CN100538749C - Extract and the method for anti-counterfeit that uses the random distribution fiber characteristics of image - Google Patents
Extract and the method for anti-counterfeit that uses the random distribution fiber characteristics of image Download PDFInfo
- Publication number
- CN100538749C CN100538749C CNB2005101214371A CN200510121437A CN100538749C CN 100538749 C CN100538749 C CN 100538749C CN B2005101214371 A CNB2005101214371 A CN B2005101214371A CN 200510121437 A CN200510121437 A CN 200510121437A CN 100538749 C CN100538749 C CN 100538749C
- Authority
- CN
- China
- Prior art keywords
- image
- fiber
- random distribution
- distribution fiber
- counterfeit
- 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.)
- Active
Links
- 239000000835 fiber Substances 0.000 title claims abstract description 234
- 238000000034 method Methods 0.000 title claims abstract description 56
- 239000000284 extract Substances 0.000 title claims description 13
- 238000000605 extraction Methods 0.000 claims description 10
- 238000010408 sweeping Methods 0.000 claims description 5
- 238000005516 engineering process Methods 0.000 description 10
- 238000013461 design Methods 0.000 description 4
- 229920001131 Pulp (paper) Polymers 0.000 description 3
- 230000015572 biosynthetic process Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000010365 information processing Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000003756 stirring Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
Images
Landscapes
- Inspection Of Paper Currency And Valuable Securities (AREA)
Abstract
The present invention is a kind of to extract and the method for anti-counterfeit that uses the random distribution fiber characteristics of image, the geometric characteristic of intersection feature, fiber lines direction character and lines by extracting the article random distribution fiber is confirmed the distinctive feature of article, the database of random distribution fiber feature is set then, when differentiating article genuine-fake, only need the fibre characteristics information on the item inspecting, can identify the true and false of article, be used to prevent that article from being palmed off.
Description
[technical field]
The present invention relates to a kind of method for anti-counterfeit, specifically relate to a kind of the extraction and the method for anti-counterfeit that uses fibre characteristics information.
[background technology]
Paper is the important substance technical foundation of the fields of circulation such as various currency, securities, bill in the modern society.In order to prevent personation, present stage has been developed various types of cheque papers, and is widely used in as banknote, bill, check, stock, bond, gift token, various bill, card, and field such as passport.
Fluorescent fiber paper is a kind of effective counterfeit prevention paper of developing in recent years, and it is by fluorescent material is being dyed on the man-made fiber carrier, and in paper-making process this fluorescent fiber is added in the paper pulp according to certain ratio and make fluorescent fiber paper.In fluorescent fiber paper manufacture process, through each road technology of papermaking, as the stirring of paper pulp, paper stretching etc., fiber presents stochastic distribution in the finished product paper.This anti-forge paper surface seems the same with common paper, but if ultraviolet light one photograph in false distinguishing, the fluorescent fiber of adding in the paper just can present the fluorescence of a certain color immediately.
Though fluorescent fiber paper has security feature efficiently, the anti-counterfeiting characteristic that is characterized out by its fluorescent fiber also can duplicate and distort.Such as, at fluorescent fiber paper, discovery has the fluorescent fiber image that utilizes fluorescent ink (even e-Pointer) imitation fluorescent fiber paper to be shown, print artificially or print Counterfeit Item, and the paper that duplicates, also can present fibre image like the fluorescent fiber stationery under uviol lamp, naked eyes can't divide evident.Like this, just bring huge potential safety hazard to securities that utilize fluorescent fiber paper made and bill etc.So, how designing a kind of anti-anti-counterfeiting technology of duplicating and distorting function that has, this anti-counterfeiting technology had both made to the public does not openly lose its false proof effect yet, becomes present problem demanding prompt solution.
[summary of the invention]
The objective of the invention is to propose a kind of being difficult to is palmed off, extraction of distorting and the method for anti-counterfeit that uses fibre characteristics information, confirm the distinctive feature of article by the characteristic information that extracts the article random distribution fiber, the special card of the true fiber database of random distribution fiber feature is set then, when differentiating article genuine-fake, thereby by the true fiber image in fibre image to be verified and the database being compared the authenticity of determining fibre image to be verified.
The technical scheme that realizes above-mentioned purpose is:
(1) a kind of method for anti-counterfeit of random distribution fiber characteristics of image, comprise step: extract the image feature information of random distribution fiber and generate true random distribution fiber image information data storehouse, thereby by the true fiber image in fibre image to be verified and the database being compared the authenticity of determining fibre image to be verified.
