CN106991419A - Method for anti-counterfeit based on tire inner wall random grain - Google Patents
Method for anti-counterfeit based on tire inner wall random grain Download PDFInfo
- Publication number
- CN106991419A CN106991419A CN201710146621.4A CN201710146621A CN106991419A CN 106991419 A CN106991419 A CN 106991419A CN 201710146621 A CN201710146621 A CN 201710146621A CN 106991419 A CN106991419 A CN 106991419A
- Authority
- CN
- China
- Prior art keywords
- image
- tire
- random
- point
- random grain
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- 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
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/018—Certifying business or products
- G06Q30/0185—Product, service or business identity fraud
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/28—Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/34—Smoothing or thinning of the pattern; Morphological operations; Skeletonisation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Development Economics (AREA)
- Evolutionary Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Entrepreneurship & Innovation (AREA)
- Accounting & Taxation (AREA)
- Evolutionary Computation (AREA)
- Economics (AREA)
- Finance (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of method for anti-counterfeit based on tire inner wall random grain.Outside coating is diffused on crude tyre inwall by the present invention, occurs random Light deformation after outside coating dry solidification, so that tire inner wall produces random grain, is obtained and random grain image one-to-one random character code by image characteristics extraction algorithm;User is taken pictures and uploaded onto the server to the texture image of tire to be verified;The checking condition code of tire to be verified is obtained by identical image characteristics extraction algorithm, according to random character code with verifying that the similarity of condition code judges whether tire texture image to be verified matches with random grain image;The match is successful, then it is crude tyre to prove the tire.Tire inner wall textural characteristics and outside coating are combined by the present invention, tire anti-counterfeiting image pickup area is solidificated in inside tires, mark that cost is low, method for anti-counterfeit is simple and easy to apply using the textural characteristics of tire inherently as anti-counterfeit recognition, are easy to industrialization to make.
Description
Technical field
The invention belongs to market after automobile, and in particular to a kind of method for anti-counterfeit based on tire inner wall random grain.
Background technology
With the fast development and the raising of national life level of urban infrastructure, China's car ownership is continuously and healthily
Increase, market potential sustained release after automobile, the O2O service platforms of market segment have reached thousands of families after domestic automobile, compete
It is abnormal fierce, bring more quality services experience to turn into the key that major O2O service platforms win user by innovation.
The insurance that domestic only several moneys cover tire guarantee belongs to product quality insurance, and tire casualty insurance still belongs to
Blank field, insurance company to tire surprisingly do not pay for it is not protecting main reason is that tire belong to consumable accessory and easily replace,
Core guarantor's difficulty of insuring that tire is directed in the case of lacking tire identification technology is very big.Not yet occur at present any for being sold
The tire method that carries out unique mark, the existing difficulty for implementing complicated and tire anti-counterfeiting detection based on antiforge method for commodities
Greatly, the cycle is long.
The content of the invention
The present invention is in view of the shortcomings of the prior art, it is proposed that a kind of method for anti-counterfeit based on tire inner wall random grain.
Outside coating is diffused on tire inner wall by the present invention, occurs random Light deformation after outside coating dry solidification, from
And tire inner wall is produced random texture, obtain special correspondingly with original texture image by image characteristics extraction algorithm
Code is levied, condition code is stored in tire false proof database;User is clapped the texture image of tire to be verified by mobile phone
According to uploading onto the server, the checking condition code of tire to be verified is obtained by identical image characteristics extraction algorithm, according to two
The similarity of condition code judges whether tire texture image to be verified matches with original image, and the match is successful then proves the tire by putting down
Platform is sold, you can the tire for enjoying platform offer surprisingly ensures service.
Beneficial effects of the present invention:
(1) feasibility is strong, and tire inner wall textural characteristics and outside coating are combined by the present invention, adopt tire anti-counterfeiting image
Collection is regions curing in inside tires, marks that cost is low, method for anti-counterfeit using the textural characteristics of tire inherently as anti-counterfeit recognition
It is simple and easy to apply, it is easy to industrialization to make.
