CN110222602A - Antiforge recognizing method, system, device end and computer readable storage medium - Google Patents
Antiforge recognizing method, system, device end and computer readable storage medium Download PDFInfo
- Publication number
- CN110222602A CN110222602A CN201910432872.8A CN201910432872A CN110222602A CN 110222602 A CN110222602 A CN 110222602A CN 201910432872 A CN201910432872 A CN 201910432872A CN 110222602 A CN110222602 A CN 110222602A
- Authority
- CN
- China
- Prior art keywords
- information
- eigenvector
- printing image
- image
- reference standard
- 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
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
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Editing Of Facsimile Originals (AREA)
Abstract
The present invention relates to a kind of Antiforge recognizing methods, system, device end and computer readable storage medium, above-mentioned Antiforge recognizing method includes the first printing image for obtaining the presumptive area shot by first terminal, feature extraction is carried out to the first printing image, obtain corresponding eigenvector information, eigenvector information includes the image edge information and image texture information of the first printing image, identification is carried out to obtain corresponding identity identification information to the first printing image, corresponding reference standard eigenvector information is obtained from eigenvector information data base querying according to identity identification information, calculate the similarity of corresponding eigenvector information Yu reference standard eigenvector information, and the true and false of product is determined according to similarity.Above-mentioned Antiforge recognizing method can utilize the true and false that product pattern prints the non-reproduction in generating process and randomness distinguishes product, greatly improve the anti-counterfeit capability of product.
Description
Technical field
The present invention relates to anti-counterfeit recognition field more particularly to a kind of Antiforge recognizing method, system, device end and computers
Readable storage medium storing program for executing.
Background technique
As intelligence manufacture continues to develop, intelligence, automation and the standardization level of commodity manufacture are continuously improved, manufacture
Level, which improves, also improves commodity forgery technology simultaneously, forges the difficulty for improving commodity true and false identification of technology, personation
Spreading unchecked for counterfeit or inferior quality goods is also easy to damage consumer.The anti-counterfeiting technology of non-reproduction is extremely urgent as a result,.
Planar bar code technology is also widely used by industrial circle, is used for the fields such as message identification and logistics the first day of the lunar month source, significant increase
The informationization and automatization level of these conventional industries processes.However the identification technologies such as two dimensional code, digital letter can only be solved
The carrying problem of breath, anti-counterfeit capability are to be improved.
Summary of the invention
In view of the above problems, it the invention proposes a kind of Antiforge recognizing method, system, device end and computer-readable deposits
Storage media can print the true and false that non-reproduction and randomness in generating process distinguish product, pole using product pattern
The big anti-counterfeit capability for improving product.
A kind of Antiforge recognizing method, Antiforge recognizing method include:
Obtain the first printing image of the presumptive area shot by first terminal;
Feature extraction is carried out to the first printing image, obtains corresponding eigenvector information, eigenvector information includes the
The image edge information and image texture information of one printing image;
Identification is carried out to obtain corresponding identity identification information to the first printing image;
Corresponding reference standard feature vector is obtained from eigenvector information data base querying according to identity identification information
Information;
The similarity of corresponding eigenvector information Yu reference standard eigenvector information is calculated, and is determined according to similarity
The true and false of product.
In one embodiment, eigenvector information database pre-establishes in the following way:
Obtain the second printing image of the presumptive area shot by second terminal;
Identification and feature extraction are carried out to the second printing image, obtain corresponding identity identification information and reference standard
Eigenvector information database is arrived in eigenvector information and associatedly storage.
In one embodiment, eigenvector information database pre-establishes in the following way:
It receives the identity identification information and reference standard eigenvector information that second terminal uploads and associatedly stores to spy
It levies vector information database, identity identification information and reference standard eigenvector information and acquisition presumptive area is shot by second terminal
The second printing image and identification and feature extraction carried out to the second printing image handle to obtain.
In one embodiment, feature extraction is carried out to the first printing image, obtains the step of corresponding eigenvector information
Suddenly include:
First printing image is corrected, change of scale and image enhancement, obtains pre-processed image information;
Pre-processed image information is handled according to default feature extraction operator, generates corresponding eigenvector information.
In one embodiment, the similarity of corresponding eigenvector information and reference standard eigenvector information is calculated
Before step further include:
Eigenvector information and reference standard eigenvector information are subjected to standard according to preset data canonical algorithm respectively
Change processing, preset data canonical algorithm use min-max standardized algorithm or Z-score standardized algorithm.
In addition, also providing a kind of Antiforge recognizing method, Antiforge recognizing method includes:
Obtain the first printing image of the presumptive area of first terminal shooting;
Feature extraction is carried out to the first printing image, obtains corresponding eigenvector information, eigenvector information includes the
The image edge information and image texture information of one printing image;
Identification is carried out to obtain corresponding identity identification information to the first printing image;
Corresponding reference standard feature vector is obtained from eigenvector information data base querying according to identity identification information
Information;
Corresponding eigenvector information and reference standard eigenvector information are sent to first terminal, so that first terminal
It calculates the similarity of corresponding eigenvector information and reference standard eigenvector information and the true of product is determined according to similarity
It is pseudo-.
In addition, also providing a kind of Antiforge recognizing method, Antiforge recognizing method includes:
The first printing image of presumptive area is shot, and feature extraction is carried out to the first printing image, obtains corresponding spy
Vector information is levied, eigenvector information includes the image edge information and image texture information of the first printing image;
To the first printing image identity identification to obtain corresponding identity identification information;
Eigenvector information and identity identification information are sent to server so that server according to identity identification information from
Eigenvector information data base querying obtains corresponding reference standard eigenvector information and calculates corresponding eigenvector information
The true and false of product is determined with the similarity of reference standard eigenvector information.