(1) characteristics of image of described random distribution fiber is an intersection feature, and the method for extracting intersection feature information may further comprise the steps:
Pick up the original-gray image of random distribution fiber image;
The binaryzation original-gray image obtains binary image;
Some horizontal scanning line (l
1, l
2..., l
n) even partitioned image, and intersect with fiber and to be the point of crossing, add up every level of sweeping and retouch the quantity that line runs into fiber intersection points, each number of hits on the horizontal direction is designated as proper vector V
1=(n
1, n
2..., n
n);
Some vertical scan line (v
1, v
2..., v
n) even partitioned image, and intersect with fiber and to be the point of crossing, add up every and sweep and vertically retouch the quantity that line runs into fiber intersection points, each number of hits on the vertical direction is designated as proper vector V
2=(m
1, m
2..., m
n);
Random distribution fiber image intersection feature database is set, with true random distribution fiber characteristic information V
1, V
2In the input database.
(2) characteristics of image of described random distribution fiber is fiber lines direction characters, and the method for extracting fiber lines direction character information may further comprise the steps:
Pick up the original-gray image of random distribution fiber image;
The binaryzation original-gray image obtains binary image;
The scanning binary image is determined the fiber connected domain, and is connected domain numbering (0～t-1);
Binary image is divided into the interval of several marks in the scope of [90 °, 90 °], the connected domain direction of each numbering falls into one of interval of the some marks that are divided, and the interval mark that the connected domain direction is fallen into is as proper vector V
3=(d
1, d
2..., d
t);
Random distribution fiber image fiber lines direction character database is set, in true random distribution fiber characteristic information V3 input database.
(3) characteristics of image of described random distribution fiber is the lines geometric characteristic, and the method for extracting how much shape facility information of single bar may further comprise the steps:
The fiber lines are classified and numbering by geometric configuration;
Pick up the original-gray image of random distribution fiber image;
The binaryzation original-gray image obtains binary image;
Scan binary image and determine the fiber connected domain, judge affiliated how much classification of connected domain, geometric configuration under the connected domain is numbered as proper vector V
4=(s
1, s
2..., s
t);
How much character shape data storehouses of random distribution fiber image line bar are set, with true random distribution fiber characteristic information V
4In the input database.
How much Shape Classification of described single bar are: point-like, wire, u shape, n shape, 5 word shapes, 2 word shape and cross-like, number consecutively is the 1-7 class respectively.
How much classification may further comprise the steps under the described judgement connected domain:
Determine the area of the outer boundary region of fiber, if extraneous region area less than given individual threshold value, promptly this fiber is enough little, then is the 1st class lines, otherwise judgement according to the following steps;
Determining the end points number of fiber, if greater than 2, then is the 7th class, otherwise judges as follows;
If in the institute's line edge that is included in a little in the end points line or on the edge, then be the 2nd class, otherwise judge according to the following steps;
If the majority point in the end points line for upwards or left to, then be the 3rd class; If the majority point in the end points line is downward or right direction, then be the 4th class; If the end points line passes fiber, then fiber is the 5th or 6 classes, further, if first opening makes progress or a left side, then is the 5th class, otherwise is the 6th class.
As a kind of embodiment of database of the present invention, described database is based upon on the server.
As the another kind of embodiment of database of the present invention, described database is the bar code record that is dispersed on the article.
(2) a kind of method for anti-counterfeit that uses the random distribution fiber characteristics of image comprises the following steps:
Extract the geometric characteristic of intersection feature, line orientations feature and the lines of random distribution fiber image;
The random distribution fiber image feature base is set, in the geometric characteristic input database with intersection feature, line orientations feature and the lines of true random distribution fiber, and the true fiber image in fibre image to be verified and the database compared, thereby judge the authenticity of fibre image to be verified.
Described intersection feature extracting method may further comprise the steps: the original-gray image of picking up the random distribution fiber image; The binaryzation original-gray image obtains binary image; Some horizontal scanning line (l
1, l
2..., l
n) even partitioned image, and intersect with fiber and to be the point of crossing, add up every level of sweeping and retouch the quantity that line runs into fiber intersection points, each number of hits on the horizontal direction is designated as proper vector V
1=(n
1, n
2..., n
n); Some vertical scan line (v
1, v
2..., v
n) even partitioned image, and intersect with fiber and to be the point of crossing, add up every and sweep and vertically retouch the quantity that line runs into fiber intersection points, each number of hits on the vertical direction is designated as proper vector V
2=(m
1, m
2..., m
n).
