CN110490269A - A kind of oil painting uniqueness discrimination method based under multispectral light source - Google Patents

A kind of oil painting uniqueness discrimination method based under multispectral light source Download PDF

Info

Publication number
CN110490269A
CN110490269A CN201910791043.9A CN201910791043A CN110490269A CN 110490269 A CN110490269 A CN 110490269A CN 201910791043 A CN201910791043 A CN 201910791043A CN 110490269 A CN110490269 A CN 110490269A
Authority
CN
China
Prior art keywords
oil painting
identified
optical signature
uniqueness
light source
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
CN201910791043.9A
Other languages
Chinese (zh)
Inventor
陈景茵
李浩铭
马天骄
刘家隆
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.)
Guangzhou Yitechnetium Culture Creative Co ltd
Original Assignee
Guangzhou Yitechnetium Culture Creative Co ltd
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 Guangzhou Yitechnetium Culture Creative Co ltd filed Critical Guangzhou Yitechnetium Culture Creative Co ltd
Priority to CN201910791043.9A priority Critical patent/CN110490269A/en
Publication of CN110490269A publication Critical patent/CN110490269A/en
Pending legal-status Critical Current

Links

Classifications

    • 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
    • G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
    • 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
    • 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
    • G06V10/50—Extraction of image or video features by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis
    • 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
    • G06V10/58—Extraction of image or video features relating to hyperspectral data

Landscapes

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

Abstract

The present invention relates to the anti-fake certificate field of oil painting, specially a kind of oil painting uniqueness discrimination method based under multispectral light source includes the following steps: that S1. obtains m × n width image of the oil painting A to be identified at multiband m, multiple light courcess angle n；S2. the optical signature of m × n width image of oil painting A to be identified is extracted respectively；S3. the optical signature of the oil painting authentic work B corresponding with oil painting A to be identified stored in oil painting authentication data library is obtained；S4. according to the optical signature of the oil painting authentic work B corresponding with A stored in the optical signature of oil painting A to be identified and database, the similarity of A and oil painting authentic work B are obtained, to identify the uniqueness of oil painting A to be identified.By extracting optical signature respectively to multiple image of the oil painting A to be identified under multispectral light source, and the optical signature of the oil painting authentic work B corresponding with A stored in the optical signature and database progress similarity is compared, the uniqueness of oil painting A to be identified, at low cost, high-efficient, identification accuracy height can be identified.