(2) texture structure uniqueness, the tire of generation is ensure that by the irregular combination of tire inner wall decorative pattern and coating
Inwall texture structure randomness and uniqueness.
(3) difficulty is copied high, it is of the invention that tire inner wall random grain characteristic information is stored in platform by picture collection
Inside, the condition code extracted by algorithm can effectively preserve the minutia of texture structure image, even if attempting to forge phase
As the condition code that extracts of tire inner wall texture structure also differ.
(4) accuracy rate is high, and based on the similitude of partial structurtes on the inside of tire, the method voted using neighbour domain rejects mistake
Matching double points obtain accurately matching point set, effectively increase the accuracy rate of condition code similarity identification.
(5) detection speed is fast, and the present invention is completely embedded into inside the existing usage scenario of electric business platform, and consumer passes through lossless
Whether shooting style quick search tire is by this gondola sales, and the cycle of detection is short, and consumer can enjoy rapidly after detecting successfully
The tire provided by platform surprisingly ensures service, effectively improves Consumer's Experience and improves loyalty of the car owner for platform
Really spend.
Brief description of the drawings
Fig. 1 is based on tire inner wall random grain false-proof method overall flow figure;
Fig. 2 is image preprocessing flow chart;
Fig. 3 is crude tyre inwall texture;
Fig. 4 is crude tyre inwall texture;
Fig. 5 is tire inner wall texture to be verified;
Scheme after Fig. 6 Gaussian smoothings and Otsu binaryzations;
Fig. 7 is figure after morphological erosion;
Fig. 8 is figure after morphological dilations;
Fig. 9 is figure after minimum bounding box positioning;
Figure 10 is characterized a yard matching result visual presentation figure.
Embodiment
Below in conjunction with drawings and examples, the invention will be further described.
As shown in figure 1, technical scheme includes four steps:
Step 1:Tire inner wall random grain is generated:
Seal, as a kind of keepsake, is the representative of legal rights in China since ancient times, be public organization, government offices,
Enterprises and institutions or the important evidence of personal proving authenticity and validity when exercising legitimate authority, trace is the print face of seal
The vestige for impressing out by ink paste, stamp-pad ink, stamp-pad ink are because with good water resistance, heat resistance, light resistance, resistance to acids and bases etc.
Advantage can be used directly in tire inner wall.In order that stamp-pad ink produces larger curtain coating amplitude of deformation on tire inner wall and expansion may be selected
Dissipate the strong stamp-pad ink of property to print, stamp-pad ink is printed onto on tire inner wall by seal, form a random grain structural images and adopt
Collect region and supply follow-up anti-counterfeit recognition.Mainly efficient combination by the following method generates unique random grain:
(1) different brands, the tire inner wall decorative pattern of different model are typically different from, and the decorative pattern of tire inner wall is uneven
Cause stamp-pad ink irregular, dispersal direction when spreading thereon random.
(2) even if same brand, the tire inner wall decorative pattern of same model are identical, the position that seal is covered is random.
(3) after stamp-pad ink is printed on tire inner wall, before the not yet thorough dry solidification of stamp-pad ink, different directions are passed through
The external force such as wind promote stamp-pad ink to occur micro-displacement, produce more notable so as to be combined with the rough decorative pattern of tire inner wall
, personalized random grain, while accelerating the rate of drying of stamp-pad ink.
Step 2:Image preprocessing:
The random grain structural images and the first-order partial derivative convolution of two-dimensional Gaussian function that step 1 is produced, make original graph
As reaching that output image is converted to gray level image after smooth effect, Gaussian smoothing, and prospect is tried to achieve by maximum variance between clusters
Optimal segmenting threshold between background, bianry image is converted gray images into according to the threshold value, to obtained bianry image
Mathematical morphology erosion operation several times is carried out, being significantly less than the background in random grain region to remove area in bianry image makes an uproar
Sound, then recovers random grain region area size by mathematical morphology dilation operation several times, finally with minimum encirclement
Box algorithm orients the area-of-interest in bianry image and exports the color texture image of area-of-interest to next step.
Image preprocessing flow chart is as shown in Figure 2.