In addition, also providing a kind of Antiforge recognizing method, Antiforge recognizing method includes:
The first printing image of presumptive area is shot, and feature extraction is carried out to the first printing image, obtains corresponding spy
Vector information is levied, eigenvector information includes the image edge information and image texture information of the first printing image;
To the first printing image identity identification to obtain corresponding identity identification information;
Eigenvector information and identity identification information are sent to server so that server according to identity identification information from
Eigenvector information data base querying obtains corresponding reference standard eigenvector information;
The reference standard eigenvector information that server is sent is received, and calculates corresponding eigenvector information and reference mark
The similarity of quasi- eigenvector information further determines the true and false of product according to similarity.
In addition, also providing a kind of anti-counterfeit recognition system, anti-counterfeit recognition system includes: first terminal and server;
First terminal is used to upload the first printing image of the presumptive area of shooting;
Server is used to carry out feature extraction to the first printing image, obtains corresponding eigenvector information, feature vector
Information includes the image edge information and image texture information of the first printing image;
Server is also used to carry out identification to obtain corresponding identity identification information, according to body to the first printing image
Part identification information obtains corresponding reference standard eigenvector information from eigenvector information data base querying;
Server is also used to calculate the similarity of corresponding eigenvector information Yu reference standard eigenvector information, and root
The true and false of product is determined according to similarity.
In one embodiment, anti-counterfeit recognition system further include:
Second terminal, for shooting the second printing image of presumptive area and being uploaded to server;
Server carries out identification and feature extraction to the second printing image, obtains corresponding identity identification information and ginseng
It examines standard feature vector information and associatedly stores to eigenvector information database.
In addition, also providing a kind of device end, including memory and processor, memory is for storing computer journey
Sequence, processor runs computer program so that device end executes above-mentioned Antiforge recognizing method.
A kind of computer readable storage medium is stored with computer program used in device end.
Above-mentioned Antiforge recognizing method, it is right by obtaining the first printing image of the presumptive area shot by first terminal
First printing image carries out feature extraction, obtains corresponding eigenvector information, and eigenvector information includes the first printing image
Image edge information and image texture information, to first printing image carry out identification to obtain corresponding identification
Information obtains corresponding reference standard feature vector from eigenvector information data base querying according to identity identification information and believes
Breath, calculates the similarity of corresponding eigenvector information Yu reference standard eigenvector information, and determine product according to similarity
The true and false, utilize the corresponding image edge information of printing image and image texture information of the presumptive area of first terminal shooting
Unique and non-reproduction, the corresponding identity identification information of combination product believe above-mentioned image edge information and image texture
It ceases corresponding eigenvector information to be compared with reference standard eigenvector information, further calculates between the two similar
Degree, the corresponding true and false information of product is finally judged according to above-mentioned similarity, does not need to do original product line larger
Variation, and the corresponding anti-counterfeit capability of product is greatly improved with lesser cost price.
Detailed description of the invention
In order to illustrate more clearly of technical solution of the present invention, letter will be made to attached drawing needed in the embodiment below
It singly introduces, it should be understood that the following drawings illustrates only certain embodiments of the present invention, therefore is not construed as to the present invention
The restriction of protection scope.In various figures, part is similarly comprised using similar number.
Fig. 1 is the applied environment figure of Antiforge recognizing method in one embodiment;
Fig. 2 is a kind of flow diagram of Antiforge recognizing method in one embodiment;
Fig. 3 (a) and Fig. 3 (b) is respectively two corresponding two dimensions of product that first terminal is shot in one embodiment
Code printing image schematic diagram;
Fig. 4 (a) and Fig. 4 (b) is respectively two corresponding two dimensions of product that first terminal is shot in one embodiment
Code printing image local sideline schematic diagram;
Fig. 5 (a) and Fig. 5 (b) be respectively in one embodiment eigenvector information and reference standard eigenvector information it is each
Self-corresponding curve synoptic diagram;
Fig. 6 is the flow diagram of eigenvector information database building method in one embodiment;
Fig. 7 is to obtain the method flow schematic diagram of corresponding eigenvector information in another embodiment;
Fig. 8 is a kind of flow diagram of Antiforge recognizing method in another embodiment;
Fig. 9 is a kind of flow diagram of Antiforge recognizing method in another embodiment;
Figure 10 is a kind of flow diagram of Antiforge recognizing method in another embodiment;
Figure 11 is a kind of flow diagram of Antiforge recognizing method in another embodiment;
Figure 12 is a kind of structural block diagram of anti-counterfeit recognition system in one embodiment.
Specific embodiment
Hereinafter, term " includes ", " having " and its cognate that can be used in various embodiments of the present invention are only
It is intended to mean that special characteristic, number, step, operation, the combination of element, component or aforementioned item, and is understood not to first
Exclude the combined presence or increase by one of one or more other features, number, step, operation, element, component or aforementioned item
A or more feature, number, step, operation, element, component or aforementioned item combination a possibility that.
Hereinafter, the various embodiments of the disclosure will be described more fully.The disclosure can have various embodiments, and
It can adjust and change wherein.It should be understood, however, that: there is no the various embodiments of the disclosure are limited to spy disclosed herein
Determine the intention of embodiment, but the disclosure should be interpreted as in the spirit and scope for covering the various embodiments for falling into the disclosure
All adjustment, equivalent and/or optinal plan.