Description

A kind of oil painting uniqueness discrimination method based under multispectral light source
Technical field
The present invention relates to the anti-fake certificate technical fields of oil painting, more particularly, to a kind of based under multispectral light source Oil painting uniqueness discrimination method.
Background technique
With rapid economic development, oil painting market is more flourishing, and trading volume and the amount of money are constantly soaring, friendship relevant to oil painting Easily, the activity such as exhibition is increasingly frequent.In the intermediate links (transaction, mortgage, outer rent etc.) of oil painting, the dispute of the oil painting true and false, oil painting are dirty The events such as damage confirmation of responsibility dispute happen occasionally.Existing oil painting identification, which depends on, identifies expert, this needs to rely on personal warp It tests, subjective ingredient is big, while expending a large amount of manpowers and time, it is difficult to ensure that the accuracy identified.Therefore, market there is an urgent need to A kind of low cost, high efficiency, high accuracy discrimination method be compared for the similarity of oil painting, and then it is unique to reach discrimination Property, the purpose for differentiating responsibility.
Rich in color, changeful, infrared light, ultraviolet light of the pigment of different color for different-waveband of oil Absorptivity, reflectivity, refractive index it is different, so the oil painting image under multispectral light source irradiation has uniqueness, Therefore, the Computer imaging analysis system in conjunction with comparison technology of the feature extraction based on oil painting under multispectral light source can be used as The technical method of oil uniqueness identification.
Summary of the invention
The present invention is directed to overcome above-mentioned defect in the prior art, provide a kind of unique based on the oil painting under multispectral light source Property discrimination method, it is at low cost, high-efficient, to identify accuracy high.
In order to achieve the above objectives, the technical solution adopted by the present invention is that: a kind of oil painting based under multispectral light source is provided Uniqueness discrimination method, includes the following steps:
S1. m × n width image of the oil painting A to be identified at multiband m, multiple light courcess angle n is obtained；
S2. the optical signature of m × n width image of oil painting A to be identified is extracted respectively；
S3. the optical signature of the oil painting authentic work B corresponding with oil painting A to be identified stored in oil painting authentication data library is obtained；
S4. according to the oil painting authentic work B corresponding with oil painting A to be identified stored in the optical signature of oil painting A to be identified and database Optical signature, the similarity of oil painting A to be identified Yu oil painting authentic work B are obtained, to identify the uniqueness of oil painting A to be identified.
In above scheme, by extracting optical signature respectively to multiple image of the oil painting A to be identified under multispectral light source, And it is the optics of the oil painting authentic work B corresponding with oil painting A to be identified stored in the optical signature of oil painting A to be identified and database is special Sign carries out similarity comparison, can identify the uniqueness of oil painting A to be identified, that is, determine whether oil painting A to be identified is oil painting authentic work B, the discrimination method is at low cost, high-efficient, identification accuracy is high.
Preferably, the acquisition methods of m × n width image of oil painting A to be identified are as follows in step S1: extraneous unknown that can shield In the dark place of light source, oil painting A to be identified is shot respectively in the multispectral light source of m wave band, n light source angle using multispectral camera Image under degree obtains m × n width image altogether.Setting can be to avoid the influence of other external environments in this way, and then improves and identify standard Exactness.
It is further preferred that n light-source angle is respectively front, left side, right side and the back side of oil painting A to be identified. Setting can obtain the oil painting image after light reflection, refraction and transmission respectively in this way, and the oil painting image of multiple angles is convenient for polygonal Degree extracts optical signature, and then is compared by the similarity of multi-angle, improves the accuracy of identification.
Preferably, remember p=m × n, the optical signature of each image respectively includes area optical feature and entirety in step S2 Optical signature, acquisition methods are as follows:
S21. each image is divided into the identical region of q area, and obtains the intensity histogram diagram data in each region respectively, Perception hash algorithm is used to obtain 64 hash values of each area grayscale histogram data respectively and as area optical feature； The area optical feature for remembering all width images is Ai,j, the integer that wherein i is 1 to p, the integer that j is 1 to q；
S22. 64 hash values and optical signature as a whole of each image are obtained respectively using perception hash algorithm；Note is all The whole optical signature of width image is Ak, wherein k be 1 to p integer.
Perceptual hash algorithm is the prior art, is one kind of hash algorithm, mainly does the search work of similar image, herein Its principle is no longer described in detail, perceptual hash algorithm has no matter how height, the brightness even color of image change, will not all change The advantages of cryptographic Hash of image；Therefore the area optical feature and entirety of each image can be accurately obtained using perception hash algorithm Optical signature, and then improve and identify accuracy.
Preferably, the area optical feature for remembering oil painting authentic work B is Bi,j, whole optical signature is Bk, to be identified in step S4 The similarity of oil painting A and oil painting authentic work B includes area optical characteristic similarity and whole optical signature similarity, oil painting to be identified The uniqueness discrimination method of A is as follows:
S41.Ai,jWith Bi,jWith the identical digit/64*100% of binary representation number, p × q area optical feature is obtained altogether Similarity data；According to p whole optical signature similarity data acquisition Parameters of Normal Distribution μ1With
S42.AkWith BkWith the identical digit/64*100% of binary representation number, p whole optical signature similarity is obtained altogether Data；According to p × q area optical characteristic similarity data acquisition Parameters of Normal Distribution μ2With
If S43. μ1>=90 and μ2>=80 andAndIt then can be concluded that oil painting A to be identified and oil painting authentic work B From same width oil painting, that is, determine the uniqueness of oil painting A to be identified.