Step 3:Tire inner wall texture feature extraction and selection:
SIFT algorithms are best feature point extraction and description operator, and the characteristic point that it is extracted has not to yardstick and rotation
Denaturation, has certain consistency simultaneously for brightness and three-dimensional view angle change.The present invention utilizes this feature of SIFT algorithms
The tire inner wall random grain that extraction step 2 is produced, key step is as follows:
(1) structure of metric space:The structure of metric space is the Analysis On Multi-scale Features for simulated image data, random line
Reason feature extraction is completed in multiscale space, ensure that the feature extracted has scale invariability, makes what platform was preserved
Crude tyre inwall texture image is consistent with the tire inner wall texture image to be verified that user gathers, and large scale preserves tire inner wall
General picture feature, small yardstick preserves the minutia of tire inner wall.
(2) detection of key point:Three-dimensional extreme point is detected in Gaussian difference scale space as the characteristic point of candidate, so
Three-dimensional extreme point is fitted with quadratic function afterwards, fitting function is
Wherein, X=(x, y, σ), in order to obtain the offset of extreme point, by formula (1) derivation and allow equation be equal to 0, formula (1)
Formula (2) is can obtain,
UtilizeValue can reject the unstable characteristic point of low contrast, to strengthen the steady of next step Feature Points Matching
Qualitative, raising anti-noise ability, because difference of Gaussian can produce strong edge response point, it is therefore desirable to reject, use candidate point
Hessian matrixes handled;
(3) direction of key point is determined:It is that each key point refers to based on the local attribute of tire inner wall random grain image
One or several fixed directions, and description of each key point is closely related with these directions, so as to realize description
Consistency.Each the direction of key point is determined in the mould of each pixel and direction in the vertex neighborhood, the mould and ladder of certain pixel
Spend direction calculating formula as follows:
θ (x, y)=tan-1((L (x, y+1)-L (x, y-1))/(L (x+1, y)-L (x-1, y))) (5)
Wherein, L (x, y) is the image on certain group metric space, and m (x, y) is the gradient modulus value in (x, y) place pixel, θ
(x, y) is the gradient direction in (x, y) place pixel, and each picture in each crucial vertex neighborhood is can obtain using formula (4) and formula (5)
The gradient-norm of element and direction, for each characteristic point for detecting in its neighborhood statistical yardstick direction histogram, try to achieve straight
The peak value of square figure as characteristic point principal direction.
(4) generation of key feature points description:After above-mentioned processing characteristic point comprising coordinate information, dimensional information,
Principal direction information etc., is rotated coordinate system according to principal direction, so that it is guaranteed that description of generation has not to image rotation
Denaturation, in order to avoid the catastrophe of description, makes the gradients affect proportion remote from description subwindow center smaller, reduces
Influence to matching error, introduces gaussian weighing function, a weights is distributed for each pixel in description subwindow, from spy
Levy that the remote pixel weight of key point is small, the information contribution closer to the pixel gradient direction of key point is bigger, to every sub- window
The gradient direction of pixel in mouthful, at this moment will be centered on key point from eight direction travel direction statistics with histogram
N data of generation in d*d windows, for each key point, the characteristic point required for the n dimensional feature vectors of generation are is retouched
State son.
After the n dimensional feature vectors of key point are formed, in order to remove the influence of illumination variation, it is necessary to carry out normalizing to them
Change is handled, and for the overall drift of image intensity value, because the gradient of image each point is that neighborhood territory pixel subtracts each other and obtained, so
Also it can remove.Because the non-linear illumination effect that tire inner wall image capture environment factor is caused can cause the ladder in some directions
Spend modulus value excessive, but faint is influenceed on gradient direction, by setting threshold value to block larger Grad so as to reduce modulus value
The influence of larger gradient, then carries out a normalized again, improves the quality of description.
(5) condition code of original texture image enters false proof database:The feature point description sub-portfolio calculated is turned into original
Preserved in the condition code typing false proof database of beginning texture image.