Hereinafter, can the term " includes " used in the various embodiments of the disclosure or " may include " instruction disclosed in
Function, operation or the presence of element, and do not limit the increase of one or more functions, operation or element.In addition, such as existing
Used in the various embodiments of the disclosure, term " includes ", " having " and its cognate are meant only to indicate special characteristic, number
Word, step, operation, the combination of element, component or aforementioned item, and be understood not to exclude first one or more other
Feature, number, step, operation, element, component or aforementioned item combined presence or increase one or more features, number,
Step, operation, element, component or aforementioned item combination a possibility that.
In the various embodiments of the disclosure, at least one of A or/and B are stated " it include appointing for the text listed file names with
What combination or all combinations.For example, statement " A or B " or " at least one of A or/and B " may include A, may include B or can wrap
Include A and B both.
The term used in the various embodiments of the disclosure is used only for the purpose of describing specific embodiments and not anticipates
In the various embodiments of the limitation disclosure.As used herein, singular is intended to also include plural form, unless context is clear
Chu it is indicated otherwise.Unless otherwise defined, otherwise all terms (including technical terms and scientific terms) used herein have
There is meaning identical with the various normally understood meanings of embodiment one skilled in the art of the disclosure.The term
(term such as limited in the dictionary generally used) is to be interpreted as having and situational meaning in the related technical field
Identical meaning and it will be interpreted as having Utopian meaning or meaning too formal, unless in the various of the disclosure
It is clearly defined in embodiment.
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to the accompanying drawings and embodiments, right
The present invention is further elaborated.
Fig. 1 is the applied environment figure of Antiforge recognizing method in one embodiment, including first terminal 100 and server 200,
First terminal 100 and server 200 can be communicated by network, and wherein first terminal 100 includes having shooting function
Mobile phone, tablet computer or individual PC.
As shown in Fig. 2, a kind of Antiforge recognizing method, the Antiforge recognizing method include:
Step S110 obtains the first printing image of the presumptive area shot by first terminal.
Wherein, the product of plant produced is usually provided with corresponding product printing image, such as one-dimension code, two dimensional code, word
Symbol and pattern, are used to carry out product identity, printing image generation be all before factory setting in corresponding product or
In product packaging, the process that product identification mark generates is commonly referred to as endowed process, the process there are two types of implementation, one is
Will the directly logical spray printing of printing image in product designated position, another kind is will to print image elder generation spray printing on adhesive label then
Label is directly being attached in product or product packaging.
First terminal has camera function, when user shoots using first terminal the printing image of the said goods presumptive area
When, available corresponding first printing image is simultaneously further sent to server.
Step S120 carries out feature extraction to the first printing image, obtains corresponding eigenvector information, feature vector letter
Breath includes the image edge information and image texture information of the first printing image.
Wherein, server, can be into one after obtaining the corresponding printing image of presumptive area shot by first terminal
Step carries out feature extraction to above-mentioned printing image, wherein generate in the printing process of printing image, for microcosmic angle, oil
The depth of the ink on each position will necessarily variant and completely random do not controlled by human factor, and ink jet-printing to produce
Its marginal position will necessarily generate STOCHASTIC DIFFUSION during air-drying on product, even unlike material or identical material are not
Same position, it is all random and irreproducible that ink dot, which spreads size, thus prints shape corresponding to the image border of image
And location information is all unique and irreproducible, and dynamics when due to spray printing is different, caused by the thickness of ink
Shade it is significantly different, cause corresponding image texture information different, thus feature can be carried out to above-mentioned printing image
It extracts, obtains corresponding eigenvector information, this feature vector information includes image edge information and image texture information.
In one embodiment, corresponding printing image is two dimensional code on product, is printed by mobile terminal to the two dimensional code
Map brushing picture is taken pictures, and corresponding printing image is obtained, and wherein the resolution ratio of mobile terminal is 800 resolutions very much, passes through shooting
High-resolution image can lay the foundation for the precision of feature extraction and identification of subsequent printing image.
As shown in Fig. 3 (a) and Fig. 3 (b), the respective printing image of identical product A and product B is taken pictures and can be obtained respectively
Corresponding two dimensional code printing image graph 3 (a) and Fig. 3 (b).
Further, for example, respectively to two dimensional code printing image graph 3 (a) a certain office corresponding with angle upper on the right side of Fig. 3 (b)
The printing image in portion carries out the feature extraction of marginal edge line profile, and available corresponding image such as Fig. 4 (a) and Fig. 4 (b) are shown.
Wherein, during carrying out feature extraction, common information between the two and respective background can be removed
Information retains differentiation information between the two, to guarantee the accuracy of feature extraction.
Although seeing on the whole, two dimensional code prints image graph 3 (a) and Fig. 3 (b) is roughly the same, due to each product
Ink jet-printing amount, the expansion rate of ink and expanding location are different in corresponding image in 2 D code jet printing process, lead to reality
Corresponding sideline chamfered shape is significantly different, takes the respective upper right corner above-mentioned two dimensional code printing image graph 3 (a) and Fig. 3 (b) respectively
Local edge line drawing figure, as shown in Fig. 4 (a) and Fig. 4 (b), the sideline chamfered shape that respectively indicates, it is clear that Fig. 4 (a) and Fig. 4 (b)
Respective chamfered shape has significant difference, thus can further extract Fig. 4 (a) and the respective chamfered shape pair of Fig. 4 (b) respectively
Answer feature;Further, since ink jet-printing amount thickness is different, lead to two dimensional code printing image graph 3 (a) and the respective entirety of Fig. 3 (b)
It is of different shades, can also further extract two dimensional code printing image graph 3 (a) and the respective whole texture information of Fig. 3 (b), most
After obtain corresponding eigenvector information.