Preferably, wave band number m is 10, and light-source angle n is 4, p=40, q=100.
Preferably, the pixel density of each image is 300ppi.
Compared with prior art, the invention has the benefit that
By extracting optical signature respectively to multiple image of the oil painting A to be identified under multispectral light source, and by oil painting A to be identified Optical signature and database in the optical signature of oil painting authentic work B corresponding with oil painting A to be identified that stores carry out similarity ratio It is right, the uniqueness of oil painting A to be identified can be identified, that is, determine whether oil painting A to be identified is oil painting authentic work B, the discrimination method at This is low, high-efficient, identification accuracy is high.
Specific embodiment
Attached drawing of the present invention only for illustration, is not considered as limiting the invention.It is following in order to more preferably illustrate Embodiment, the certain components of attached drawing have omission, zoom in or out, and do not represent the size of actual product；For art technology For personnel, the omission of the possibility of certain known features and its explanation be will be understood by attached drawing.
Embodiment
The present embodiment provides a kind of oil painting uniqueness discrimination methods based under multispectral light source, include the following steps:
S1. m × n width image of the oil painting A to be identified at multiband m, multiple light courcess angle n is obtained；
S2. the optical signature of m × n width image of oil painting A to be identified is extracted respectively；
S3. the optical signature of the oil painting authentic work B corresponding with oil painting A to be identified stored in oil painting authentication data library is obtained；
S4. according to the oil painting authentic work B corresponding with oil painting A to be identified stored in the optical signature of oil painting A to be identified and database Optical signature, the similarity of oil painting A to be identified Yu oil painting authentic work B are obtained, to identify the uniqueness of oil painting A to be identified.
When oil painting authentic work comes out for the first time, according to the operation of S1 to S3, carries out optical signature in advance to oil painting authentic work and adopt Collection, and its optical signature is stored in database profession, using the comparison sample identified as later oil painting.
The present invention is incited somebody to action by extracting optical signature respectively to multiple image of the oil painting A to be identified under multispectral light source The optical signature of the oil painting authentic work B corresponding with oil painting A to be identified stored in the optical signature and database of oil painting A to be identified into Row similarity compares, and can identify the uniqueness of oil painting A to be identified, that is, determine whether oil painting A to be identified is oil painting authentic work B, should Discrimination method is at low cost, high-efficient, identification accuracy is high.
Wherein, the acquisition methods of m × n width image of oil painting A to be identified are as follows in step S1: can shield extraneous unknown light In the dark place in source, oil painting A to be identified is shot respectively in the multispectral light source of m wave band, n light-source angle using multispectral camera Under image, altogether obtain m × n width image.Setting can be to avoid the influence of other external environments in this way, and then improves and identify accurately Degree.
In the present embodiment, n light-source angle is respectively front, left side, right side and the back side of oil painting A to be identified.This Sample setting can obtain the oil painting image after light reflection, refraction and transmission respectively, and the oil painting image of multiple angles is convenient for multi-angle Optical signature is extracted, and then is compared by the similarity of multi-angle, the accuracy of identification is improved.
In addition, the optical signature of each image respectively includes area optical feature and whole light in note p=m × n, step S2 Feature is learned, acquisition methods are as follows:
S21. each image is divided into the identical region of q area, and obtains the intensity histogram diagram data in each region respectively, Perception hash algorithm is used to obtain 64 hash values of each area grayscale histogram data respectively and as area optical feature； The area optical feature for remembering all width images is Ai,j, the integer that wherein i is 1 to p, the integer that j is 1 to q；
S22. 64 hash values and optical signature as a whole of each image are obtained respectively using perception hash algorithm；Note is all The whole optical signature of width image is Ak, wherein k be 1 to p integer.
Perceptual hash algorithm is the prior art, is one kind of hash algorithm, mainly does the search work of similar image, herein Its principle is no longer described in detail, perceptual hash algorithm has no matter how height, the brightness even color of image change, will not all change The advantages of cryptographic Hash of image；Therefore the area optical feature and entirety of each image can be accurately obtained using perception hash algorithm Optical signature, and then improve and identify accuracy.
Wherein, the area optical feature for remembering oil painting authentic work B is Bi,j, whole optical signature is Bk, oil to be identified in step S4 The similarity for drawing A and oil painting authentic work B includes area optical characteristic similarity and whole optical signature similarity, oil painting A to be identified Uniqueness discrimination method it is as follows:
S41.Ai,jWith Bi,jWith the identical digit/64*100% of binary representation number, p × q area optical feature is obtained altogether Similarity data；According to p whole optical signature similarity data acquisition Parameters of Normal Distribution μ1With
S42.AkWith BkWith the identical digit/64*100% of binary representation number, p whole optical signature similarity is obtained altogether Data；According to p × q area optical characteristic similarity data acquisition Parameters of Normal Distribution μ2With
If S43. μ1>=90 and μ2>=80 andAndIt then can be concluded that oil painting A to be identified and oil painting authentic work B From same width oil painting, that is, determine the uniqueness of oil painting A to be identified.
In the present embodiment, wave band number m is 10, and light-source angle n is 4, p=40, q=100.
In the present embodiment, the pixel density of each image is 300ppi.
Obviously, the above embodiment of the present invention is only intended to clearly illustrate technical solution of the present invention example, and It is not the restriction to a specific embodiment of the invention.It is all made within the spirit and principle of claims of the present invention Any modifications, equivalent replacements, and improvements etc. should all be comprised within the scope of protection of the claims of the present invention.