Described fiber lines direction character extracting method may further comprise the steps: the original-gray image of picking up the random distribution fiber image; The binaryzation original-gray image obtains binary image; The scanning binary image is determined the fiber connected domain, and is connected domain numbering (0～t-1); Binary image is divided into the interval of several marks in the scope of [90 °, 90 °], the connected domain direction of each numbering falls into one of interval of the some marks that are divided, and the interval mark that the connected domain direction is fallen into is as proper vector V
3=(d
1, d
2..., d
t).
The geometric characteristic extracting method of described fiber lines may further comprise the steps: the original-gray image of picking up the random distribution fiber image; The binaryzation original-gray image obtains binary image; Scan binary image and determine the fiber connected domain, judge affiliated how much classification of connected domain, geometric configuration under the connected domain is numbered as proper vector V
4=(s
1, s
2..., s
t).
As a kind of embodiment of database of the present invention, described database is based upon on the server.
As the another kind of embodiment of database of the present invention, described database is the bar code record that is dispersed on the article.
The present invention adopts technique scheme, its beneficial technical effects is: 1) method for anti-counterfeit of the present invention is to be based on complicated random distribution fiber feature owing to what extract, be difficult to the people for duplicating and controlling, come false proof in the mode of setting up the fiber information storehouse simultaneously, the counterfeiter still is the fibre characteristics information that is difficult to duplicate or distort some stochastic distribution on the technology technically, the security that has improved anti-counterfeiting technology.2) method of the present invention is convenient to computer information processing, is simple intersection feature, fiber lines direction character and geometric characteristic with the characteristic quantification of random distribution fiber, make things convenient for computer information processing and with the comparison that is verified article.
[description of drawings]
Below by embodiment also in conjunction with the accompanying drawings, the present invention is described in further detail:
Fig. 1 is the fiber geometries figure that the present invention extracted and used the method for anti-counterfeit of random distribution fiber characteristics of image.
Fig. 2 is the fiber geometries figure that the present invention extracted and used the method for anti-counterfeit of random distribution fiber characteristics of image.
Fig. 3 is the fiber geometries figure that the present invention extracted and used the method for anti-counterfeit of random distribution fiber characteristics of image.
Fig. 4 is the fiber geometries figure that the present invention extracted and used the method for anti-counterfeit of random distribution fiber characteristics of image.
Fig. 5 is the fiber geometries figure that the present invention extracted and used the method for anti-counterfeit of random distribution fiber characteristics of image.
[embodiment]
In false proof article paper manufacture process, can add the fiber of some.The present invention extracts and uses the random distribution fiber characteristic information of any false proof article paper to come false proof, and the fiber characteristics that duplicates stochastic distribution is very difficult, based on this characteristic that is difficult to duplicate, so anti-counterfeiting technology of the present invention can disclose, and Authentication devices is set the third party, use for public's fake certification.
Described false proof thing paper product all adds false proof fluorescent fiber, each road technology of fiber process papermaking, as the stirring of paper pulp, paper stretching etc., fiber presents stochastic distribution in the article paper, bar number, direction, shape and the mutual relationship etc. that are embodied in fiber present the characteristics of stochastic distribution.In the method for anti-counterfeit of the present invention, extract satisfy in the random distribution fiber characteristic information of article paper that the class spacing is very big, in the class apart from very little special fiber proper vector group, described fiber characteristics Vector Groups is the distinctive feature between false proof article and the forgery article.
The present invention extracts and the method for anti-counterfeit that uses the random distribution fiber characteristics of image, and the geometric characteristic of intersection feature, fiber lines direction character and lines by extracting the article random distribution fiber is confirmed the distinctive feature of article.The geometric characteristic of described crunode feature, fiber lines direction character and lines also can be distinguished the distinctiveness fibre characteristics information that extracts separately as article.
After the geometric characteristic of the intersection feature of having extracted the article random distribution fibers, fiber lines direction character and lines, the database of true random distribution fiber characteristic information is set, the geometric characteristic of crunode feature, fiber lines direction character and the lines of the random distribution fiber that extracts is pressed the described database of taxonomy of goods typing.When differentiating article genuine-fake, the true fiber image in fibre image to be verified and the database is compared, thereby judge the authenticity of fibre image to be verified.
(1) described intersection feature is divided into horizontal line point of crossing and vertical curve point of crossing two classes, and the method for extracting intersection feature may further comprise the steps:
Pick up the original-gray image of random distribution fiber image, these fibers are actuated to visible, and pick up, just obtained the fiber gray level image of stochastic distribution by imageing sensor.
The binaryzation original-gray image obtains binary image.Binarization method is, sets a numerical value (being called binary-state threshold), each pixel in the check image, and when its value during greater than threshold value, this pixel is set to white (representing with 1), otherwise is black (representing with 0), has so just obtained bianry image.
