CN106485256A - Double label relative position information construction methods based on SIFT feature point - Google Patents

Double label relative position information construction methods based on SIFT feature point Download PDF

Info

Publication number
CN106485256A
CN106485256A CN201610883528.7A CN201610883528A CN106485256A CN 106485256 A CN106485256 A CN 106485256A CN 201610883528 A CN201610883528 A CN 201610883528A CN 106485256 A CN106485256 A CN 106485256A
Authority
CN
China
Prior art keywords
label
sift feature
feature point
lower floor
sift
Prior art date
Legal status (The legal status 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 status listed.)
Pending
Application number
CN201610883528.7A
Other languages
Chinese (zh)
Inventor
宋育锋
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Individual
Original Assignee
Individual
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 Individual filed Critical Individual
Priority to CN201610883528.7A priority Critical patent/CN106485256A/en
Publication of CN106485256A publication Critical patent/CN106485256A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/46—Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features
    • G06V10/462—Salient features, e.g. scale invariant feature transforms [SIFT]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/22—Matching criteria, e.g. proximity measures

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Artificial Intelligence (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • General Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Image Analysis (AREA)

Abstract

The invention discloses a kind of double label relative position information construction methods based on SIFT feature point, it is specially double labels for commodity and shoot the picture being formed, sift characteristic point according to detection, build the relative position relation of double labels in picture, using relative position relation as picture symbolic characteristic, false proof for commodity.The inventive method enriches the description information of picture by introducing SIFT feature point, and amount of calculation is little, and robustness is good.

Description

Double label relative position information construction methods based on SIFT feature point
Technical field
The present invention relates to image false-proof field, more particularly, to a kind of double label relative position informations based on SIFT feature point Construction method.
Background technology
At present, picture anti-counterfeiting technology is widely used in the fields such as the circulation of commodity, sale.A part of method adopts information Digital watermark in security fields.Another part method concentrates on the geometric error modeling feature extracting picture.But these sides above-mentioned Method often focuses on pixel level, and shortcoming is that the most of pixel needing region will participate in, computationally intensive.
Content of the invention
The technical problem to be solved in the present invention is for defect of the prior art, provides one kind to be based on SIFT feature point Double label relative position information construction methods.
The technical solution adopted for the present invention to solve the technical problems is：A kind of double labels based on SIFT feature point are relatively Positional information construction method, described pair of label includes upper label and lower floor's label；Described upper label entirety boundary under In the bounds of layer label；The texture information of described lower floor label is being used for including abundant shade of gray characteristic information The texture information that SIFT feature is extracted；
Comprise the following steps：
1) real goods carrying double-layer label are selected, the picture that its pair of label is shot with n secondary difference angle is designated as Pi, i=1 ..., n；Select wherein one pair as template picture, be designated as P1, the relative displacement of levels label therein is as ginseng According to thing；Select another pair as picture to be measured, be designated as P2；To picture P1, P2Middle levels label carries out SIFT feature point detection, Obtain P1, P2The SIFT feature point set F of upper label1, F2；P1, P2The SIFT feature point set f of lower floor's label1, f2；On obtaining respectively The physical coordinates of lower floor's label SIFT feature point, the coordinate information of the SIFT feature point of storage at least 70%；
2) to P1, P2Lower floor's label carries out SIFT feature registration, obtains P1And P2In corresponding SIFT registration point set f1 ' With collection f2 ', the location coordinate information of all SIFT feature points in storage f1 ' and f2 '；
3) position coordinateses utilizing all SIFT feature points in f1 ' and f2 ' construct P2To P1Affine transformation matrix M, by P1 And P2Lower floor's label normalize to same physical location, brought with weakening different shooting angles putting position different with commodity Impact；
4) to P1、P2Upper label carries out SIFT feature registration, obtains P1And P2In corresponding SIFT registration point set F1' and F2', store F1' and F2' in all SIFT feature points location coordinate information；To F2' in SIFT feature point utilize affine matrix meter Calculate the SIFT feature point set F after affine transformation2”；
5) calculate F2" and F1' in the distance between corresponding SIFT feature point d, this distance reflects P2Label phase at the middle and upper levels Relative position information to lower floor's label.
By such scheme, the area of described lower floor label is more than or equal to 3 times of upper label area.
By such scheme, described lower floor label shading at least can guarantee that and detects 300 SIFT feature points.
The beneficial effect comprise that：The inventive method enriches the description letter of picture by introducing SIFT feature point Breath, amount of calculation is little, and robustness is good.
Brief description
Below in conjunction with drawings and Examples, the invention will be further described, in accompanying drawing：
Fig. 1 is double label schematic diagrams of the embodiment of the present invention；
Fig. 2 is double label characteristics point detects schematic diagrams of the embodiment of the present invention.
Specific embodiment
In order that the objects, technical solutions and advantages of the present invention become more apparent, with reference to embodiments, to the present invention It is further elaborated.It should be appreciated that specific embodiment described herein, only in order to explain the present invention, is not used to limit Determine the present invention.
As shown in figure 1, a kind of double label relative position information construction methods based on SIFT feature point, described pair of label bag Include upper label and lower floor's label；Described upper label entirety boundary is in the bounds of lower floor's label；Described lower floor mark The texture information signed includes abundant shade of gray characteristic information, extracts for SIFT feature；
Comprise the following steps：
1) real goods carrying double-layer label are selected, the picture that its pair of label is shot with n secondary difference angle is designated as Pi, i=1 ..., n；Select wherein one pair as template picture, be designated as P1, the relative displacement of levels label therein is as ginseng According to thing；Select another pair as picture to be measured, be designated as P2；As Fig. 2, to picture P1, P2Middle levels label carries out SIFT feature point Detection, obtains P1, P2The SIFT feature point set F of upper label1, F2；P1, P2The SIFT feature point set f of lower floor's label1, f2, picture As Fig. 2 after SIFT feature detection；Obtain the physical coordinates of levels label SIFT feature point, the SIFT of storage 70% respectively The coordinate information of characteristic point；
2) to P1, P2Lower floor's label carries out SIFT feature registration, obtains P1And P2In corresponding SIFT registration point set f1 ' With collection f2 ', the location coordinate information of all SIFT feature points in storage f1 ' and f2 '；
3) position coordinateses utilizing all SIFT feature points in f1 ' and f2 ' construct P2To P1Affine transformation matrix M, by P1 And P2Lower floor's label normalize to same physical location, brought with weakening different shooting angles putting position different with commodity Impact；
4) to P1, P2Upper label carries out SIFT feature registration, obtains P1And P2In corresponding SIFT registration point set F1' and F2', store F1' and F2' in all SIFT feature points location coordinate information；To F2' in SIFT feature point utilize affine matrix meter Calculate the SIFT feature point set F after affine transformation2”.
5) calculate F2" and F1' in the distance between corresponding SIFT feature point d, this distance reflects P2Label phase at the middle and upper levels Relative position information to lower floor's label.
It should be appreciated that for those of ordinary skills, can be improved according to the above description or be converted, And all these modifications and variations all should belong to the protection domain of claims of the present invention.

