CN109993739A - A kind of seal distinguishing method between true and false and device - Google Patents
A kind of seal distinguishing method between true and false and device Download PDFInfo
- Publication number
- CN109993739A CN109993739A CN201910255966.2A CN201910255966A CN109993739A CN 109993739 A CN109993739 A CN 109993739A CN 201910255966 A CN201910255966 A CN 201910255966A CN 109993739 A CN109993739 A CN 109993739A
- Authority
- CN
- China
- Prior art keywords
- seal
- image
- binaryzation
- polygon
- chapter
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
- G06T7/001—Industrial image inspection using an image reference approach
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10024—Color image
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
- Collating Specific Patterns (AREA)
Abstract
The present invention relates to the method and devices that a kind of seal true and false identifies, this method comprises: obtaining polygon seal image, default polygon chapter mould image and chapter modulus block domain information to be identified;Pretreatment generates binaryzation seal image and seal location information, regenerate the binaryzation seal image of scaling and the RGB seal image of dividing processing generation and the horizontal linear for extracting RGB seal image, seal frame and chapter mould frame are obtained again, and rotating seal image, then obtain binaryzation rotating seal image;And by binaryzation rotating seal image and default polygon chapter mould images match, maximum match parameter is generated, if maximum match parameter is greater than preset matching threshold, is true seal, is otherwise dummy seal.Whether use above-mentioned polygon seal distinguishing method between true and false to realize can accurately identify seal when identifying polygon seal for true seal, improve the accuracy that true seal is examined, and to the accurate rejection of the seal image of dummy seal, improve the practicability of seal identification.
Description
Technical field
The present invention relates to Digital Image Processing and identification technology field, and in particular to a kind of seal distinguishing method between true and false and dress
It sets.
Background technique
With the development with scientific and technological level, production technique and the technique of modern society's seal are substantially improved, in seal
The forgery technology of seal also have developed rapidly while production technique improves, and the seal forged can then cause social development seriously
Loss, it is therefore, imperative to the identification of the seal true and false.
Chop current is divided into circular stamp, oval seal, rectangular seal etc., traditional manual folding auxiliary people by shape
The seal distinguishing method between true and false of eye can not adapt to the development of modern society, and the method for manually identifying seal is time-consuming takes
It is not high to identify accuracy for power.Therefore with the continuous development of image procossing and identification technology and computer hardware, the seal true and false is certainly
Dynamic identification is widely used, but chop current automatic identification method is only to identify the size on machine to seal, shape in seal
Shape, content are simply matched, and accuracy is lower, can not efficient detection go out true and false seal, and research seal is automatic both at home and abroad
The mechanism of identification algorithm is simultaneously few, although there are many professional person also carry out seal discrimination method research, all be sound out
The research of property, practicability is not high, and to polygon, and identification accuracy rate when rectangular seal identifies true seal is lower, and dummy seal has
It is the phenomenon that missing inspection, not very practical.
Summary of the invention
In view of this, it is an object of the invention to overcome the deficiencies of the prior art and provide a kind of seal distinguishing method between true and false
With device, solve in the prior art when carrying out the identification of the seal true and false, identification accuracy is lower, to satisfactory true seal
It is low to identify percent of pass, the problem of practicability not high problem.
In order to achieve the above object, the present invention adopts the following technical scheme:
A kind of seal distinguishing method between true and false, comprising:
Obtain polygon seal image to be identified and default polygon chapter mould image;
Obtain the chapter modulus block domain information of the default polygon chapter mould image;
Binary conversion treatment is carried out to the polygon seal image to be identified, generates binaryzation seal image;
The seal location information of polygon seal image to be identified described in coarse localization;
Processing is split to the polygon seal image to be identified according to the seal location information, generates RGB color
Color mode seal image;
The binaryzation seal image is zoomed in and out, the binaryzation seal image of scaling is generated;
The largest connected region for extracting the binaryzation seal image of the scaling obtains the polygon seal figure to be identified
The seal frame of picture;And the chapter mould frame of the default polygon chapter mould image is obtained according to chapter modulus block domain information;
The seal frame and the binaryzation seal image are synthesized, generates synthesisization seal image, and obtain the conjunction
At the mass center for changing seal image;
Extract the horizontal linear in the rgb color mode seal image;
The horizontal linear and the chapter mould frame of horizontal direction are subjected to angle automatching, obtain the inclination of horizontal direction
Angle;According to the inclination angle, point rotates the rgb color mode seal image centered on the mass center, generates angular registration
Rotating seal image;
Binary conversion treatment is carried out to the rotating seal image of the angular registration, generates binaryzation rotating seal image;
Image information matching is carried out to the binaryzation rotating seal image and the default polygon chapter mould image, is obtained
Maximum match parameter;
Judge whether the maximum match parameter is greater than preset matching threshold, if the maximum match parameter is greater than described
Preset matching threshold then identifies that the polygon seal to be identified is true seal, is otherwise dummy seal.