8 horizontal scanning line (l are set
1, l
2..., l
8) even partitioned image, and intersect with fiber and to be the point of crossing, add up every level of sweeping and retouch the quantity that line runs into fiber intersection points, each number of hits on the horizontal direction is designated as proper vector V
1=(n
1, n
2..., n
8);
8 vertical scan line (v are set
1, v
2..., v
8) even partitioned image, and intersect with fiber and to be the point of crossing, add up every and sweep and vertically retouch the quantity that line runs into fiber intersection points, each number of hits on the vertical direction is designated as proper vector V
2=(m
1, m
2..., m
8);
With random distribution fiber characteristic information V
1, V
2In the input database.
(2) in conjunction with the direction of utilizing fiber, can remedy the deficiency in the intersection feature, the directional characteristic performing step of described fiber lines is as follows:
Pick up the original-gray image of random distribution fiber image, these fibers are actuated to visible, and pick up, just obtained the fiber gray level image of stochastic distribution by imageing sensor.
The binaryzation original-gray image obtains binary image.For bianry image, the fiber lines show as one by one independently connected domain.
Determine the fiber connected domain, and be the connected domain numbering.Concrete grammar is: each pixel in the scan image, from top to bottom, scanning from left to right.When scanning a connected domain, make marks, sequence number is since 0, is labeled as same mark at the pixel of same connected domain, and is not labeled as 1,2,3 successively at the mark of same connected domain, etc., establish total t bar fiber, then sequence number is from 0～t-1.
To the fiber of each sequence number, calculate the direction of the connected domain that all pixels constituted of forming this fiber, its scope is [90,90].[90 °, 90 °] interval is divided into 16 intervals, i.e. (90 ° ,-78.75 °), (78.75 ° 67.5 °) ..., (78.75 °, 90 °), and each interval done different marks, for example 0～15.Like this, the fiber of each sequence number just is divided into one of 16 intervals, and these marks as proper vector, are obtained V
3=(d
1, d
2..., d
t).Binary image is divided into the interval of several marks in the scope of [90 °, 90 °], the connected domain direction of each numbering falls into one of interval of the some marks that are divided, and the interval mark that the connected domain direction is fallen into is as proper vector V
3=(d
1, d
2..., d
t).
With random distribution fiber characteristic information V
3In the input database.
(3) the geometric characteristic extracting method of fiber lines
1) because distinctive flexibility of fiber lines and rigidity characteristic, the fiber lines present diversified shape in image, in addition, two or many s' intersection situation also may appear in position relation between the lines, in order to express the difference of these shapes, and simplify the difficult judgment that complex-shaped property is brought as far as possible, the fiber lines are divided into following 7 classes (definition):
The 1st class is " point-like " lines: all in given periphery, lines present shape a little to lines in the stretching, extension of both direction.
The 2nd class is " wire " lines: the institute's line edge that is included in a little in the end points line is interior or on the edge, as shown in Figure 1, this lines are approximately straight line.
The 3rd class is " U shape " lines: upwards or 2 curves of left opening, and as shown in Figure 3.
The 4th class is " n shape " lines: 2 curves of downward or right opening.As shown in Figure 2
The 5th class is " 5 word shape " lines: first opening makes progress or 3 times left curves, as shown in Figure 4.
The 6th class is " 2 word shape " lines, and Open Side Down or 3 times right curves for first.
The 7th class is the cross-like lines: the cross shaped head of 2 or many fibers, as shown in Figure 5.
2) performing step of the geometric characteristic extracting method of described fiber lines is as follows:
Pick up the original-gray image of random distribution fiber image, each pixel on the original color image is the RGB tlv triple, promptly use (r, g, b) expression, and each data is the nonzero integers in 255 scopes, gray processing adopts the gray processing method of standard, be g=0.299*r+0.587*g+0.114*b, wherein, g is the gray-scale value (nonzero integers in 255 scopes) of the corresponding pixel of gray level image.
The binaryzation original-gray image obtains binary image;
Scan binary image and determine the fiber connected domain, judge affiliated how much classification of connected domain, geometric configuration under the connected domain is numbered as proper vector V
4=(s
1, s
2..., s
t)
With random distribution fiber characteristic information V
4In the input database.
