2016206173A1 WO 2016206173 A1 WO2016206173 A1 WO 2016206173A1 CN 2015086300 W CN2015086300 W CN 2015086300W WO 2016206173 A1 WO2016206173 A1 WO 2016206173A1
Authority
WO
WIPO (PCT)
Prior art keywords
chaotic
information
graphic
tag
label
Prior art date
Application number
PCT/CN2015/086300
Other languages
English (en)
French (fr)
Inventor
马中发
Original Assignee
杭州沃朴物联科技有限公司
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 杭州沃朴物联科技有限公司 filed Critical 杭州沃朴物联科技有限公司
Publication of WO2016206173A1 publication Critical patent/WO2016206173A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce

Definitions

  • the invention belongs to the technical field of anti-counterfeiting, and in particular relates to an anti-counterfeiting system and method based on chaotic graphic labels.
  • Anti-counterfeiting technology is a cross-technology involving many disciplines. It is often difficult to use anti-counterfeit labels. To a certain extent, the products to be protected are closely related to the institutions or manufacturers indicated on the labels. With the rapid development of anti-counterfeiting technology and market economy, anti-counterfeiting technology has not only been widely used in passports, certificates, currency, tickets and securities produced by banks, customs, taxation, finance, public security, government departments, etc., but also widely used in tobacco. Anti-counterfeiting of pesticides, fertilizers, auto parts and other commodities in wine, food, medicine, health care products, cosmetics, clothing, optical disc products and production materials.
  • Anti-counterfeiting technology plays an important role in controlling and combating counterfeiting and counterfeiting activities, maintaining a good market competition environment and protecting intellectual property rights. China's counterfeiting and counterfeiting are rampant, and merchants are paying more and more attention to the protection of intellectual property rights. Therefore, the demand for new anti-counterfeiting technology is very urgent.
  • the core of the anti-counterfeiting system is the security label.
  • anti-counterfeit labels are widely used, such as laser hologram labels, watermark labels, electronic code labels, digital coded labels, barcode labels, two-dimensional code labels, and RFID labels. In principle, these anti-counterfeit labels can be copied and cannot guarantee absolute security.
  • the existing backtracking query type anti-counterfeiting technology is based on artificial anti-counterfeiting labels, such as digital codes, barcodes, two-dimensional codes, and RFID, etc., and can be copied and forged in principle; existing non-replicable natural structure graphic labels are complicated and need to be processed. The human eye judges.
  • the object of the present invention is to provide an anti-counterfeiting system and method based on chaotic graphic label, which aims to solve the problem that the existing backtracking query type anti-counterfeiting technology can be copied and forged based on the artificial anti-counterfeit label, and the non-replicable natural structure graphic label is not complicated. And it needs the judgment of the human eye.
  • an anti-counterfeiting method based on a chaotic graphic label and the anti-counterfeiting method based on the chaotic graphic label comprises:
  • the image recognition technology is used to extract the feature information of the chaotic graphic label, and the ID of the chaotic graphic label, the characteristic information and the information of the target object associated with the label are stored in the cloud data;
  • the same image recognition technology is used to calculate the feature information of the chaotic graphic label, and the corresponding tag characteristic information is found in the cloud database by using the ID of the chaotic graphic label;
  • Another object of the present invention is to provide a tag information entry system and a tag verification system, the tag information entry system and tag verification system comprising:
  • the tag information input system uses image recognition technology to extract feature information of the chaotic graphic tag, and then stores the ID of the chaotic graphic tag, the feature information, and the information of the object associated with the tag into the cloud data;
  • the tag verification system uses the same image recognition technology to calculate the feature information of the chaotic graphic tag, and then uses the ID of the chaotic graphic tag to find the corresponding tag feature information in the cloud database, and finally judges whether the matching is based on the similarity of the two feature information. If it matches, the object information associated with the current tag is returned. If it does not match, the matching failure is returned.
  • Step one using multi-angle graphic scanning technology, each time the graphic scanning is shifted by a certain angle from the last scanning;
  • Step 2 using a corner detection algorithm to calculate corner points of chaotic graphic labels of different viewing angles, and Perform feature matching to obtain matching feature points between adjacent perspective images;
  • Step 3 according to the principle of multi-view geometry of the computer, using a 7-point algorithm to calculate the basic matrix between adjacent matching images;
  • Step 4 Using the basic matrix described above, calculate a corresponding diagonal line of each feature point in any image on its adjacent matching image, and use a partial template matching algorithm to calculate other matching features existing on the polar line. point;
  • Step 5 calculating a parallax of the matching feature points of the matched image to obtain a disparity map of the adjacent matching image
  • Step 6 using the parallax principle, calculating the feature point space coordinates in the chaotic graphic label from the disparity map of the adjacent matching image obtained in the above step, and further obtaining the relative position information of the feature points in the chaotic graphic label;
  • Step 7 extracting texture information of a point distribution in a chaotic graphic label corresponding to each perspective by using a structure-based image texture feature extraction algorithm
  • Step 8 Combining the object information associated with the tag with the feature relative position information and texture information of the acquired chaotic graphic tag constitutes composite data information, assigning the ID information, and storing the composite data information and the ID information into the back-end database.
  • Step one using multi-angle graphic scanning technology, each time the graphic scanning is shifted by a certain angle from the last scanning;
  • Step 2 Using the corner detection algorithm, calculate the corner points of the chaotic graphic labels of different viewing angles, and perform feature matching to obtain matching feature points between adjacent viewing angle images;
  • Step 3 according to the principle of computer multi-vision geometry, utilize 7 a point algorithm calculates a base matrix between adjacent matching images;
  • Step 4 Using the basic matrix described above, calculate a corresponding diagonal line of each feature point in any image on its adjacent matching image, and use a partial template matching algorithm to calculate other matching features existing on the polar line. point;
  • Step 5 calculating a parallax of the matching feature points of the matched image to obtain a disparity map of the adjacent matching image
  • Step 6 using the parallax principle, calculating the disparity map of the adjacent matching images obtained in the above steps The feature point space coordinates in the chaotic graphic label are obtained, and the relative position information of the feature points in the chaotic graphic label is further obtained;
  • Step 7 extracting texture information of a point distribution in a chaotic graphic label corresponding to each perspective by using a structure-based image texture feature extraction algorithm