Step S130 carries out identification to the first printing image to obtain corresponding identity identification information.
Wherein, the first printing image contains the corresponding identification information of product, and server is obtaining first terminal hair
After the first printing image sent, the corresponding body of product can be obtained by carrying out identification to the first printing image
Part identification information.
For example, identification information may include product coding or electronic supervision code etc..
Step S140 obtains corresponding reference standard from eigenvector information data base querying according to identity identification information
Eigenvector information.
Wherein, server is previously provided with eigenvector information database, and this feature vector information database goes out with product
It is arranged that corresponding printing image is corresponding when factory on product or product packaging, due to the corresponding printing figure being arranged when factory
As containing the identification information of product, thus in this feature vector information database, the identity of each actual products is believed
Breath has corresponding extracted eigenvector information.
Server carries out identification by the first printing image and obtains corresponding identity identification information, and further basis should
Identity identification information is inquired from pre-set characteristic vector data library and further obtains corresponding reference standard feature
Vector information.
Step S150, calculates the similarity of corresponding eigenvector information Yu reference standard eigenvector information, and according to
Similarity determines the true and false of product.
Server, further will be right after obtaining eigenvector information corresponding to the first printing image of presumptive area
The eigenvector information answered is compared with reference standard eigenvector information, similarity between the two is calculated, according to similar
The size of degree determines the true and false of product.
Above-mentioned Antiforge recognizing method, the corresponding image border of printing image using the presumptive area of first terminal shooting are believed
The uniqueness and non-reproduction of breath and image texture information, the corresponding identity identification information of combination product, by above-mentioned image side
Edge information and the corresponding eigenvector information of image texture information are compared with reference standard eigenvector information, are further counted
Similarity between the two is calculated, the corresponding true and false information of product is finally judged according to above-mentioned similarity, is not needed to original
Product line does biggish variation, and is greatly improved the corresponding anti-counterfeit capability of product with lesser cost price.
In one embodiment, the threshold value of above-mentioned similarity can be set, such as similarity threshold is set as
0.85, it, can if it is 0.27 that server, which calculates corresponding eigenvector information and the similarity of reference standard eigenvector information,
Directly determine that first prints image to forge image, and then judges the corresponding product of the first printing image for adulterant.
For example, Fig. 5 (a) and Fig. 5 (b) respectively represent eigenvector information and mark with reference as shown in Fig. 5 (a) and Fig. 5 (b)
Quasi- eigenvector information, the two similarity are 0.2708, it is clear that can product representated by process decision chart 5 (a) directly be directly pseudo-
Product.
Wherein, the above-mentioned similarity value calculated is smaller, shows that difference is bigger between the two, shows the first printing image
It is bigger a possibility that adulterant when corresponding product.
Wherein, the setting of similarity threshold can reasonably be chosen according to big data algorithm, and then can set conjunction
The similarity threshold of reason further increases the accuracy of above-mentioned deterministic process.
In one embodiment, as shown in fig. 6, features described above vector information database is established in the following way:
Step S162 obtains the second printing image of the presumptive area shot by second terminal.
Wherein, second terminal is usually provided with camera shooting acquisition device, and the PC with photographic device can also be used in second terminal
Terminal or tablet computer etc., camera shooting acquisition device generally use satisfactory industrial photographic device.
Wherein, to reduce cost, corresponding second can be installed in each product line in follow-up process additional before factory
Terminal acquires the process of the second printing image of presumptive area, for the ease of the second printing figure of second terminal acquisition presumptive area
Picture can directly can be with to the product to be dispatched from the factory on each production line by second terminal by needing to match with sensor
Second printing image of acquisition presumptive area is simultaneously further uploaded to server.
Step S164 carries out identification and feature extraction to the second printing image, obtains corresponding identity identification information
It stores with reference standard eigenvector information and associatedly and arrives eigenvector information database.
Similarly, in the second printing image corresponding to the setting of each product, the second printing image generally comprises factory
The corresponding identity identification information of product;Further, since the depth of the ink on each position will necessarily be variant, and completely with
Machine is not controlled by human factor, its marginal position can also generate STOCHASTIC DIFFUSION during air-drying on ink jet-printing to product, no
Even the different parts of same material or identical material, ink dot diffusion size be all it is random and irreproducible, thus
Shape corresponding to printing image border corresponding to each product and location information are unique and irreproducible, and
Dynamics when spray printing is different, and image texture information caused by the thickness of ink is different, thus server is to the second printing image
Identification and feature extraction, available corresponding identity identification information and reference mark eigenvector information are carried out, is gone forward side by side
One step is by identity identification information and reference standard eigenvector information associated storage to eigenvector information database.
Wherein, server to second printing image carry out feature extraction during, can remove common information and
Respective background information retains apparent differentiation information, to guarantee the accuracy of feature extraction.
When presetting eigenvector information database, this feature vector information database and product export by server
It is arranged that corresponding printing image is corresponding on product or product packaging, the identification judgement for the true and false of subsequent product is established
Basis.