3) how much classification may further comprise the steps under the described judgement connected domain:
A, determine the area of the outer boundary region of fiber, if extraneous region area less than given individual threshold value, promptly this fiber is enough little, then is the 1st class lines, otherwise judgement according to the following steps;
If b, the end points number of deciding fiber greater than 2, then are the 7th class, otherwise judge as follows;
If in the institute's line edge that is included in a little in the c end points line or on the edge, then be the 2nd class, otherwise judge according to the following steps;
If the majority point in the d end points line for upwards or left to, then be the 3rd class; If the majority point in the end points line is downward or right direction, then be the 4th class; If the end points line passes fiber, then fiber is the 5th or 6 classes, further, if first opening makes progress or a left side, then is the 5th class, otherwise is the 6th class.
As a kind of embodiment of database of the present invention, described database is based upon on the server.The random distribution fiber image feature information centralized stores of gathering is in the data in server storehouse.True fiber image in fibre image to be verified and the database is compared, thereby judge the authenticity of fibre image to be verified.
As the another kind of embodiment of database of the present invention, described database is the bar code record that is dispersed on the article.When fake certification, computing machine reads in the bar code about random distribution fiber characteristic information data by bar-code reader, compares with the fiber characteristics that is verified article or bill again.
In the enforcement of bar code record random distribution fiber characteristics of image, adopted encryption technology, digital signature is encrypted and obtained to the characteristic information of random distribution fiber.Digital signature is made based on the identity identifying technology of rivest, shamir, adelman, mainly contains two kinds of operations of checking computing (V computing) of adopting signature computing (S computing) that private key finishes and employing PKI to finish in rivest, shamir, adelman.Wherein, utilize the sign signature result of computing of private key to be called digital signature, represented private key holder's signature, so require the secret private key of preserving of private key holder.
And PKI only is used to verify computing, can disclose.Private key and PKI occur in pairs, use each PKI and by the checking computing, can verify the true and false of the digital signature that use corresponding private key and signature computing are done, and the formation of digital signature and checking principle are:
M=HASH (M) formula (1)
S=f1 (m, sk1) formula (2)
V=f2 (S, pk1) formula (3)
In the formula, M is called expressly, is original character and the numerical information that is used to make digital signature, and this expressly is the special characteristic information of gathering of random distribution fiber among the present invention.M is called HASH expressly, is that M is carried out the numerical information that HASH asks the digest value computing to obtain; F1 is a signature function, and this argument of function is m and private key sk1, and functional value is digital signature S, represents with the form of numeral.F2 is the checking function, and this argument of function is digital signature S and PKI pk1 (sk1 becomes pair of secret keys with private key), and functional value is identifying code V, f1 and f2 inverse operation each other, if V equals m, then explanation checking result is correct; If V is not equal to m, authentication error is described then.
Digital signature is actually between plaintext M and digital signature result S and sets up the process that certain is got in touch, and the people that private key is only held in this contact can sign and issue out digital signature, so this digital signature can be as the anti-counterfeiting identification code of article.Because the fake producer does not have private key, even he can grasp the feature formation method of random distribution fiber image so, oneself generates fibre image, can not obtain correct digital signature, still can be identified by the checking computing.And for the owner of article or bill, he holds the private key of signature and the PKI of checking usefulness, so can sign and issue and certifying digital signature.Simultaneously the owner of article or bill also can license to the PKI of checking usefulness and needed algorithm reliable third-party authentication mechanism, but this third-party authentication mechanism certifying digital signature but but can not sign and issue out digital signature.
In the process of extracting the random distribution fiber feature, owing to the variation of material behavior, the error of equipment, the reasons such as error of image processing software, authentication is identified as counterfeit with real article or bill probably.For fear of the generation of this situation, need provide certain image and show other range of allowable error.The size of error provides after will taking all factors into consideration above-mentioned factor, if design error is too small, may be counterfeit articles with true article identification; Counterfeit articles may be identified as true article again if the error design is excessive.Described error also will be considered the difference that different application requires except considering above-mentioned factor, mustn't change along with the difference of using.For example, in bank's a large sum of money bill anti-counterfeit is used,, wish the error range that design is less for the true and false of strictness checking bill.
For this reason, the present invention designs a bar code area on the appropriate location of article or bill, can be the bar code of one dimension or two dimension, is used for writing down fibre characteristics information and identification allowable error thereof at random.And when checking, when extracting the random distribution fiber image feature information, can be with reference to the identification allowable error information in this bar code, to help judging whether the fibre image characteristic information that is extracted is correct.
Read the bar code information of article or bill by bar code reader.Computing machine has interface and the bitcom that carries out communication with bar code reader, bar code figure can be converted to numerical information and be kept in the computing machine.Computing machine is according to fibre characteristics information at random that comes from the bar code conversion and fiber identification permissible error, with fibre image characteristic information that is verified and the data of changing out from bar code, compare one by one by feature, if the deviation of each characteristic is all in the permissible error scope of correspondence, then judge to be verified article or bill, otherwise be judged to be vacation for true.