  • Step 8 identifying the two-dimensional code information on the label, and reading the label ID; in step 9, using the label ID to obtain the previously entered label information in the database, and the relative position information and label of the corner point of the chaotic graphic label obtained above Texture features for similarity metrics;
  • Step 10 if there is a tag in the database that has similarity between the chaotic image information and the acquired tag information, and the tag is greater than the set threshold, confirm that the associated object is authentic, and send back the object information associated with the tag; if not, if not , confirm that the related products are not genuine.
  • the present invention utilizes the non-reproducible features of chaotic graphics to design anti-counterfeit labels, and has the absolute advantage of low cost.
  • the invention utilizes the principle of image recognition and multi-view geometry to calculate the surface spatial relationship of chaotic graphics, and realizes the intelligent recognition of chaotic graphic labels according to the structural uniqueness and similarity measure of chaotic labels, and the identification is fast and accurate. Increase the user experience.
  • the invention extracts chaotic graphic texture features, and performs comparison verification in the intelligent recognition process to avoid errors caused by insufficient accuracy of chaotic graphic label spatial structure recognition, and further improves the reliability of the verification result.
  • FIG. 1 is a flowchart of an anti-counterfeiting method based on a chaotic graphic label according to an embodiment of the present invention
  • FIG. 2 is a schematic structural diagram of an anti-counterfeiting system based on a chaotic graphic label according to an embodiment of the present invention.
  • the invention utilizes a non-replicable natural chaotic structure graphic as an anti-counterfeit label, and utilizes image scanning, computer vision, graphic feature extraction, and structural texture information extraction to realize a novel anti-counterfeiting system, which has the label difficult to copy, fast and accurate identification, and use. Easy to use and wide range of features.
  • the anti-counterfeiting method based on the chaotic graphic label in the embodiment of the present invention includes:
  • the image recognition technology is used to extract the feature information of the chaotic graphic label, and the ID of the chaotic graphic label, the characteristic information and the information of the target object associated with the label are stored in the cloud data;
  • the same image recognition technology is used to calculate the feature information of the chaotic graphic label, and the corresponding tag characteristic information is found in the cloud database by using the ID of the chaotic graphic label;
  • the present invention includes two parts: a tag information entry system and a tag verification system;
  • the tag information input system uses image recognition technology to extract feature information of the chaotic graphic tag, and then stores the ID of the chaotic graphic tag, the feature information, and the information of the object associated with the tag into the cloud data;
  • the tag verification system uses the same image recognition technology to calculate the feature information of the chaotic graphic tag, and then uses the ID of the chaotic graphic tag to find the corresponding tag feature information in the cloud database, and finally judges whether the matching is based on the similarity of the two feature information. If it matches, the object information associated with the current tag is returned. If it does not match, the matching failure is returned.
  • Step one using multi-angle graphic scanning technology, each time the graphic scanning is shifted by a certain angle from the last scanning;
  • Step 2 using a corner detection algorithm to calculate corner points of chaotic graphic labels of different viewing angles, and Perform feature matching to obtain matching feature points between adjacent perspective images;
  • Step 3 according to the principle of multi-view geometry of the computer, using a 7-point algorithm to calculate the basic matrix between adjacent matching images;
  • Step 4 Using the basic matrix described above, calculate a corresponding diagonal line of each feature point in any image on its adjacent matching image, and use a partial template matching algorithm to calculate other matching features existing on the polar line. point;
  • Step 5 calculating a parallax of the matching feature points of the matched image to obtain a disparity map of the adjacent matching image
  • Step 6 using the parallax principle, calculating the feature point space coordinates in the chaotic graphic label from the disparity map of the adjacent matching image obtained in the above step, and further obtaining the relative position information of the feature points in the chaotic graphic label;
  • Step 7 extracting texture information of a point distribution in a chaotic graphic label corresponding to each perspective by using a structure-based image texture feature extraction algorithm
  • step 8 the object information associated with the tag and the feature relative position information and texture information of the chaotic graphic tag obtained above constitute composite data information, and the ID information is assigned, and the composite data information and the ID information are stored in the background database.
  • the ID information of the chaotic graphic label is placed in the two-dimensional code area of the chaotic graphic label in the form of a two-dimensional code for easy identification.
  • Step one using multi-angle graphic scanning technology, each time the graphic scanning is shifted by a certain angle from the last scanning;
  • Step 2 Using the corner detection algorithm, calculate the corner points of the chaotic graphic labels of different viewing angles, and perform feature matching to obtain matching feature points between adjacent viewing angle images;
  • Step 3 according to the principle of computer multi-vision geometry, utilize 7 a point algorithm calculates a base matrix between adjacent matching images;
  • Step 4 Using the basic matrix described above, calculate a corresponding diagonal line of each feature point in any image on its adjacent matching image, and use a partial template matching algorithm to calculate other matching features existing on the polar line.
  • Step 5 calculating a parallax of the matching feature points of the matched image to obtain a disparity map of the adjacent matching image
  • Step 6 using the parallax principle, calculating the feature point space coordinates in the chaotic graphic label from the disparity map of the adjacent matching image obtained in the above step, and further obtaining the relative position information of the feature points in the chaotic graphic label;
  • Step 7 extracting texture information of a point distribution in a chaotic graphic label corresponding to each perspective by using a structure-based image texture feature extraction algorithm
  • Step 8 identifying the two-dimensional code information on the label, and reading the label ID; in step 9, using the label ID to obtain the previously entered label information in the database, and the relative position information and label of the corner point of the chaotic graphic label obtained above Texture features for similarity metrics;
  • Step 10 if there is a tag in the database that has similarity between the chaotic image information and the acquired tag information, and the tag is greater than the set threshold, confirm that the associated object is authentic, and send back the object information associated with the tag; if not, if not , confirm that the related products are not genuine.