In one embodiment, features described above vector information database can also be established in the following way:
Server receives the identity identification information that second terminal uploads and reference standard eigenvector information and associatedly deposits
Eigenvector information database, identity identification information and reference standard eigenvector information is stored up to be obtained in advance by second terminal shooting
Determine the second printing image in region and the second printing image progress identification and feature extraction are handled to obtain.
Wherein, the method for building up due to the limitation of current network bandwidth and speed, with characteristic vector data library shown in fig. 6
It compares, the corresponding identity identification information of the second printing image and reference standard eigenvector information can be located by second terminal
Reason, it is usually necessary to use the computers for having photographic device and processor for second terminal at this time, as identity identification information and reference
The respective identification of standard feature vector information and characteristic extraction procedure are still identical as the process of server end.
In one embodiment, as shown in fig. 7, step S120 includes:
First printing image is corrected, change of scale and image enhancement, obtains pretreatment image letter by step S122
Breath.
Wherein, first terminal obtains in the process of the first printing image since image taking or deformation of products can bring figure
Distortion of image problem needs to carry out image image distortion correction to improve discrimination.
Further, image of the above-mentioned distortion correction after processed can also be transformed to multiple and different scales, passed through
The minutia of analysis picture under different scales guarantees that the eigenvector information extracted is reliable and stable.
Further, can by time domain and frequency domain both direction to image enhancement processing, to strengthen detail textures information
And strengthen colour fading detailed information, and then improve detail textures feature identification degree while inhibiting to be likely to occur when image taking bright
The uneven influence to subsequent extracted eigenvector information process of degree.
It can also strengthen in the image after above-mentioned process than shallower image detail, and then improve feature difference
Change identification.
First printing image is corrected, change of scale and image enhancement, obtains pre-processed image information, is subsequent
Eigenvector information extraction lays the foundation.
Step S124 handles pre-processed image information according to default feature extraction operator, generates corresponding feature
Vector information.
It for pre-processed image information, generally requires to remove common information therein and background information, retain apparent
Differentiation information, to guarantee the accuracy of feature extraction.
Wherein, presetting feature extraction operator can be used common extraction algorithm, such as SIFT (Scale invariant features transform),
LBP operator (local binary patterns) and HOG (histograms of oriented gradients) method, can also be used other traditional extraction algorithms, example
Such as SURF, ORB or HAAR algorithm.
Wherein, for generally chosen in the extraction process of pre-processed image information corresponding edge contour information in image,
The location information of edge pixel point and texture information in image are as emphasis.
In one embodiment, as shown in figure 8, step S150 includes:
Step S170, respectively by eigenvector information and reference standard eigenvector information according to preset data canonical algorithm
It is standardized, the preset data canonical algorithm is calculated using min-max standardized algorithm or Z-score standardization
Method.
Wherein, the similarity of eigenvector information and reference standard eigenvector information between the two is calculated in step S150
It also needs respectively to mark eigenvector information and reference standard eigenvector information according to preset data canonical algorithm before
Quasi-ization processing, in order to subsequent calculating process and comparison procedure.
Wherein, preset data canonical algorithm generallys use min-max standardized algorithm or Z-score standardized algorithm.
In addition, Antiforge recognizing method includes: as shown in figure 9, also providing a kind of Antiforge recognizing method
Step S210 obtains the first printing image of the presumptive area of first terminal shooting.
Step S220 carries out feature extraction to the first printing image, obtains corresponding eigenvector information, feature vector letter
Breath includes the image edge information and image texture information of the first printing image.
Step S230 carries out identification to the first printing image to obtain corresponding identity identification information.
Step S240 obtains corresponding reference standard from eigenvector information data base querying according to identity identification information
Eigenvector information.
Corresponding eigenvector information and reference standard eigenvector information are sent to first terminal by step S250, with
First terminal is set to calculate the similarity of corresponding eigenvector information and reference standard eigenvector information and true according to similarity
The true and false of fixed output quota product.
Wherein, above-mentioned Antiforge recognizing method shown in Fig. 9 is compared with Antiforge recognizing method shown in Fig. 2, only step
S250 is different from step S150, this is because eigenvector information and reference standard eigenvector information are sent to the by server
One terminal, the calculating and judgement of similarity are carried out by first terminal, and first terminal carries out the mistake of calculating and the judgement of similarity
Journey is substantially identical as the process of server end.
In addition, as shown in Figure 10, also providing a kind of Antiforge recognizing method, Antiforge recognizing method includes:
Step S310 shoots the first printing image of presumptive area, and carries out feature extraction to the first printing image, obtains
Corresponding eigenvector information, eigenvector information include the image edge information and image texture letter of the first printing image
Breath.
Step S320, to the first printing image identity identification to obtain corresponding identity identification information.
Eigenvector information and identity identification information are sent to server by step S330, so that server is according to identity
Identification information obtains corresponding reference standard eigenvector information from eigenvector information data base querying and calculates corresponding spy
The similarity of vector information and reference standard eigenvector information is levied to determine the true and false of product.
Wherein, compared with Antiforge recognizing method shown in Fig. 2, the first terminal in Figure 10 in step S310 is predetermined in shooting
Further progress feature extraction after the first printing image in region, obtains corresponding eigenvector information, then in step S320
First terminal further to first printing image carry out identification to obtain corresponding identity identification information, wherein Figure 10
In identification procedure and characteristic extraction procedure first terminal complete, then directly upload corresponding eigenvector information
With identity identification information to server.