Claims (3)

1. a kind of double label relative position information construction methods based on SIFT feature point, the described pair of label include upper label and Lower floor's label；Described upper label entirety boundary is in the bounds of lower floor's label；The texture information of described lower floor label It is the texture information extracting for SIFT feature including abundant shade of gray characteristic information；
Comprise the following steps：
1) real goods carrying double-layer label are selected, the picture that its pair of label is shot with n secondary difference angle is designated as Pi, i= 1,…,n；Select wherein one pair as template picture, be designated as P1, the relative displacement of levels label therein is as object of reference； Select another pair as picture to be measured, be designated as P2；To picture P1, P2Middle levels label carries out SIFT feature point detection, obtains P1, P2The SIFT feature point set F of upper label1, F2；P1, P2The SIFT feature point set f of lower floor's label1, f2；Obtain levels respectively The physical coordinates of label SIFT feature point, the coordinate information of the SIFT feature point of storage at least 70%；
2) to P1, P2Lower floor's label carries out SIFT feature registration, obtains P1And P2In corresponding SIFT registration point set f1 ' and collection F2 ', the location coordinate information of all SIFT feature points in storage f1 ' and f2 '；
3) position coordinateses utilizing all SIFT feature points in f1 ' and f2 ' construct P2To P1Affine transformation matrix M, by P1And P2 Lower floor's label normalize to same physical location, to weaken the shadow that different shooting angles putting position different with commodity brings Ring；
4) to P1、P2Upper label carries out SIFT feature registration, obtains P1And P2In corresponding SIFT registration point set F1' and F2', Storage F1' and F2' in all SIFT feature points location coordinate information；To F2' in SIFT feature point to utilize affine matrix to calculate imitative Penetrate the SIFT feature point set F after conversion2”；
5) calculate F2" and F1' in the distance between corresponding SIFT feature point d, this distance reflects P2At the middle and upper levels label relatively under The relative position information of layer label.
2. the double label relative position information construction methods based on SIFT feature point according to claim 1, its feature exists In the area of described lower floor label is more than or equal to 3 times of upper label area.
3. the double label relative position information construction methods based on SIFT feature point according to claim 1, its feature exists In described lower floor label shading at least can detect 300 SIFT feature points.
CN201610883528.7A 2016-10-10 2016-10-10 Double label relative position information construction methods based on SIFT feature point Pending CN106485256A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610883528.7A CN106485256A (en) 2016-10-10 2016-10-10 Double label relative position information construction methods based on SIFT feature point

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610883528.7A CN106485256A (en) 2016-10-10 2016-10-10 Double label relative position information construction methods based on SIFT feature point

Publications (1)