Landscapes

  • Business, Economics & Management (AREA)
  • Accounting & Taxation (AREA)
  • Development Economics (AREA)
  • Economics (AREA)
  • Finance (AREA)
  • Marketing (AREA)
  • Strategic Management (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Image Analysis (AREA)

Abstract

一种基于混沌图形标签的防伪系统及方法，利用图像识别技术提取混沌图形标签的特征信息，将混沌图形标签的ID、特征信息以及与标签关联的标的物的信息存入云端数据中；利用相同的图像识别技术计算出混沌图形标签的特征信息，利用混沌图形标签的ID在云端数据库中找出对应标签特征信息；根据两个特征信息相似度，判断是否匹配，匹配，则返回当前标签关联的标的物信息，不匹配，则返回匹配失败；系统包括：标签信息录入系统和标签验证系统。其利用具有不可复制的自然混沌结构图形作为防伪标签，并利用图像扫描、计算机视觉、图形特征提取、海量图形检索技术，实现防伪；具有标签难以复制、鉴别快速准确、使用简便且范围广。

Description

一种基于混沌图形标签的防伪系统及方法 技术领域
本发明属于防伪技术领域，尤其涉及一种基于混沌图形标签的防伪系统及方法。
背景技术
防伪技术是一门涉及诸多学科的交叉技术，通常利用防伪标签的难以复制性，在一定程度上将所防护的产品与标签上标明的机构或厂家进行紧密关联。随着防伪技术与市场经济的快速发展，防伪技术不但已经广泛银行、海关、税务、金融、公安、政府部门等制作的护照、证件、货币、票证和有价证券上，而且也广泛应用于烟酒、食品、药品、保健品、化妆品、服装、光盘制品以及生产资料中的农药、化肥、汽车零部件等商品的防伪上。防伪技术对控制并打击造假制假活动,维护良好的市场竞争环境，保护知识产权具有重要作用。我国制假造假泛滥，商家对知识产权保护的保护越来越重视，因此对新型防伪技术的需求非常迫切。防伪系统的核心就是防伪标签。目前广泛应用的防伪标签多种多样,如激光全息标签、水印标签、电码标签、数字编码标签、条形码标签、二维码标签以及RFID标签等。原则上来讲，这些防伪标签都可以被复制，无法保证绝对防伪。除了标签与目标物之间的关联关系的保证非常重要以外，防伪系统的另一个关键就是验伪的准确性。虽然法国Prooftag公司推出的气泡标签(Bubble Tag)的气泡部分理论上几乎不可复制，但由于气泡部分不具防揭功能，而防揭部分又完全有可能被修复或复制，导致该标签的综合不可复制性大大降低。
现有回溯查询型防伪技术基于人造防伪标签，如数字编码、条形码、二维码以及RFID等，在原则上都可被复制和伪造；现有不可复制的自然结构图形标签对不过程复杂且需要人眼判断。
发明内容
本发明的目的在于提供一种基于混沌图形标签的防伪系统及方法，旨在解决现有回溯查询型防伪技术基于人造防伪标签都可被复制和伪造，不可复制的自然结构图形标签对不过程复杂且需要人眼判断的问题。
本发明是这样实现的，一种基于混沌图形标签的防伪方法，所述基于混沌图形标签的防伪方法包括：
首先利用图像识别技术提取混沌图形标签的特征信息，将混沌图形标签的ID、特征信息以及与标签关联的标的物的信息存入云端数据中；
其次利用相同的图像识别技术计算出混沌图形标签的特征信息，利用混沌图形标签的ID在云端数据库中找出对应标签特征信息；
最后根据两个特征信息相似度，判断是否匹配，如果匹配，则返回当前标签关联的标的物信息，如果不匹配，则返回匹配失败。
本发明的另一目的在于提供一种标签信息录入系统与标签验证系统，所述标签信息录入系统与标签验证系统包括：