In addition, as shown in figure 11, also providing a kind of Antiforge recognizing method, Antiforge recognizing method includes:
Step S410, first terminal shoots the first printing image of presumptive area, and carries out feature to the first printing image
It extracts, obtains corresponding eigenvector information, eigenvector information includes the image edge information and figure of the first printing image
As texture information.
Step S420, to the first printing image identity identification to obtain corresponding identity identification information.
Eigenvector information and identity identification information are sent to server by step S430, so that server is according to identity
Identification information obtains corresponding reference standard eigenvector information from eigenvector information data base querying.
Step S440 receives the reference standard eigenvector information that server is sent, and calculates corresponding feature vector letter
The similarity of breath and reference standard eigenvector information, further determines the true and false of product according to similarity.
Wherein, compared with Antiforge recognizing method method shown in Fig. 2, by identification, feature extraction and phase in Figure 11
It is carried out in first terminal like the process of calculating and the judgement of degree.
In addition, as shown in figure 12, also providing a kind of anti-counterfeit recognition system 300, anti-counterfeit recognition system includes: first terminal
301 and server 302;
First terminal 301 is used to upload the first printing image of the presumptive area of shooting;
Server 302 is used to carry out feature extraction to the first printing image, obtains corresponding eigenvector information, feature to
Amount information includes the image edge information and image texture information of the first printing image;
Server 302 is also used to carry out identification to the first printing image to obtain corresponding identity identification information, root
Corresponding reference standard eigenvector information is obtained from eigenvector information data base querying according to identity identification information;
Server 302 is also used to calculate the similarity of corresponding eigenvector information Yu reference standard eigenvector information,
And the true and false of product is determined according to similarity.
In one embodiment, above-mentioned anti-counterfeit recognition system further include:
Second terminal, for shooting the second printing image of presumptive area and being uploaded to server 302;
Server 302 carries out identification and feature extraction to the second printing image, obtains corresponding identity identification information
It stores with reference standard eigenvector information and associatedly and arrives eigenvector information database.
Wherein, the process of feature extraction and identification can both be completed in first terminal 301, can also be in server
302 ends are completed, and similarly, the calculating of similarity and deterministic process can both be completed in first terminal 301, can also be in server
302 ends are completed, and treatment process is corresponding.
In addition, also providing a kind of device end, including memory and processor, memory is for storing computer journey
Sequence, processor runs computer program so that device end executes above-mentioned Antiforge recognizing method.
A kind of computer readable storage medium is stored with computer program used in device end.
In several embodiments provided herein, it should be understood that disclosed device and method can also pass through
Other modes are realized.The apparatus embodiments described above are merely exemplary, for example, flow chart and structure in attached drawing
Figure shows the system frame in the cards of the device of multiple embodiments according to the present invention, method and computer program product
Structure, function and operation.In this regard, each box in flowchart or block diagram can represent a module, section or code
A part, a part of the module, section or code includes one or more for implementing the specified logical function
Executable instruction.It should also be noted that function marked in the box can also be to be different from the implementation as replacement
The sequence marked in attached drawing occurs.For example, two continuous boxes can actually be basically executed in parallel, they are sometimes
It can execute in the opposite order, this depends on the function involved.It is also noted that in structure chart and/or flow chart
The combination of each box and the box in structure chart and/or flow chart, can function or movement as defined in executing it is dedicated
Hardware based system realize, or can realize using a combination of dedicated hardware and computer instructions.
In addition, each functional module or unit in each embodiment of the present invention can integrate one independence of formation together
Part, be also possible to modules individualism, an independent part can also be integrated to form with two or more modules.
It, can be with if the function is realized and when sold or used as an independent product in the form of software function module
It is stored in a computer readable storage medium.Based on this understanding, technical solution of the present invention is substantially in other words
The part of the part that contributes to existing technology or the technical solution can be embodied in the form of software products, the meter
Calculation machine software product is stored in a storage medium, including some instructions are used so that a computer equipment (can be intelligence
Can mobile phone, personal computer, server or network equipment etc.) execute each embodiment the method for the present invention whole or
Part steps.And storage medium above-mentioned include: USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only Memory),
Random access memory (RAM, Random Access Memory), magnetic or disk etc. be various to can store program code
Medium.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, can easily think of the change or the replacement, and should all contain
Lid is within protection scope of the present invention.
Claims (12)
1. a kind of Antiforge recognizing method, which is characterized in that the Antiforge recognizing method includes:
Obtain the first printing image of the presumptive area shot by first terminal;
Feature extraction is carried out to the first printing image, obtains corresponding eigenvector information, described eigenvector packet
Include the image edge information and image texture information of the first printing image;
Identification is carried out to obtain corresponding identity identification information to the first printing image;
Corresponding reference standard feature vector is obtained from eigenvector information data base querying according to the identity identification information
Information;
The similarity of the corresponding eigenvector information Yu the reference standard eigenvector information is calculated, and according to the phase
Like the true and false for spending determining product.
2. Antiforge recognizing method according to claim 1, which is characterized in that described eigenvector information database is by such as
Under type pre-establishes:
Obtain the second printing image of the presumptive area shot by second terminal;
Identification and feature extraction are carried out to the second printing image, obtain corresponding identity identification information and reference standard feature
Eigenvector information database is arrived in vector information and associatedly storage.
3. Antiforge recognizing method according to claim 1, which is characterized in that described eigenvector information database is by such as
Under type pre-establishes:
Receive second terminal upload identity identification information and reference standard eigenvector information and associatedly storage arrive feature to
Information database, the identity identification information and reference standard eigenvector information is measured to be made a reservation for by second terminal shooting
The second printing image in region simultaneously handles to obtain to the second printing image progress identification and feature extraction.