Step 4:Condition code similarity judges:Assuming that some characteristic point A in the original texture image that platform is preserved, its
The vectorial D that corresponding feature point description produces for step 3A, user submit texture image to be verified in existing characteristics point B and
C, wherein characteristic point B are the point nearest with characteristic point A Euclidean distances, and C is time closest approach, the two corresponding feature point description son point
Wei not vector DBWith vectorial DCIf Euclidean distance ratio meets formula (6), then it is assumed that feature in the texture image to be verified that user submits
The candidate matches point that point B is characteristic point A in original texture image.
Due to the influence for the factor such as the partial structurtes of tire inner wall are similar, based on the feature that Euclidean distance is similarity measurement
The matching of mistake is there may be in matching result.Just because of the similitude of partial structurtes on the inside of tire, in correct match point
Surrounding certainly exists more correct matching results, if not correct matching result, then match point is seldom in its neighborhood, very
To not there is match point, thus can add up the match point of surrounding local principal direction and on to the candidate matches point
Contribution, and judge the matching double points of texture image to be verified and original texture image apart from the degree of correlation and the local angle degree of correlation
Whether in threshold range, it is correct match point if in threshold range, otherwise abandons the candidate matches point, institute of the present invention
The method of the neighbour domain ballot of use can reject error matching points, obtain accurately matching point set.Eventually through matching point set
Quantity and threshold value be compared the similarity degree of judging characteristic intersymbol.
Embodiment:
Step 1:Tire inner wall random grain is generated
Crude tyre inwall texture as shown in figure 3, the rough texture of inwall is produced along with tire manufacturing process,
Different brands, the tire inner wall decorative pattern of different model are typically different from, and general tire is black, and the present embodiment is using white
Color stamp-pad ink makes random grain, and white stamp-pad ink is printed on tire inner wall by square seal, and the wind for passing through different directions
Make stamp-pad ink rapid draing and occur small displacement.Embodiments of the invention are only chosen a type of seal and illustrated, including but
Be not limited only to the seal of this type, the seal of other shapes buckle be imprinted on it is within the scope of the present invention on the inside of tire.
The tire inner wall texture structure completed is as shown in Figure 4.In order to simulate actual scene, from different perspectives to same
Two photos of individual inwall texture structure collection, the crude tyre inwall texture image that a width retains as platform, as shown in figure 4,
An other width is that the tire texture image to be verified that consumer shoots is as shown in Figure 5.
Step 2:Image preprocessing
Texture structure coloured image in Fig. 5 is converted into gray level image by gray processing, removes and makes an uproar by Gaussian smoothing
After sound, and the optimal segmenting threshold between foreground and background is tried to achieve by maximum variance between clusters, by the threshold value by gray-scale map
As being converted into bianry image, as shown in Figure 6.
Morphological erosion computing several times (such as Fig. 7) is passed through to obtained bianry image, to remove area in bianry image
It is significantly less than the ambient noise in random grain region, random grain area is then expanded by mathematical morphology dilation operation several times
Domain area is in order to minimum bounding box algorithm positioning (such as Fig. 8).
Finally area-of-interest in bianry image is oriented with minimum bounding box algorithm and by area-of-interest with red
Color circle marks (such as Fig. 9), and the coloured image in the region is exported to next step.Use the effective of digital image processing techniques
Random grain region is within the scope of the present invention on the inside of the tire that integrated positioning goes out.
Step 3:Tire inner wall random grain feature extraction and selection
Detect that three-dimensional extreme point, as candidate feature point, is then picked by formula (2) first in Gaussian difference scale space
Except the characteristic point of low contrast, three-dimensional extreme point is accurately positioned, for all extreme points encountered in the present embodiment,
If the fitting extreme value of extreme pointThen think sampled point of this point for low contrast, and remove it.
There is a very strong response along edge direction difference of Gaussian function, and the extreme point in edge is difficult positioning,
Therefore the extreme point in edge is unstable, is also easily influenceed by a small amount of noise, using Hessian matrix disposal candidate features
λ in point, the present embodiment formula (3)010 are taken, is retained when formula (3) is set up as characteristic point, otherwise regard the point as strong side
Edge response point carries out rejecting processing.