标签信息录入系统，利用图像识别技术提取混沌图形标签的特征信息，然后将混沌图形标签的ID、特征信息以及与标签关联的标的物的信息存入云端数据中；
标签验证系统，利用相同的图像识别技术计算出混沌图形标签的特征信息，然后利用混沌图形标签的ID在云端数据库中找出对应标签特征信息，最后根据两个特征信息相似度，判断是否匹配，如果匹配，则返回当前标签关联的标的物信息，如果不匹配，则返回匹配失败。
进一步，所述标签信息录入系统的具体步骤如下：
步骤一，采用多角度图形扫描技术，每次图形扫描时较上次扫描错开一定角度；
步骤二，利用角点检测算法，计算出不同视角的混沌图形标签的角点，并 进行特征匹配，得到相邻视角图像之间的匹配特征点；
步骤三，根据计算机多视几何原理，利用7点算法计算相邻匹配图像间的基础矩阵；
步骤四，利用上述的基础矩阵，计算出任一图像中的每个特征点在其相邻匹配图像上对应的对极线，利用局部模板匹配算法，计算出对极线上存在的其他匹配的特征点；
步骤五，计算匹配图像的匹配特征点视差，得到相邻匹配图像视差图；
步骤六，利用视差原理，由上述步骤中得到的相邻匹配图像的视差图计算出混沌图形标签中的特征点空间坐标，并进一步得到混沌图形标签中的特征点相对位置信息；
步骤七，利用基于结构的图像纹理特征提取算法，提取每个视角对应的混沌图形标签中点位分布的纹理信息；
步骤八，将标签关联的标的物信息与获取的混沌图形标签的特征相对位置信息及纹理信息构成复合数据信息，赋予ID信息，并将复合数据信息及ID信息存入后台数据库。
进一步，所述标签信息识别系统具体的步骤如下：
步骤一，采用多角度图形扫描技术，每次图形扫描时较上次扫描错开一定角度；
步骤二，利用角点检测算法，计算出不同视角的混沌图形标签的角点，并进行特征匹配，得到相邻视角图像之间的匹配特征点；步骤三，根据计算机多视几何原理，利用7点算法计算相邻匹配图像间的基础矩阵；
步骤四，利用上述的基础矩阵，计算出任一图像中的每个特征点在其相邻匹配图像上对应的对极线，利用局部模板匹配算法，计算出对极线上存在的其他匹配的特征点；
步骤五，计算匹配图像的匹配特征点视差，得到相邻匹配图像视差图；
步骤六，利用视差原理，由上述步骤中得到的相邻匹配图像的视差图计算 出混沌图形标签中的特征点空间坐标，并进一步得到混沌图形标签中的特征点相对位置信息；
步骤七，利用基于结构的图像纹理特征提取算法，提取每个视角对应的混沌图形标签中点位分布的纹理信息；
步骤八，识别出标签上的二维码信息，读取标签ID；步骤九，利用标签ID在数据库中获得之前录入的标签信息，并与以上获得的混沌图形标签的角点相对位置信息、标签纹理特征进行相似性度量；
步骤十，如果在数据库中存在混沌图像信息与上述获取的标签信息相似度大于设定阈值的标签，则确认关联标的物为正品，并发送回与标签相关联的标的物信息；反之如果不存在，则确认关联产品为非正品。
本发明提供的基于混沌图形标签的防伪系统及方法，相对现有的防伪技术，具有以下优势：
(1)本发明利用混沌图形的不可复制特征设计防伪标签，具有成本低廉的绝对优势。
(2)本发明利用图像识别与多视几何原理，计算出混沌图形的表面空间关系，并根据混沌标签的结构唯一性与相似性度量，实现混沌图形标签的智能识别，鉴别快速准确，极大程度提升用户体验。
(3)本发明提取混沌图形纹理特征，并在智能识别过程中进行对比验证，避免由于混沌图形标签空间结构识别的准确性不足而引起的误差，进一步提高了验证结果的可靠性。
附图说明
图1是本发明实施例提供的基于混沌图形标签的防伪方法流程图；
图2是本发明实施例提供的基于混沌图形标签的防伪系统结构示意图。
具体实施方式
为了使本发明的目的、技术方案及优点更加清楚明白，以下结合实施例，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。