4. Antiforge recognizing method according to claim 1, which is characterized in that described to carry out spy to the first printing image
Sign is extracted, and the step of obtaining corresponding eigenvector information includes:
The first printing image is corrected, change of scale and image enhancement, obtains pre-processed image information;
The pre-processed image information is handled according to default feature extraction operator, generates corresponding eigenvector information.
5. Antiforge recognizing method according to claim 1, which is characterized in that described to calculate the corresponding feature vector letter
Before the step of similarity of breath and the reference standard eigenvector information further include:
Described eigenvector information and the reference standard eigenvector information are carried out according to preset data canonical algorithm respectively
Standardization, the preset data canonical algorithm use min-max standardized algorithm or Z-score standardized algorithm.
6. a kind of Antiforge recognizing method, which is characterized in that the Antiforge recognizing method includes:
Obtain the first printing image of the presumptive area of first terminal shooting;
Feature extraction is carried out to the first printing image, obtains corresponding eigenvector information, described eigenvector packet
Include the image edge information and image texture information of the first printing image;
Identification is carried out to obtain corresponding identity identification information to the first printing image;
Corresponding reference standard feature vector is obtained from eigenvector information data base querying according to the identity identification information
Information;
The corresponding eigenvector information and the reference standard eigenvector information are sent to the first terminal, so that
The first terminal calculates the similarity and root of the corresponding eigenvector information and the reference standard eigenvector information
The true and false of product is determined according to the similarity.
7. a kind of Antiforge recognizing method, which is characterized in that the Antiforge recognizing method includes:
The first printing image of presumptive area is shot, and carries out feature extraction to the first printing image is stated, obtains corresponding feature
Vector information, described eigenvector information include the image edge information and image texture information of the first printing image;
To the first printing image identity identification to obtain corresponding identity identification information;
Described eigenvector information and the identity identification information are sent to server, so that the server is according to the body
Part identification information from eigenvector information data base querying obtain corresponding reference standard eigenvector information and calculate it is described right
The similarity of the eigenvector information answered and the reference standard eigenvector information is to determine the true and false of product.
8. a kind of Antiforge recognizing method, which is characterized in that the Antiforge recognizing method includes:
The first printing image of presumptive area is shot, and carries out feature extraction to the first printing image is stated, obtains corresponding feature
Vector information, described eigenvector information include the image edge information and image texture information of the first printing image;
To the first printing image identity identification to obtain corresponding identity identification information;
Described eigenvector information and the identity identification information are sent to server, so that the server is according to the body
Part identification information obtains corresponding reference standard eigenvector information from eigenvector information data base querying;
The reference standard eigenvector information that the server is sent is received, and calculates the corresponding eigenvector information and institute
The similarity for stating reference standard eigenvector information determines the true and false of product according to the similarity.
9. a kind of anti-counterfeit recognition system, which is characterized in that the anti-counterfeit recognition system includes: first terminal and server;
The first terminal is used to upload the first printing image of the presumptive area of shooting;
The server is used to carry out feature extraction to the first printing image, obtains corresponding eigenvector information, described
Eigenvector information includes the image edge information and image texture information of the first printing image;
The server is also used to carry out identification to the first printing image to obtain corresponding identity identification information, root
Corresponding reference standard eigenvector information is obtained from eigenvector information data base querying according to the identity identification information;
The server is also used to calculate the phase of the corresponding eigenvector information with the reference standard eigenvector information
Like degree, and determine according to the similarity true and false of product.
10. anti-counterfeit recognition system according to claim 9, the anti-counterfeit recognition system further include:
Second terminal, for shooting the second printing image of presumptive area and being uploaded to the server;
The server is also used to carry out identification and feature extraction to the second printing image, obtains corresponding identification letter
Eigenvector information database is arrived in breath and reference standard eigenvector information and associatedly storage.
11. a kind of device end, which is characterized in that including memory and processor, the memory is for storing computer
Program, the processor runs the computer program so that the device end perform claim requires described in any one of 1 to 8
Antiforge recognizing method.