Gradient-norm and the direction of each pixel each crucial vertex neighborhood Nei are can obtain using formula (4) and (5), because key point
Direction determined in the mould of each pixel and direction in the vertex neighborhood, so need to set up a direction histogram for key point,
Direction histogram will be equally divided into 36 blocks 360 °, and the interval width of each block is 10 °, and the gradient direction of pixel is determined
Modulus value is determined to be placed in which histogrammic block, the area is added after the modulus value of this pixel then is carried out into Gauss weighting
Block, is formed the direction histogram of the key point.
The corresponding direction of block where the peak-peak of direction histogram is the principal direction of the key point;Reference axis is revolved
Go to after the principal direction of key point, key point surrounding neighbors are divided into 4 × 4 subregion, and calculate the figure in each 4 × 4 region
As fritter is in the gradient orientation histogram in 8 directions, the accumulated value of each gradient direction is calculated, one is consequently formed by 8 dimensions
The seed point of vector representation, therefore 4 × 4 seed points produce the characteristic vector of 4 × 4 × 8=128 dimension, the spy of 128 dimension
It is condition code to levy vector, and the minutia with tire inner wall random grain, characteristic vector pickup result is as follows:
In order to eliminate influence of the illumination variation to characteristic vector, it is necessary to make standardization, linear to characteristic vector
Illumination variation, characteristic vector is standardized as unit length, for non-linear illumination variation, first sets threshold value to make unit character
The value of vector is no more than 0.2, and characteristic vector then is standardized as into unit length again.The present embodiment is only with SIFT feature extraction side
Method is illustrated, and other have the Robust Algorithm of Image Corner Extraction of yardstick, rotational invariance also within the scope of the present invention.
Step 4:Condition code is matched and similarity judges
Assuming that A be Fig. 4 original texture images in some characteristic point, its corresponding condition code be step 3 produce to
Measure DA, Fig. 5 user submit texture image to be verified in existing characteristics point B and C, the two corresponding condition code is respectively vectorial DB
With vectorial DC, wherein characteristic point B is the point nearest with characteristic point A Euclidean distances, and C is time closest approach, if Euclidean distance is than meeting
T is typically set to 0.8 in formula (6), the present embodiment, then it is considered that A and B can be with Corresponding matching, then using the ballot of neighbour domain
Method rejects error matching points.
Figure 10 shows that the two ends of straight line connection are algorithm judgement for Fig. 4 and Fig. 5 Partial Feature Point matchings result visualization
Same characteristic point, is compared in judgement similarity degree, the present embodiment according to point set quantity is matched in area-of-interest with threshold value
Threshold value is 50, thinks that two condition codes are similar if the quantity that point set is matched in two area-of-interests is more than or equal to 50, original line
Manage image it is identical with texture image to be verified, it is on the contrary then think two texture images difference.
Claims (8)
1. the method for anti-counterfeit based on tire inner wall random grain, it is characterised in that this method is specifically:
Outside coating is diffused on crude tyre inwall, occurs random Light deformation after outside coating dry solidification, so that wheel
Tire inwall produces random grain, is obtained and the one-to-one random character of random grain image by image characteristics extraction algorithm
Code, random character code is stored in tire false proof database;
User is taken pictures and uploaded onto the server to the texture image of tire to be verified by intelligent terminal;Pass through identical figure
As feature extraction algorithm obtains the checking condition code of tire to be verified, according to random character code with verifying that the similarity of condition code is sentenced
Whether the tire texture image to be verified that breaks matches with random grain image;The match is successful, then it is crude tyre to prove the tire.
2. method for anti-counterfeit according to claim 1, it is characterised in that:Described outside coating is spread by means of external force.
3. method for anti-counterfeit according to claim 1, it is characterised in that:Using needing to enter image before image characteristics extraction algorithm
Row pretreatment, be specifically:
By the random grain structural images of generation and the first-order partial derivative convolution of two-dimensional Gaussian function, original image is set to reach smoothly
Effect;
Output image is converted to gray level image after Gaussian smoothing, is tried to achieve by maximum variance between clusters between foreground and background most
Good segmentation threshold, bianry image is converted gray images into according to the threshold value;
Mathematical morphology erosion operation several times is carried out to obtained bianry image, is significantly less than with removing area in bianry image
The ambient noise in random grain region;
Random grain region area size is recovered by mathematical morphology dilation operation several times;
Area-of-interest in bianry image is oriented with minimum bounding box algorithm and by the color texture figure of area-of-interest
As output.