Optionally, the chapter modulus block domain information for obtaining default polygon chapter mould image, comprising:
Binary conversion treatment is carried out to the default polygon chapter mould image, obtains binaryzation chapter mould image;
Positive and negative to binaryzation chapter mould image progress to deposit processing, generation is just depositing binaryzation chapter mould image and is depositing binaryzation with counter
Chapter mould image;
To it is described just depositing binaryzation chapter mould image and it is counter deposit binaryzation chapter mould image carry out numeric area segmentation generation described in
Chapter modulus block domain information.
Optionally, the seal location information of the positioning polygon seal image to be identified, comprising: extract described two
The largest connected region of the seal of value seal image, the largest connected region of seal are the seal location information.
Optionally, the rotating seal image to the angular registration carries out binary conversion treatment are as follows: uses hsv color mould
Type algorithm carries out binary conversion treatment to the rotating seal image of the angular registration.
Optionally, further includes: the binaryzation rotating seal image is accurately positioned, the binaryzation rotation is obtained
The true center position coordinates of seal image;
The each side for cutting the binaryzation rotating seal image obtains the printed text image of seal, and the seal is printed
Texts and pictures picture and the default polygon chapter mould image carry out image alignment.
Optionally, the maximum match parameter includes: seal width maximum matching rate, seal pixel maximum matching rate, print
Chapter numeric area maximum matching rate and anti-counterfeit of seals line maximum matching rate;
It is described that image information matching is carried out to the binaryzation rotating seal image and the default polygon chapter mould image,
Obtain maximum match parameter, comprising:
The matching of seal width is carried out to the binaryzation rotating seal image and the default polygon chapter mould image, is obtained
The seal width maximum matching rate;
Seal pixel matching is carried out to the binaryzation rotating seal image and the default polygon chapter mould image, is obtained
The seal pixel maximum matching rate;
Seal digital coding region is carried out to the binaryzation rotating seal image and the default polygon chapter mould image
Matching, obtains the seal numeric area maximum matching rate;
Anti-counterfeit of seals line region is carried out to the binaryzation rotating seal image and the default polygon chapter mould image
Match, obtains the anti-counterfeit of seals line maximum matching rate.
A kind of seal true and false identification device, comprising: first obtain module, for obtain polygon seal image to be identified with
Default polygon chapter mould image;
Second obtains module, for extracting the chapter modulus block domain information of the default polygon chapter mould image;
First binary processing module is generated for carrying out binary conversion treatment to the polygon seal image to be identified
Binaryzation seal image;
Coarse localization module is positioned for the seal location information to the polygon seal image to be identified;
Dividing processing module, for being divided according to the seal location information the polygon seal image to be identified
Processing is cut, rgb color mode seal image is generated;
Zoom module generates the binaryzation seal image of scaling for zooming in and out to the binaryzation seal image;
First extraction module, the largest connected region of the binaryzation seal image for extracting the scaling obtain described
The seal frame of polygon seal image to be identified;And the default polygon chapter is obtained according to chapter modulus block domain information
The chapter mould frame of mould image;
Synthesis module generates the seal figure of synthesisization for synthesizing the seal frame and the binaryzation seal image
Picture, and obtain the mass center of the synthesisization seal image;
Second extraction module, for extracting the horizontal linear of the rgb color mode seal image;
Angular registration module, for the chapter mould frame of the horizontal linear and horizontal direction to be carried out angle automatching,
The inclination angle of horizontal direction is obtained, and according to the inclination angle, point rotates the rgb color mode print centered on the mass center
Chapter image generates the rotating seal image of angular registration;
Second binary processing module carries out binary conversion treatment for the rotating seal image to the angular registration, raw
At binaryzation rotating seal image;
Image information matching module, for the binaryzation rotating seal image and the default polygon chapter mould image
Image information matching is carried out, maximum match parameter is obtained;
Judgment module, for judging whether the maximum match parameter is greater than preset matching threshold, if maximum described
It is greater than the preset matching threshold with parameter, then identifies that the polygon seal to be identified is true seal, be otherwise dummy seal.