Claims (14)
1, a kind of method for anti-counterfeit that extracts the random distribution fiber characteristics of image, it is characterized in that, described method comprises step: the image feature information that extracts random distribution fiber, the characteristics of image of described random distribution fiber comprises quantitative intersection feature, and the lines geometric characteristic of the fiber lines direction character of vector and vector, and generate true random distribution fiber image information data storehouse, thereby by the true fiber image in fibre image to be verified and the database being compared the authenticity of determining fibre image to be verified.
2, the method for anti-counterfeit of extraction random distribution fiber characteristics of image according to claim 1 is characterized in that: the method for extracting intersection feature information may further comprise the steps:
Pick up the original-gray image of random distribution fiber image;
The binaryzation original-gray image obtains binary image;
Some horizontal scanning line l
1, l
2..., l
nEven partitioned image, and intersect with fiber and to be the point of crossing, add up every level of sweeping and retouch the quantity that line runs into fiber intersection points, each number of hits on the horizontal direction is designated as proper vector V
1=(n
1, n
2..., n
n);
Some vertical scan line ν
1, ν
2..., ν
nEven partitioned image, and intersect with fiber and to be the point of crossing is added up every and is swept and vertically retouch the quantity that line runs into fiber intersection points, and each number of hits on the vertical direction is designated as proper vector V
2=(m
1, m
2..., m
n);
Random distribution fiber image intersection feature database is set, with true random distribution fiber characteristic information V
1, V
2In the input database.
3, the method for anti-counterfeit of extraction random distribution fiber characteristics of image according to claim 1 is characterized in that: the method for extracting fiber lines direction character information may further comprise the steps:
Pick up the original-gray image of random distribution fiber image;
The binaryzation original-gray image obtains binary image;
The scanning binary image is determined the fiber connected domain, and is connected domain numbering 0～t-1;
Binary image is divided into the interval of several marks in the scope of [90 °, 90 °], the direction of the connected domain of each numbering falls into one of interval of the some marks that are divided, and the interval mark that the connected domain direction is fallen into is as proper vector V
3=(d
1, d
2..., d
t);
Random distribution fiber image fiber lines direction character database is set, with true random distribution fiber characteristic information V
3In the input database.
4, the method for anti-counterfeit of extraction random distribution fiber characteristics of image according to claim 1 is characterized in that: the method for extracting how much shape facility information of single bar may further comprise the steps:
The fiber lines are classified and numbering by geometric configuration;
Pick up the original-gray image of random distribution fiber image;
The binaryzation original-gray image obtains binary image;
Scan binary image and determine the fiber connected domain, judge affiliated how much classification of connected domain, geometric configuration under the connected domain is numbered as proper vector V
4=(s
1, s
2..., s
t);
How much character shape data storehouses of random distribution fiber image line bar are set, with true random distribution fiber characteristic information V
4In the input database.
5, the method for anti-counterfeit of extraction random distribution fiber characteristics of image according to claim 4, it is characterized in that: how much Shape Classification of described single bar are: point-like, wire, u shape, n shape, 5 word shapes, 2 word shape and cross-like, number consecutively is the 1-7 class respectively.
6, the method for anti-counterfeit of extraction random distribution fiber characteristics of image according to claim 5 is characterized in that: how much classification may further comprise the steps under the described judgement connected domain:
Determine the area of the outer boundary region of fiber, if extraneous region area less than given individual threshold value, promptly this fiber is enough little, then is the 1st class lines, otherwise judgement according to the following steps;
Determining the end points number of fiber, if greater than 2, then is the 7th class, otherwise judges as follows;
If in the institute's line edge that is included in a little in the end points line or on the edge, then be the 2nd class, otherwise judge according to the following steps;
If the majority point in the end points line for upwards or left to, then be the 3rd class; If the majority point in the end points line is downward or right direction, then be the 4th class; If the end points line passes fiber, then fiber is the 5th or 6 classes, further, if first opening makes progress or a left side, then is the 5th class, otherwise is the 6th class.
7, according to the method for anti-counterfeit of any described extraction random distribution fiber characteristics of image of claim 1-6, it is characterized in that: described database is based upon on the server, and perhaps described database is the bar code record that is dispersed in the image feature information that contains random distribution fiber on the article.
8, the method for anti-counterfeit of extraction random distribution fiber characteristics of image according to claim 7 is characterized in that: the image feature information to the random distribution fiber that adopts the bar code record is encrypted.