4. method for anti-counterfeit according to claim 1, it is characterised in that:The feature that described image characteristics extraction algorithm is extracted
Point has consistency to yardstick and rotation.
5. method for anti-counterfeit according to claim 4, it is characterised in that:Described image characteristics extraction algorithm, which is extracted, to be used
SIFT algorithms.
6. method for anti-counterfeit according to claim 1, it is characterised in that:Random character code is with verifying that the similarity of condition code is sentenced
It is disconnected to use Euclidean distance ratio.
7. method for anti-counterfeit according to claim 6, it is characterised in that:Match point around accumulative candidate matches point is in part
Principal direction and the contribution on to the candidate matches point, and judge of tire texture image to be verified and random grain image
The degree of correlation and the local angle degree of correlation are adjusted the distance whether in threshold range with point, if in threshold range;It is then correct matching
Point, otherwise abandons the candidate matches point.
8. the method for anti-counterfeit according to any one of claim 1-7, it is characterised in that:Described outside coating is using print
Oil.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710146621.4A CN106991419A (en) | 2017-03-13 | 2017-03-13 | Method for anti-counterfeit based on tire inner wall random grain |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710146621.4A CN106991419A (en) | 2017-03-13 | 2017-03-13 | Method for anti-counterfeit based on tire inner wall random grain |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106991419A true CN106991419A (en) | 2017-07-28 |
Family
ID=59412108
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710146621.4A Pending CN106991419A (en) | 2017-03-13 | 2017-03-13 | Method for anti-counterfeit based on tire inner wall random grain |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106991419A (en) |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108846681A (en) * | 2018-05-30 | 2018-11-20 | 于东升 | For the method for anti-counterfeit and device of woodwork, anti-fake traceability system |
CN108876402A (en) * | 2018-05-30 | 2018-11-23 | 于东升 | For the method for anti-counterfeit and device of leather and fur products, anti-fake traceability system |
CN108960849A (en) * | 2018-05-30 | 2018-12-07 | 于东升 | For the method for anti-counterfeit and device of paper products, anti-fake traceability system |
CN110264556A (en) * | 2019-06-10 | 2019-09-20 | 张慧 | A kind of generation method without the random complex texture of repetition |
CN110378425A (en) * | 2019-07-23 | 2019-10-25 | 北京隆普智能科技有限公司 | A kind of method and its system that intelligent image compares |
CN110942076A (en) * | 2019-11-27 | 2020-03-31 | 清华大学 | Method and system for generating anti-counterfeiting mark of ceramic product |
CN113793337A (en) * | 2021-11-18 | 2021-12-14 | 汶上海纬机车配件有限公司 | Locomotive accessory surface abnormal degree evaluation method based on artificial intelligence |
CN113989628A (en) * | 2021-10-27 | 2022-01-28 | 哈尔滨工程大学 | Underwater signal lamp positioning method based on weak direction gradient |
CN114757317A (en) * | 2022-06-16 | 2022-07-15 | 武汉朗修科技有限公司 | Method for making and verifying anti-counterfeiting texture pattern |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1794303A (en) * | 2005-12-29 | 2006-06-28 | 兆日科技(深圳)有限公司 | Antiforge method of abstracting using random distributed fiber image characteristics |
CN101577069A (en) * | 2008-05-09 | 2009-11-11 | 北京中科诚通科技发展有限公司 | Random texture anti-fake product, anti-fake method and mobile phone distinguishing method |
CN101582162A (en) * | 2008-05-14 | 2009-11-18 | 上海锦渡信息科技有限公司 | Virtu identifying method based on texture analysis |