本发明利用具有不可复制的自然混沌结构图形作为防伪标签，并利用图像扫描、计算机视觉、图形特征提取、以及结构纹理信息提取，实现一种新型防伪系统，具有标签难以复制、鉴别快速准确、使用简便和使用范围广的特点。
下面结合附图1对本发明的应用原理作详细的描述。
如图1所示，本发明实施例的基于混沌图形标签的防伪方法包括：
首先利用图像识别技术提取混沌图形标签的特征信息，将混沌图形标签的ID、特征信息以及与标签关联的标的物的信息存入云端数据中；
其次利用相同的图像识别技术计算出混沌图形标签的特征信息，利用混沌图形标签的ID在云端数据库中找出对应标签特征信息；
最后根据两个特征信息相似度，判断是否匹配，如果匹配，则返回当前标签关联的标的物信息，如果不匹配，则返回匹配失败。
如图2所示，本发明包括标签信息录入系统与标签验证系统两个部分；
标签信息录入系统，利用图像识别技术提取混沌图形标签的特征信息，然后将混沌图形标签的ID、特征信息以及与标签关联的标的物的信息存入云端数据中；
标签验证系统，利用相同的图像识别技术计算出混沌图形标签的特征信息，然后利用混沌图形标签的ID在云端数据库中找出对应标签特征信息，最后根据两个特征信息相似度，判断是否匹配，如果匹配，则返回当前标签关联的标的物信息，如果不匹配，则返回匹配失败。
标签信息录入具体的步骤如下：
步骤一，采用多角度图形扫描技术，每次图形扫描时较上次扫描错开一定角度；
步骤二，利用角点检测算法，计算出不同视角的混沌图形标签的角点，并 进行特征匹配，得到相邻视角图像之间的匹配特征点；
步骤三，根据计算机多视几何原理，利用7点算法计算相邻匹配图像间的基础矩阵；
步骤四，利用上述的基础矩阵，计算出任一图像中的每个特征点在其相邻匹配图像上对应的对极线，利用局部模板匹配算法，计算出对极线上存在的其他匹配的特征点；
步骤五，计算匹配图像的匹配特征点视差，得到相邻匹配图像视差图；
步骤六，利用视差原理，由上述步骤中得到的相邻匹配图像的视差图计算出混沌图形标签中的特征点空间坐标，并进一步得到混沌图形标签中的特征点相对位置信息；
步骤七，利用基于结构的图像纹理特征提取算法，提取每个视角对应的混沌图形标签中点位分布的纹理信息；
步骤八，将标签关联的标的物信息与上述获取的混沌图形标签的特征相对位置信息及纹理信息构成复合数据信息，赋予ID信息，并将复合数据信息及ID信息存入后台数据库。
混沌图形标签的ID信息以二维码的形式放在混沌图形标签的二维码区域处，便于识别。
标签信息识别具体的步骤如下：
步骤一，采用多角度图形扫描技术，每次图形扫描时较上次扫描错开一定角度；
步骤二，利用角点检测算法，计算出不同视角的混沌图形标签的角点，并进行特征匹配，得到相邻视角图像之间的匹配特征点；步骤三，根据计算机多视几何原理，利用7点算法计算相邻匹配图像间的基础矩阵；
步骤四，利用上述的基础矩阵，计算出任一图像中的每个特征点在其相邻匹配图像上对应的对极线，利用局部模板匹配算法，计算出对极线上存在的其他匹配的特征点；
步骤五，计算匹配图像的匹配特征点视差，得到相邻匹配图像视差图；
步骤六，利用视差原理，由上述步骤中得到的相邻匹配图像的视差图计算出混沌图形标签中的特征点空间坐标，并进一步得到混沌图形标签中的特征点相对位置信息；
步骤七，利用基于结构的图像纹理特征提取算法，提取每个视角对应的混沌图形标签中点位分布的纹理信息；
步骤八，识别出标签上的二维码信息，读取标签ID；步骤九，利用标签ID在数据库中获得之前录入的标签信息，并与以上获得的混沌图形标签的角点相对位置信息、标签纹理特征进行相似性度量；
步骤十，如果在数据库中存在混沌图像信息与上述获取的标签信息相似度大于设定阈值的标签，则确认关联标的物为正品，并发送回与标签相关联的标的物信息；反之如果不存在，则确认关联产品为非正品。
以上所述仅为本发明的较佳实施例而已，并不用以限制本发明，凡在本发明的精神和原则之内所作的任何修改、等同替换和改进等，均应包含在本发明的保护范围之内。