9, a kind of method for anti-counterfeit that uses the random distribution fiber characteristics of image is characterized in that: comprise the following steps:
The geometric characteristic of the lines of the quantitative intersection feature of extraction random distribution fiber image, the line orientations feature of vector and vector;
The random distribution fiber image feature base is set, in the geometric characteristic input database with intersection feature, line orientations feature and the lines of true random distribution fiber, and the true fiber image in fibre image to be verified and the database compared, thereby judge the authenticity of fibre image to be verified.
10, the method for anti-counterfeit of use random distribution fiber characteristics of image according to claim 9 is characterized in that: described intersection feature extracting method may further comprise the steps:
Pick up the original-gray image of random distribution fiber image;
The binaryzation original-gray image obtains binary image;
Some horizontal scanning line l
1, l
2..., l
nEven partitioned image, and intersect with fiber and to be the point of crossing, add up every level of sweeping and retouch the quantity that line runs into fiber intersection points, each number of hits on the horizontal direction is designated as proper vector V
1=(n
1, n
2..., n
n);
Some vertical scan line ν
1, ν
2..., ν
nEven partitioned image, and intersect with fiber and to be the point of crossing is added up every and is swept and vertically retouch the quantity that line runs into fiber intersection points, and each number of hits on the vertical direction is designated as proper vector V
2=(m
1, m
2..., m
n).
11, the method for anti-counterfeit of use random distribution fiber characteristics of image according to claim 9 is characterized in that: described line orientations feature extracting method may further comprise the steps:
Pick up the original-gray image of random distribution fiber image;
The binaryzation original-gray image obtains binary image;
The scanning binary image is determined the fiber connected domain, and is connected domain numbering 0～t-1;
Binary image is divided into the interval of several marks in the scope of [90 °, 90 °], the connected domain direction of each numbering falls into one of interval of the some marks that are divided, and the interval mark that the connected domain direction is fallen into is as proper vector V
3=(d
1, d
2..., d
t).
12, the method for anti-counterfeit of use random distribution fiber characteristics of image according to claim 9 is characterized in that: the geometric characteristic extracting method of described lines may further comprise the steps:
Pick up the original-gray image of random distribution fiber image;
The binaryzation original-gray image obtains binary image;
Scan binary image and determine the fiber connected domain, judge affiliated how much classification of connected domain, geometric configuration under the connected domain is numbered as proper vector V
4=(s
1, s
2..., s
t).
13, according to the method for anti-counterfeit of any described use random distribution fiber characteristics of image of claim 9-12, it is characterized in that: described database is based upon on the server, and perhaps described database is the bar code record that is dispersed in the image feature information that contains random distribution fiber on the article.
14, the method for anti-counterfeit of use random distribution fiber characteristics of image according to claim 13 is characterized in that: the image feature information to the random distribution fiber that adopts the bar code record is encrypted.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2005101214371A CN100538749C (en) | 2005-12-29 | 2005-12-29 | Extract and the method for anti-counterfeit that uses the random distribution fiber characteristics of image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2005101214371A CN100538749C (en) | 2005-12-29 | 2005-12-29 | Extract and the method for anti-counterfeit that uses the random distribution fiber characteristics of image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1794303A CN1794303A (en) | 2006-06-28 |
| CN100538749C true CN100538749C (en) | 2009-09-09 |
Family
ID=36805721
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2005101214371A Active CN100538749C (en) | 2005-12-29 | 2005-12-29 | Extract and the method for anti-counterfeit that uses the random distribution fiber characteristics of image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100538749C (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104134150A (en) * | 2014-07-30 | 2014-11-05 | 刘思岩 | Computer assisted Chinese art paper painting and calligraphy identification method |
| WO2016177301A1 (en) * | 2015-05-01 | 2016-11-10 | 励元科技(上海)有限公司 | Micro-texture anti-counterfeit method |
| CN104809622A (en) * | 2015-05-01 | 2015-07-29 | 海南亚元防伪技术研究所 | Ultra-limit texture anti-counterfeiting method |