CN102254153A (en) * | 2011-06-27 | 2011-11-23 | 常州市明磊橡塑制品有限公司 | Random texture anti-counterfeiting method |
CN102354410A (en) * | 2011-05-31 | 2012-02-15 | 上海傲卓防伪材料技术有限公司 | Random-texture anti-counterfeit method, system thereof and discriminator |
EP2485178A1 (en) * | 2009-09-28 | 2012-08-08 | Shanghai KOS Security Paper Technology Co., Ltd. | Anti-counterfeit method for random texture and recognizer thereof |
CN104217221A (en) * | 2014-08-27 | 2014-12-17 | 重庆大学 | Method for detecting calligraphy and paintings based on textural features |
CN104464514A (en) * | 2013-09-20 | 2015-03-25 | 李勇 | Novel motorcycle tire anti-counterfeiting device |
CN106096971A (en) * | 2015-05-01 | 2016-11-09 | 海南亚元防伪技术研究所(普通合伙) | Ink is along method for anti-counterfeit and ink along anti-fake network identification system |
CN106327534A (en) * | 2016-08-31 | 2017-01-11 | 杭州沃朴物联科技有限公司 | Tire inner wall texture identification method based on locating block |
-
2017
- 2017-03-13 CN CN201710146621.4A patent/CN106991419A/en active Pending
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1794303A (en) * | 2005-12-29 | 2006-06-28 | 兆日科技(深圳)有限公司 | Antiforge method of abstracting using random distributed fiber image characteristics |
CN101577069A (en) * | 2008-05-09 | 2009-11-11 | 北京中科诚通科技发展有限公司 | Random texture anti-fake product, anti-fake method and mobile phone distinguishing method |
CN101582162A (en) * | 2008-05-14 | 2009-11-18 | 上海锦渡信息科技有限公司 | Virtu identifying method based on texture analysis |
EP2485178A1 (en) * | 2009-09-28 | 2012-08-08 | Shanghai KOS Security Paper Technology Co., Ltd. | Anti-counterfeit method for random texture and recognizer thereof |
CN102354410A (en) * | 2011-05-31 | 2012-02-15 | 上海傲卓防伪材料技术有限公司 | Random-texture anti-counterfeit method, system thereof and discriminator |
CN102254153A (en) * | 2011-06-27 | 2011-11-23 | 常州市明磊橡塑制品有限公司 | Random texture anti-counterfeiting method |
CN104464514A (en) * | 2013-09-20 | 2015-03-25 | 李勇 | Novel motorcycle tire anti-counterfeiting device |
CN104217221A (en) * | 2014-08-27 | 2014-12-17 | 重庆大学 | Method for detecting calligraphy and paintings based on textural features |
CN106096971A (en) * | 2015-05-01 | 2016-11-09 | 海南亚元防伪技术研究所(普通合伙) | Ink is along method for anti-counterfeit and ink along anti-fake network identification system |
CN106327534A (en) * | 2016-08-31 | 2017-01-11 | 杭州沃朴物联科技有限公司 | Tire inner wall texture identification method based on locating block |
Non-Patent Citations (4)
Title |
---|
孙俊等: ""汽车轮胎号识别中的预处理问题"", 《江苏大学学报(自然科学版)》 * |
洪霞等: ""基于二维最大熵阈值分割的SIFT图像匹配算法"", 《半导体光电》 * |
胡小青等: ""基于邻域投票和Harris-SIFT特征的低空遥感影像匹配"", 《测绘工程》 * |
陈永常: ""纹理防伪技术的原理与特点"", 《包装技术》 * |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108846681A (en) * | 2018-05-30 | 2018-11-20 | 于东升 | For the method for anti-counterfeit and device of woodwork, anti-fake traceability system |
CN108876402A (en) * | 2018-05-30 | 2018-11-23 | 于东升 | For the method for anti-counterfeit and device of leather and fur products, anti-fake traceability system |
CN108960849A (en) * | 2018-05-30 | 2018-12-07 | 于东升 | For the method for anti-counterfeit and device of paper products, anti-fake traceability system |
CN110264556A (en) * | 2019-06-10 | 2019-09-20 | 张慧 | A kind of generation method without the random complex texture of repetition |
CN110378425A (en) * | 2019-07-23 | 2019-10-25 | 北京隆普智能科技有限公司 | A kind of method and its system that intelligent image compares |
CN110942076A (en) * | 2019-11-27 | 2020-03-31 | 清华大学 | Method and system for generating anti-counterfeiting mark of ceramic product |
CN110942076B (en) * | 2019-11-27 | 2020-10-16 | 清华大学 | Method and system for generating anti-counterfeiting mark of ceramic product |
CN113989628A (en) * | 2021-10-27 | 2022-01-28 | 哈尔滨工程大学 | Underwater signal lamp positioning method based on weak direction gradient |