Claims (4)

  1. 一种基于混沌图形标签的防伪系统，其特征在于，所述基于混沌图形标签的防伪系统包括：
    标签信息录入系统，利用图像识别技术提取混沌图形标签的特征信息，然后将混沌图形标签的ID、特征信息以及与标签关联的标的物的信息存入云端数据中；
    标签验证系统，利用相同的图像识别技术计算出混沌图形标签的特征信息，然后利用混沌图形标签的ID在云端数据库中找出对应标签特征信息，根据两个特征信息相似度，判断是否匹配，如果匹配，则返回当前标签关联的标的物信息，如果不匹配，则返回匹配失败。
  2. 如权利要求1所述的基于混沌图形标签的防伪系统，其特征在于，所述标签信息录入系统的具体步骤如下：
    步骤一，采用多角度图形扫描技术，每次图形扫描时较上次扫描错开一定角度；
    步骤二，利用角点检测算法，计算出不同视角的混沌图形标签的角点，并进行特征匹配，得到相邻视角图像之间的匹配特征点；
    步骤三，根据计算机多视几何原理，利用7点算法计算相邻匹配图像间的基础矩阵；
    步骤四，利用基础矩阵，计算出任一图像中的每个特征点在相邻匹配图像上对应的对极线，利用局部模板匹配算法，计算出对极线上存在的其他匹配的特征点；
    步骤五，计算匹配图像的匹配特征点视差，得到相邻匹配图像视差图；
    步骤六，利用视差原理，由得到的相邻匹配图像的视差图计算出混沌图形标签中的特征点空间坐标，并进一步得到混沌图形标签中的特征点相对位置信息；
    步骤七，利用基于结构的图像纹理特征提取算法，提取每个视角对应的混沌图形标签中点位分布的纹理信息；
    步骤八，将标签关联的标的物信息与获取的混沌图形标签的特征相对位置信息及纹理信息构成复合数据信息，赋予ID信息，并将复合数据信息及ID信息存入后台数据库。
  3. 如权利要求1所述的基于混沌图形标签的防伪系统，其特征在于，所述标签信息识别系统具体的步骤如下：
    步骤一，采用多角度图形扫描技术，每次图形扫描时较上次扫描错开；
    步骤二，利用角点检测算法，计算出不同视角的混沌图形标签的角点，并进行特征匹配，得到相邻视角图像之间的匹配特征点；
    步骤三，根据计算机多视几何原理，利用7点算法计算相邻匹配图像间的基础矩阵；
    步骤四，利用基础矩阵，计算出任一图像中的每个特征点在相邻匹配图像上对应的对极线，利用局部模板匹配算法，计算出对极线上存在的其他匹配的特征点；
    步骤五，计算匹配图像的匹配特征点视差，得到相邻匹配图像视差图；
    步骤六，利用视差原理，由得到的相邻匹配图像的视差图计算出混沌图形标签中的特征点空间坐标，并进一步得到混沌图形标签中的特征点相对位置信息；
    步骤七，利用基于结构的图像纹理特征提取算法，提取每个视角对应的混沌图形标签中点位分布的纹理信息；
    步骤八，识别出标签上的二维码信息，读取标签ID；步骤九，利用标签ID在数据库中获得之前录入的标签信息，并与以上获得的混沌图形标签的角点相对位置信息、标签纹理特征进行相似性度量；
    步骤十，如果在数据库中存在混沌图像信息与上述获取的标签信息相似度大于设定阈值的标签，则确认关联标的物为正品，并发送回与标签相关联的标的物信息；反之如果不存在，则确认关联产品为非正品。
  4. 一种基于混沌图形标签的防伪方法，其特征在于，所述基于混沌图形标 签的防伪方法包括：
    首先利用图像识别技术提取混沌图形标签的特征信息，将混沌图形标签的ID、特征信息以及与标签关联的标的物的信息存入云端数据中；
    其次利用相同的图像识别技术计算出混沌图形标签的特征信息，利用混沌图形标签的ID在云端数据库中找出对应标签特征信息；
    最后根据两个特征信息相似度，判断是否匹配，如果匹配，则返回当前标签关联的标的物信息，如果不匹配，则返回匹配失败。
PCT/CN2015/086300 2015-06-23 2015-08-06 一种基于混沌图形标签的防伪系统及方法 WO2016206173A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201510349037.XA CN105023163B (zh) 2015-06-23 2015-06-23 一种基于混沌图形标签的防伪系统及方法
CN201510349037.X 2015-06-23