12. a kind of computer readable storage medium, which is characterized in that it is stored with device end described in claim 11 and is used
The computer program.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910432872.8A CN110222602A (en) | 2019-05-23 | 2019-05-23 | Antiforge recognizing method, system, device end and computer readable storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910432872.8A CN110222602A (en) | 2019-05-23 | 2019-05-23 | Antiforge recognizing method, system, device end and computer readable storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110222602A true CN110222602A (en) | 2019-09-10 |
Family
ID=67818195
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910432872.8A Pending CN110222602A (en) | 2019-05-23 | 2019-05-23 | Antiforge recognizing method, system, device end and computer readable storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110222602A (en) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110879968A (en) * | 2019-10-18 | 2020-03-13 | 中国农业大学 | Anti-counterfeiting identification method and device for fruits and vegetables |
CN111461102A (en) * | 2020-04-22 | 2020-07-28 | 艾科芯(深圳)智能科技有限公司 | Anti-counterfeiting identification method, device, equipment terminal and readable storage medium |
CN111523605A (en) * | 2020-04-28 | 2020-08-11 | 新疆维吾尔自治区烟草公司 | Image identification method and device, electronic equipment and medium |
CN111932282A (en) * | 2020-09-22 | 2020-11-13 | 北京大鱼梦想科技有限公司 | Anti-counterfeiting detection method and device |
CN112766404A (en) * | 2021-01-29 | 2021-05-07 | 安徽工大信息技术有限公司 | Chinese mitten crab authenticity identification method and system based on deep learning |
CN113656631A (en) * | 2021-08-20 | 2021-11-16 | 上海烟草集团有限责任公司 | Method for generating product identity, method, device and equipment for identifying identity |
CN113888198A (en) * | 2021-12-07 | 2022-01-04 | 北京微点科技有限公司 | Anti-counterfeiting method based on anti-counterfeiting characteristic correction |
CN114218423A (en) * | 2022-02-21 | 2022-03-22 | 广东联邦家私集团有限公司 | 5G-based non-labeling solid wood board identity digitalization method, device and system |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101748659A (en) * | 2009-12-18 | 2010-06-23 | 兆日科技(深圳)有限公司 | Random distribution fiber and substance anti-counterfeiting method |
CN102982606A (en) * | 2011-09-07 | 2013-03-20 | 深圳兆日科技股份有限公司 | Anti-fake method using physical feature recognition and anti-fake system using physical feature recognition |
CN103177452A (en) * | 2013-04-19 | 2013-06-26 | 海南大学 | Intelligent texture anti-counterfeiting method based on DWT-DCT (Dreamweaver Template-Discrete Cosine Transform) transformation |
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 |
-
2019
- 2019-05-23 CN CN201910432872.8A patent/CN110222602A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101748659A (en) * | 2009-12-18 | 2010-06-23 | 兆日科技(深圳)有限公司 | Random distribution fiber and substance anti-counterfeiting method |
CN102982606A (en) * | 2011-09-07 | 2013-03-20 | 深圳兆日科技股份有限公司 | Anti-fake method using physical feature recognition and anti-fake system using physical feature recognition |
CN103177452A (en) * | 2013-04-19 | 2013-06-26 | 海南大学 | Intelligent texture anti-counterfeiting method based on DWT-DCT (Dreamweaver Template-Discrete Cosine Transform) transformation |
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 |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110879968A (en) * | 2019-10-18 | 2020-03-13 | 中国农业大学 | Anti-counterfeiting identification method and device for fruits and vegetables |
CN111461102A (en) * | 2020-04-22 | 2020-07-28 | 艾科芯(深圳)智能科技有限公司 | Anti-counterfeiting identification method, device, equipment terminal and readable storage medium |
CN111523605A (en) * | 2020-04-28 | 2020-08-11 | 新疆维吾尔自治区烟草公司 | Image identification method and device, electronic equipment and medium |
CN111932282A (en) * | 2020-09-22 | 2020-11-13 | 北京大鱼梦想科技有限公司 | Anti-counterfeiting detection method and device |
CN112766404A (en) * | 2021-01-29 | 2021-05-07 | 安徽工大信息技术有限公司 | Chinese mitten crab authenticity identification method and system based on deep learning |
CN113656631A (en) * | 2021-08-20 | 2021-11-16 | 上海烟草集团有限责任公司 | Method for generating product identity, method, device and equipment for identifying identity |
CN113888198A (en) * | 2021-12-07 | 2022-01-04 | 北京微点科技有限公司 | Anti-counterfeiting method based on anti-counterfeiting characteristic correction |
CN114218423A (en) * | 2022-02-21 | 2022-03-22 | 广东联邦家私集团有限公司 | 5G-based non-labeling solid wood board identity digitalization method, device and system |
CN114218423B (en) * | 2022-02-21 | 2022-05-20 | 广东联邦家私集团有限公司 | 5G-based non-labeling solid wood board identity digitalization method, device and system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110222602A (en) | Antiforge recognizing method, system, device end and computer readable storage medium | |
CN109657595B (en) | Key feature region matching face recognition method based on stacked hourglass network | |
CN103914858B (en) | Document Image Compression Method And Its Application In Document Authentication | |
CN110427972B (en) | Certificate video feature extraction method and device, computer equipment and storage medium | |
CN108701234A (en) | Licence plate recognition method and cloud system | |
CN110163152A (en) | Antiforge recognizing method, method for anti-counterfeit, system, device end and storage medium | |
US20110249897A1 (en) | Character recognition | |
JP6708981B2 (en) | Individual identifier extraction device | |
CN110298353B (en) | Character recognition method and system | |
EP3252669B1 (en) | Method and device for classifying scanned documents | |
CN108734520B (en) | Jade price evaluation method and device based on machine learning | |
US20190287266A1 (en) | Individual identifying device | |
CN111461102A (en) | Anti-counterfeiting identification method, device, equipment terminal and readable storage medium | |
WO2014162168A1 (en) | System and method for describing image outlines | |
WO2015158784A1 (en) | Pattern recognition system | |
CN110533704B (en) | Method, device, equipment and medium for identifying and verifying ink label | |
EP2748754B1 (en) | Forensic authentication system and method | |
US9058517B1 (en) | Pattern recognition system and method using Gabor functions | |
CN114648771A (en) | Character recognition method, electronic device and computer readable storage medium | |
CN113256644A (en) | Bill image segmentation method, device, medium, and apparatus | |
CN113421257B (en) | Method and device for correcting rotation of text lines of dot matrix fonts | |
CN110415424B (en) | Anti-counterfeiting identification method and device, computer equipment and storage medium | |
CN113537216B (en) | Dot matrix font text line inclination correction method and device | |
CN109325489A (en) | The recognition methods of image and device, storage medium, electronic device | |
CN113850100B (en) | Method for correcting two-dimensional code and electronic equipment |
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 |