CN113793337A (en) * | 2021-11-18 | 2021-12-14 | 汶上海纬机车配件有限公司 | Locomotive accessory surface abnormal degree evaluation method based on artificial intelligence |
CN113793337B (en) * | 2021-11-18 | 2022-02-08 | 汶上海纬机车配件有限公司 | Locomotive accessory surface abnormal degree evaluation method based on artificial intelligence |
CN114757317A (en) * | 2022-06-16 | 2022-07-15 | 武汉朗修科技有限公司 | Method for making and verifying anti-counterfeiting texture pattern |
CN114757317B (en) * | 2022-06-16 | 2022-09-27 | 武汉朗修科技有限公司 | Method for making and verifying anti-fake grain pattern |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106991419A (en) | Method for anti-counterfeit based on tire inner wall random grain | |
CN107545239B (en) | Fake plate detection method based on license plate recognition and vehicle characteristic matching | |
Luvizon et al. | A video-based system for vehicle speed measurement in urban roadways | |
US9014432B2 (en) | License plate character segmentation using likelihood maximization | |
CN103605953B (en) | Vehicle interest target detection method based on sliding window search | |
CN112686812B (en) | Bank card inclination correction detection method and device, readable storage medium and terminal | |
CN101859382B (en) | License plate detection and identification method based on maximum stable extremal region | |
CN106683119B (en) | Moving vehicle detection method based on aerial video image | |
CN108268867B (en) | License plate positioning method and device | |
CN103699905B (en) | Method and device for positioning license plate | |
CN106610969A (en) | Multimodal information-based video content auditing system and method | |
Alam et al. | Indian traffic sign detection and recognition | |
CN109740478A (en) | Vehicle detection and recognition methods, device, computer equipment and readable storage medium storing program for executing | |
Türkyılmaz et al. | License plate recognition system using artificial neural networks | |
CN104166841A (en) | Rapid detection identification method for specified pedestrian or vehicle in video monitoring network | |
CN104933398B (en) | vehicle identification system and method | |
CN104182973A (en) | Image copying and pasting detection method based on circular description operator CSIFT (Colored scale invariant feature transform) | |
CN108491498A (en) | A kind of bayonet image object searching method based on multiple features detection | |
CN103745197B (en) | A kind of detection method of license plate and device | |
Jagtap et al. | Multi-style license plate recognition using artificial neural network for Indian vehicles | |
CN115244542A (en) | Method and device for verifying authenticity of product | |
CN111126393A (en) | Vehicle appearance refitting judgment method and device, computer equipment and storage medium | |
CN107578011A (en) | The decision method and device of key frame of video | |
Lin et al. | Convolutional neural networks for face anti-spoofing and liveness detection | |
Tao et al. | Smoke vehicle detection based on spatiotemporal bag-of-features and professional convolutional neural network |
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 | ||
TA01 | Transfer of patent application right |
Effective date of registration: 20180925 Address after: 311400 626, 6 floor, 7 floor, Silver Lake innovation center, 9 Fu Fu Road, Fuyang District, Hangzhou, Zhejiang. Applicant after: Good fast auto parts (Hangzhou) Co., Ltd. Address before: 311400 Hangzhou, Fuyang, Zhejiang province 9 Yin Hu Street, No. 9, No. 9, six, floor 610, innovation center. Applicant before: Turvey round network technology (Hangzhou) Co., Ltd. |
|
TA01 | Transfer of patent application right | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170728 |
|
RJ01 | Rejection of invention patent application after publication |