Publications (1)

Publication Number Publication Date
WO2016206173A1 true WO2016206173A1 (zh) 2016-12-29

Family

ID=54413114

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2015/086300 WO2016206173A1 (zh) 2015-06-23 2015-08-06 一种基于混沌图形标签的防伪系统及方法

Country Status (2)

Country Link
CN (1) CN105023163B (zh)
WO (1) WO2016206173A1 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110428027A (zh) * 2019-07-22 2019-11-08 杭州沃朴物联科技有限公司 基于lcd防伪标签的识别验伪方法、装置、设备及介质
CN110533704A (zh) * 2019-07-22 2019-12-03 杭州沃朴物联科技有限公司 油墨标签的识别验伪方法、装置、设备及介质
CN113935343A (zh) * 2021-10-12 2022-01-14 南通大学 一种基于字符串转图片加密的商品防伪码生成方法

Families Citing this family (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103559473B (zh) * 2013-10-28 2017-08-01 汝思信息技术（上海）有限公司 利用特征图像来实现承印物防伪的方法和系统
CN106960351A (zh) * 2016-01-11 2017-07-18 深圳市安普盛科技有限公司 一种商品防伪、验证方法和系统及条码扫描装置
CN106203427A (zh) * 2016-07-06 2016-12-07 四川大学 一种基于外包装特征的香烟真假识别系统
CN106485256A (zh) * 2016-10-10 2017-03-08 宋育锋 基于sift特征点的双标签相对位置信息构建方法
CN107633048B (zh) * 2017-09-15 2021-02-26 国网重庆市电力公司电力科学研究院 一种图像标注鉴别方法及系统
CN109754243A (zh) * 2017-11-01 2019-05-14 杭州沃朴物联科技有限公司 基于防伪标识的收款二维码防调包系统及方法
CN108373001A (zh) * 2018-02-01 2018-08-07 王学斌 一种智能工业机器人自动化仓储系统
CN109767826A (zh) * 2019-01-21 2019-05-17 河西学院 一种医疗摄影数据的获取方法及医疗摄影成像系统
CN110222602A (zh) * 2019-05-23 2019-09-10 艾科芯（深圳）智能科技有限公司 防伪识别方法、系统、设备终端和计算机可读存储介质
CN110443623A (zh) * 2019-07-22 2019-11-12 杭州沃朴物联科技有限公司 针对防伪标签的识别验伪方法、装置、设备及介质
CN112598008B (zh) * 2020-12-25 2021-12-03 上海大学 不可复制防伪标签的薄膜图案数据库建立及分类识别方法

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1108781A (zh) * 1994-12-03 1995-09-20 浙江大学 不可复制的票证防伪印记产生方法
US20100111445A1 (en) * 2008-11-05 2010-05-06 Chih-Yi Yang Portable image-extracting device for identifying tiny images and method of the same
CN102663600A (zh) * 2012-01-17 2012-09-12 付强 基于数字水印和条码的防伪系统及其防伪方法和应用
CN103310256A (zh) * 2013-06-03 2013-09-18 何南炎 基于图像对比的连续实物防伪标签的制备及鉴别方法
CN104112392A (zh) * 2013-04-20 2014-10-22 万战斌 一种防伪标签和检伪方法