Publication Number Publication Date
CN106485256A true CN106485256A (en) 2017-03-08

Family

ID=58269607

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610883528.7A Pending CN106485256A (en) 2016-10-10 2016-10-10 Double label relative position information construction methods based on SIFT feature point

Country Status (1)

Country Link
CN (1) CN106485256A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106981250A (en) * 2016-10-10 2017-07-25 宋育锋 Antifalsification label based on article surface texture

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101556647A (en) * 2009-05-20 2009-10-14 哈尔滨理工大学 mobile robot visual orientation method based on improved SIFT algorithm
CN101782969A (en) * 2010-02-26 2010-07-21 浙江大学 Reliable image characteristic matching method based on physical positioning information
CN102486830A (en) * 2010-12-01 2012-06-06 无锡锦腾智能科技有限公司 Object micro texture identifying method based on spatial alternation consistency
CN102737261A (en) * 2011-04-08 2012-10-17 吴善斌 Barcode inquiring marker and barcode generation method thereof
CN105023163A (en) * 2015-06-23 2015-11-04 杭州沃朴物联科技有限公司 Anti-counterfeiting system based on chaotic graphic label and method
CN105117757A (en) * 2015-08-26 2015-12-02 复旦大学无锡研究院 Quick response code encryption and decryption method based on random textures
CN105488512A (en) * 2015-11-27 2016-04-13 南京理工大学 Sift feature matching and shape context based test paper inspection method
CN105741295A (en) * 2016-02-01 2016-07-06 福建师范大学 High-resolution remote sensing image registration method based on local invariant feature point

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101556647A (en) * 2009-05-20 2009-10-14 哈尔滨理工大学 mobile robot visual orientation method based on improved SIFT algorithm
CN101782969A (en) * 2010-02-26 2010-07-21 浙江大学 Reliable image characteristic matching method based on physical positioning information
CN102486830A (en) * 2010-12-01 2012-06-06 无锡锦腾智能科技有限公司 Object micro texture identifying method based on spatial alternation consistency
CN102737261A (en) * 2011-04-08 2012-10-17 吴善斌 Barcode inquiring marker and barcode generation method thereof
CN105023163A (en) * 2015-06-23 2015-11-04 杭州沃朴物联科技有限公司 Anti-counterfeiting system based on chaotic graphic label and method
CN105117757A (en) * 2015-08-26 2015-12-02 复旦大学无锡研究院 Quick response code encryption and decryption method based on random textures
CN105488512A (en) * 2015-11-27 2016-04-13 南京理工大学 Sift feature matching and shape context based test paper inspection method
CN105741295A (en) * 2016-02-01 2016-07-06 福建师范大学 High-resolution remote sensing image registration method based on local invariant feature point

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
冷俊敏 等: "三维显示技术现状与发展", 《中国印刷与包装研究》 *
王鹏翔: "基于SIFT的商标图像检索", 《中国优秀硕士学位论文全文数据库 信息科技辑（月刊）》 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106981250A (en) * 2016-10-10 2017-07-25 宋育锋 Antifalsification label based on article surface texture

Similar Documents

Publication Publication Date Title
CN105931295B (en) A kind of geologic map Extracting Thematic Information method
CN106446894B (en) A method of based on outline identification ball-type target object location
CN106250895B (en) A kind of remote sensing image region of interest area detecting method
CN104156965B (en) A kind of automatic quick joining method of Mine Monitoring image
CN104167003A (en) Method for fast registering remote-sensing image
CN104778679A (en) Gaofen-1 satellite data-based control point graphic element rapid-matching method
CN103154972A (en) Text-based 3D augmented reality
Sheng et al. Automated image registration for hydrologic change detection in the lake-rich Arctic
CN102254144A (en) Robust method for extracting two-dimensional code area in image
CN101551732A (en) Method for strengthening reality having interactive function and a system thereof
CN111160291B (en) Human eye detection method based on depth information and CNN
CN106682641A (en) Pedestrian identification method based on image with FHOG- LBPH feature
CN104123529A (en) Human hand detection method and system thereof
CN104933738A (en) Visual saliency map generation method based on local structure detection and contrast
CN103544491A (en) Optical character recognition method and device facing complex background
CN104143077B (en) Pedestrian target search method and system based on image
CN106709952B (en) A kind of automatic calibration method of display screen
CN106778510B (en) Method for matching high-rise building characteristic points in ultrahigh-resolution remote sensing image
CN103632376A (en) Method for suppressing partial occlusion of vehicles by aid of double-level frames
CN107527366A (en) A kind of camera tracking towards depth camera
CN106407973A (en) Robust recognition method for AR code marked on cylinder
JP2008288684A (en) Person detection device and program
JP2003141546A (en) Image processing method
CN106485256A (en) Double label relative position information construction methods based on SIFT feature point
CN105243661A (en) Corner detection method based on SUSAN operator

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20170308