Claims (7)

1. a kind of oil painting uniqueness discrimination method based under multispectral light source, which comprises the steps of:
S1. m × n width image of the oil painting A to be identified at multiband m, multiple light courcess angle n is obtained；
S2. the optical signature of m × n width image of oil painting A to be identified is extracted respectively；
S3. the optical signature of the oil painting authentic work B corresponding with oil painting A to be identified stored in oil painting authentication data library is obtained；
S4. according to the oil painting authentic work B corresponding with oil painting A to be identified stored in the optical signature of oil painting A to be identified and database Optical signature, the similarity of oil painting A to be identified Yu oil painting authentic work B are obtained, to identify the uniqueness of oil painting A to be identified.
2. a kind of oil painting uniqueness discrimination method based under multispectral light source according to claim 1, which is characterized in that The acquisition methods of m × n width image of oil painting A to be identified are as follows in step S1: in the dark place that can shield extraneous unknown light source, making Image of the oil painting A to be identified respectively under the multispectral light source of m wave band, n light-source angle is shot with multispectral camera, is obtained altogether Obtain m × n width image.
3. a kind of oil painting uniqueness discrimination method based under multispectral light source according to claim 2, which is characterized in that N light-source angle is respectively front, left side, right side and the back side of oil painting A to be identified.
4. a kind of oil painting uniqueness discrimination method based under multispectral light source according to claim 1, which is characterized in that Remember p=m × n, the optical signature of each image respectively includes area optical feature and whole optical signature, acquisition side in step S2 Method is as follows:
S21. each image is divided into the identical region of q area, and obtains the intensity histogram diagram data in each region respectively, Perception hash algorithm is used to obtain 64 hash values of each area grayscale histogram data respectively and as area optical feature； The area optical feature for remembering all width images is Ai,j, the integer that wherein i is 1 to p, the integer that j is 1 to q；
S22. 64 hash values and optical signature as a whole of each image are obtained respectively using perception hash algorithm；Note is all The whole optical signature of width image is Ak, wherein k be 1 to p integer.
5. a kind of oil painting uniqueness discrimination method based under multispectral light source according to claim 4, which is characterized in that The area optical feature for remembering oil painting authentic work B is Bi,j, whole optical signature is Bk, oil painting A and oil painting authentic work to be identified in step S4 The similarity of B includes area optical characteristic similarity and whole optical signature similarity, the uniqueness identification side of oil painting A to be identified Method is as follows:
S41.Ai,jWith Bi,jWith the identical digit/64*100% of binary representation number, p × q area optical feature is obtained altogether Similarity data；According to p whole optical signature similarity data acquisition Parameters of Normal Distribution μ1With
S42.AkWith BkWith the identical digit/64*100% of binary representation number, p whole optical signature similarity is obtained altogether Data；According to p × q area optical characteristic similarity data acquisition Parameters of Normal Distribution μ2With
If S43. μ1>=90 and μ2>=80 andAndIt then can be concluded that oil painting A to be identified and oil painting authentic work B comes Derived from same width oil painting, that is, determine the uniqueness of oil painting A to be identified.
6. a kind of oil painting uniqueness discrimination method based under multispectral light source according to claim 4, which is characterized in that Wave band number m is 10, and light-source angle n is 4, p=40, q=100.
7. a kind of oil painting uniqueness discrimination method based under multispectral light source according to claim 2, which is characterized in that The pixel density of each image is 300ppi.
CN201910791043.9A 2019-08-26 2019-08-26 A kind of oil painting uniqueness discrimination method based under multispectral light source Pending CN110490269A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910791043.9A CN110490269A (en) 2019-08-26 2019-08-26 A kind of oil painting uniqueness discrimination method based under multispectral light source