| CN104834747B (en) * | 2015-05-25 | 2018-04-27 | 中国科学院自动化研究所 | Short text classification method based on convolutional neural networks |
| CN207380997U (en) * | 2016-09-30 | 2018-05-18 | 北京柯斯元科技有限公司 | A kind of anti-counterfeit sign with random texture |
| CN106991419A (en) * | 2017-03-13 | 2017-07-28 | 特维轮网络科技（杭州）有限公司 | Method for anti-counterfeit based on tire inner wall random grain |
| CN107204068B (en) * | 2017-05-27 | 2019-08-27 | 深圳怡化电脑股份有限公司 | A kind of paper money management system, method and Paper Currency Identification |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1293118A (en) * | 2000-09-28 | 2001-05-02 | 李维华 | Antiforge methode by setting up combined information of fiber characteristics with digital code |
| CN1293424A (en) * | 2000-09-28 | 2001-05-02 | 李维华 | Antiforge method by setting up and using fibre characteristics information |
| CN1349186A (en) * | 2001-11-21 | 2002-05-15 | 孙显林 | Random grain anti-fake method |
| CN1430175A (en) * | 2001-12-30 | 2003-07-16 | 北京兆日科技有限责任公司 | Encryption antiforgery method based on substance characteristics |
| CN2715245Y (en) * | 2003-09-22 | 2005-08-03 | 兆日科技（深圳）有限公司 | Texture password label |
-
2005
- 2005-12-29 CN CNB2005101214371A patent/CN100538749C/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1293118A (en) * | 2000-09-28 | 2001-05-02 | 李维华 | Antiforge methode by setting up combined information of fiber characteristics with digital code |
| CN1293424A (en) * | 2000-09-28 | 2001-05-02 | 李维华 | Antiforge method by setting up and using fibre characteristics information |
| CN1349186A (en) * | 2001-11-21 | 2002-05-15 | 孙显林 | Random grain anti-fake method |
| CN1430175A (en) * | 2001-12-30 | 2003-07-16 | 北京兆日科技有限责任公司 | Encryption antiforgery method based on substance characteristics |
| CN2715245Y (en) * | 2003-09-22 | 2005-08-03 | 兆日科技（深圳）有限公司 | Texture password label |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1794303A (en) | 2006-06-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP1569166B1 (en) | Counterfeit and tamper resistant labels with randomly occurring features | |
| CN2715245Y (en) | Texture password label | |
| RU2470361C2 (en) | Secure barcode | |
| CN101537751B (en) | Novel method for processing anti-counterfeit printing | |
| Alekhya et al. | Fake currency detection using image processing and other standard methods | |
| Mirza et al. | Paper currency verification system based on characteristic extraction using image processing | |
| CN103165020B (en) | Anti-fake product and manufacturing method thereof and anti-fake detection device and method thereof | |
| CN103761799A (en) | Bill anti-counterfeit method and device based on texture image characteristics | |
| CN101540005A (en) | Method for processing random figure code | |
| JP2015515388A (en) | Security element and method of checking the authenticity of the stamp | |
| CN100538749C (en) | Extract and the method for anti-counterfeit that uses the random distribution fiber characteristics of image | |
| CN107204068A (en) | A kind of paper money management system, method and Paper Currency Identification | |
| CN104239928B (en) | Anti-counterfeiting printed matter and corresponding anti-counterfeiting method | |
| US12175479B2 (en) | Method and systems for authenticating and securing physical items with registering a 3D digital representation as a non-fungible token | |
| EP1418542B1 (en) | Watermarking a security document | |
| CN201242832Y (en) | Stealth inner structure pattern false proof confidential document and product thereof | |
| CN102982606A (en) | Anti-fake method using physical feature recognition and anti-fake system using physical feature recognition | |
| CN1385823A (en) | Ticket and security ciphering and identifying method based on identified fibre random distribution | |
| CN106780953B (en) | Paper money counterfeit distinguishing method and system based on double-crown-word number | |
| Tkachenko et al. | Authentication of rotogravure print-outs using a regular test pattern | |
| Upadhyaya et al. | Counterfeit currency detection techniques | |
| CN100405405C (en) | Fiber image antifake method | |
| CN101261749B (en) | Anti-counterfeit technology implementation system for banknote digital information | |
| Shoeb et al. | Software system to detect counterfeit Egyptian currency | |
| CN1271584C (en) | Digital multiple antifakery labels |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C56 | Change in the name or address of the patentee |
Owner name: SHENZHEN ZHAORI TECHNOLOGY CO., LTD. Free format text: FORMER NAME: ZHAORI SCIENCE + TECHNOLOGY (SHENZHEN) CO., LTD. |
|
| CP01 | Change in the name or title of a patent holder |
Address after: 518040 Guangdong city of Shenzhen province Futian District Che Kung Temple Tairan Industrial Zone 213 6C Patentee after: Shenzhen Sinosun Technology Co., Ltd. Address before: 518040 Guangdong city of Shenzhen province Futian District Che Kung Temple Tairan Industrial Zone 213 6C Patentee before: Sinosun Technology (Shenzhen) Co., Ltd. |