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN100531028C (zh) * 2005-09-12 2009-08-19 西安交通大学 基于混沌特性的图像防伪方法
JP2011253424A (ja) * 2010-06-03 2011-12-15 Bill Haraguchi 画像認識装置及び画像認識方法並びに情報提供システム
CN102968927B (zh) * 2012-11-02 2015-11-18 吴建辉 防伪标签、防伪标签的识别装置及方法
CN203930844U (zh) * 2013-09-29 2014-11-05 苏州大学 基于图像识别和nfc的字画防伪识别系统
CN103870862A (zh) * 2014-03-03 2014-06-18 汤永平 利用分离图形随机结合进行防伪的方法及实现
CN104504360B (zh) * 2014-09-28 2018-11-13 中国人民公安大学 古陶瓷自动认证方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1108781A (zh) * 1994-12-03 1995-09-20 浙江大学 不可复制的票证防伪印记产生方法
US20100111445A1 (en) * 2008-11-05 2010-05-06 Chih-Yi Yang Portable image-extracting device for identifying tiny images and method of the same
CN102663600A (zh) * 2012-01-17 2012-09-12 付强 基于数字水印和条码的防伪系统及其防伪方法和应用
CN104112392A (zh) * 2013-04-20 2014-10-22 万战斌 一种防伪标签和检伪方法
CN103310256A (zh) * 2013-06-03 2013-09-18 何南炎 基于图像对比的连续实物防伪标签的制备及鉴别方法

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110428027A (zh) * 2019-07-22 2019-11-08 杭州沃朴物联科技有限公司 基于lcd防伪标签的识别验伪方法、装置、设备及介质
CN110533704A (zh) * 2019-07-22 2019-12-03 杭州沃朴物联科技有限公司 油墨标签的识别验伪方法、装置、设备及介质
CN110533704B (zh) * 2019-07-22 2022-11-11 杭州沃朴物联科技有限公司 油墨标签的识别验伪方法、装置、设备及介质
CN110428027B (zh) * 2019-07-22 2023-06-23 杭州沃朴物联科技有限公司 基于lcd防伪标签的识别验伪方法、装置、设备及介质
CN113935343A (zh) * 2021-10-12 2022-01-14 南通大学 一种基于字符串转图片加密的商品防伪码生成方法
CN113935343B (zh) * 2021-10-12 2022-06-21 南通大学 一种基于字符串转图片加密的商品防伪码生成方法

Also Published As

Publication number Publication date
CN105023163A (zh) 2015-11-04
CN105023163B (zh) 2018-08-21

Similar Documents

Publication Publication Date Title
WO2016206173A1 (zh) 一种基于混沌图形标签的防伪系统及方法
US10915749B2 (en) Authentication of a suspect object using extracted native features
US10614302B2 (en) Controlled authentication of physical objects
US10192140B2 (en) Database for detecting counterfeit items using digital fingerprint records
CN105513221B (zh) 一种基于三维人脸识别的atm机防欺诈装置及系统
US11748768B2 (en) Commodity anti-counterfeit verification system based on natural biological information
CN104537544A (zh) 带覆盖层的基于背景纹理特征提取的商品二维码防伪方法及系统
CN109472337B (zh) 一种基于随机特征的标签防伪方法和装置
MXPA05001124A (es) Etiquetas a prueba de falsificacion y manipulacion con caracteristicas de incidencia aleatoria.
CN205068462U (zh) 嵌入有机器可读图像的卡、附接卡的产品和连接卡的产品
CN106096348B (zh) 一种基于多维码的证卡验证系统及方法
Cozzella et al. Drug packaging security by means of white-light speckle
CN105005904A (zh) 一种基于rfid编码的艺术品溯源方法
KR101595766B1 (ko) 정품 인증용 라벨 및 그 라벨을 이용한 인증방법
US20240193621A1 (en) Method and system for protection against counterfeiting
CN105117920B (zh) 一种图片电子凭证的防伪方法及装置
US20140064598A1 (en) Mobile and desktop application for extracting geometric elements and mapping to a master key-code database
CN103065547A (zh) 商品复合防伪码结构及其生成和验证方法
Su et al. Robust 2D engineering CAD graphics hashing for joint topology and geometry authentication via covariance-based descriptors
Dalla Preda et al. Exploring NFT Validation through Digital Watermarking
Priya et al. Fake Product Detection Using Diverse Technologies
Haoliang et al. The Application of Digital Signature Technology and fingerprint identification in 2D Barcode person identity
Ahlawat et al. Offline signature verification using local binary pattern and octave pattern
Sanh et al. Effective know-your-customer method for secure and trustworthy non-fungible tokens in media assets
US20240070683A1 (en) Method and system for protection against counterfeiting

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: 15896068

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: 15896068

Country of ref document: EP

Kind code of ref document: A1