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910791043.9A CN110490269A (en) 2019-08-26 2019-08-26 A kind of oil painting uniqueness discrimination method based under multispectral light source

Publications (1)

Publication Number Publication Date
CN110490269A true CN110490269A (en) 2019-11-22

Family

ID=68554264

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910791043.9A Pending CN110490269A (en) 2019-08-26 2019-08-26 A kind of oil painting uniqueness discrimination method based under multispectral light source

Country Status (1)

Country Link
CN (1) CN110490269A (en)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103116755A (en) * 2013-01-27 2013-05-22 深圳市书圣艺术品防伪鉴定有限公司 Automatic painting and calligraphy authenticity degree detecting system and method thereof
CN103149164A (en) * 2013-03-04 2013-06-12 暨南大学 Traditional Chinese painting verification method and device based on spectral imaging technology
CN109635878A (en) * 2018-12-24 2019-04-16 山东环渤海艺术大数据科技有限公司 A kind of painting and calligraphy fidelity identification method and device

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103116755A (en) * 2013-01-27 2013-05-22 深圳市书圣艺术品防伪鉴定有限公司 Automatic painting and calligraphy authenticity degree detecting system and method thereof
CN103149164A (en) * 2013-03-04 2013-06-12 暨南大学 Traditional Chinese painting verification method and device based on spectral imaging technology
CN109635878A (en) * 2018-12-24 2019-04-16 山东环渤海艺术大数据科技有限公司 A kind of painting and calligraphy fidelity identification method and device

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
陈勇昌: "基于不变特征的数字水印与感知哈希图像认证技术研究", 《中国优秀博硕士学位论文全文数据库(博士) 信息科技辑》 *

Similar Documents

Publication Publication Date Title
Kasaei et al. New morphology-based method for robustiranian car plate detection and recognition
CN105842173B (en) A kind of EO-1 hyperion material discrimination method
US8577088B2 (en) Method and system for collecting information relating to identity parameters of a vehicle
CN106203415B (en) bank card number automatic identification device based on digital image processing
CN110008909B (en) Real-name system business real-time auditing system based on AI
CN105632015B (en) A kind of bank money fingerprint characteristic anti false authentication method based on cloud platform
Darade et al. Automatic recognition of fake Indian currency note
WO2019197628A1 (en) Method and system for optical product authentication
Prabhakar et al. A novel design for vehicle license plate detection and recognition
Li et al. An effective detection method based on physical traits of recaptured images on LCD screens
CN105741404B (en) The discrimination method of bill, device and system
CN105488486A (en) Face recognition method and device for preventing photo attack
CN112149690A (en) Tracing method and tracing system based on biological image feature recognition
Suresh et al. Indian currency recognition and verification using image processing
Khan et al. Car Number Plate Recognition (CNPR) system using multiple template matching
KR101781351B1 (en) Forgery passport inspection method and recording medium thereof
Lamsal COUNTERFIET PAPER BANKNOTE IDENTIFICATION BASED ON COLOR AND TEXTURE
CN110490269A (en) A kind of oil painting uniqueness discrimination method based under multispectral light source
CN111316305A (en) System and method for authenticating a consumer product
CN111739036B (en) Hyperspectrum-based file handwriting counterfeiting detection method
Bammey Analysis and experimentation on the ManTraNet image forgery detector
Ishiyama et al. Melon authentication by agri-biometrics-identifying individual fruits using a single image of rind pattern
Kanojia et al. Recognition and verification of Indian currency notes using digital image processing
Forero Banknotes classification system through image processing and pattern recognition for people with visual impairment
Swami et al. Currency recognition and fake currency identification using image processing

Legal Events

Date Code Title Description
PB01 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: 20191122