Optionally, the second acquisition module includes: binary conversion treatment submodule, positive and negative to deposit processing submodule, digital block
Regional partition submodule;
The binary conversion treatment submodule is obtained for carrying out binary conversion treatment to the default polygon chapter mould image
Binaryzation chapter mould image;
It is described it is positive and negative deposit processing submodule, for the binaryzation chapter mould image carry out it is positive and negative deposit processing, generate and just deposit
Binaryzation chapter mould image deposits binaryzation chapter mould image with counter;
The numeric area divides submodule, for just depositing binaryzation chapter mould image to described and counter depositing binaryzation chapter mould figure
As carrying out numeric area segmentation, chapter modulus block domain information is generated.
Optionally, further includes: precise positioning module cuts module;
The precise positioning module obtains described two for carrying out precise positioning to the binaryzation rotating seal image
The true center position coordinates of value rotating seal image;
The cutting module obtains the printed text figure of seal for cutting each side of the binaryzation rotating seal image
Picture, and the printed text image of the seal and the default polygon chapter mould image are subjected to image alignment.
Optionally, the maximum match parameter includes: seal width maximum matching rate, seal pixel maximum matching rate, print
Chapter numeric area maximum matching rate and anti-counterfeit of seals line maximum matching rate;
Described image information matches module, comprising:
Seal width matched sub-block, for the binaryzation rotating seal image and the default polygon chapter mould figure
As carrying out the matching of seal width, the seal width maximum matching rate is obtained;
Seal pixel matching submodule, for the binaryzation rotating seal image and the default polygon chapter mould figure
As carrying out seal pixel matching, the seal pixel maximum matching rate is obtained;
Seal digital coding Region Matching submodule, for the binaryzation rotating seal image with it is described preset it is polygon
Shape chapter mould image carries out seal digital coding Region Matching, obtains the seal numeric area maximum matching rate;
Anti-counterfeit of seals line Region Matching submodule, for the binaryzation rotating seal image and the default polygon
Chapter mould image carries out anti-counterfeit of seals line Region Matching, obtains the anti-counterfeit of seals line maximum matching rate.
Technical solution provided by the present application can include the following benefits:
A kind of seal distinguishing method between true and false is disclosed in the present invention, comprising: obtain polygon seal image to be identified, default
Polygon chapter mould image and chapter modulus block domain information;Seal image is pre-processed, binaryzation seal image and seal position are generated
Information regenerates the binaryzation seal image of scaling and the RGB seal image of dividing processing generation and extracts RGB seal image
Horizontal linear, then extract the largest connected region of the binaryzation seal image of scaling, obtains seal frame and chapter mould frame, and with
Rotating seal image, binaryzation are put centered on the mass center for the seal image that synthesized seal frame and binaryzation seal image are beaten
The rotating seal image of process angle registration obtains binaryzation rotating seal image;And it by binaryzation rotating seal image and presets
Polygon chapter mould image is matched, and generates maximum match parameter, and compare the size of the match parameter Yu preset matching threshold value,
If maximum match parameter is greater than preset matching threshold, it is true seal, is otherwise dummy seal.It is true using above-mentioned polygon seal
Pseudo- discrimination method, which is realized, can accurately identify whether seal is true seal when identifying polygon seal, improve what true seal was examined
Accuracy, and to the accurate rejection of the seal image of dummy seal, improve the practicability of seal identification.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 is the method flow diagram that a kind of seal true and false that one embodiment of the application provides identifies;
Fig. 2 is the structure drawing of device that a kind of seal true and false that one embodiment of the application provides identifies.
Specific embodiment
To make the object, technical solutions and advantages of the present invention clearer, technical solution of the present invention will be carried out below
Detailed description.Obviously, described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.Base
Embodiment in the present invention, those of ordinary skill in the art are obtained all without making creative work
Other embodiment belongs to the range that the present invention is protected.
Fig. 1 is the method flow diagram that a kind of seal true and false that one embodiment of the application provides identifies, as shown in Figure 1, this reality
A kind of seal true and false for applying example identifies, comprising:
Step 101 obtains polygon seal image to be identified and default polygon chapter mould image.
In the present embodiment, polygon seal image to be identified and default polygon chapter mould image are obtained by scanner, and
The dots per inch (Dots Per Inch, DPI) of scanner are consistent with the DPI of default polygon chapter mould image, thus can
It determines the physical size of polygon seal image to be identified, and then just can determine the position of polygon seal image to be identified,
The pixel number of polygon seal image to be identified has so been determined that naturally, to determine that polygon seal figure to be identified
The physical size of picture.Therefore, the polygon seal image to be identified acquired is stored in default polygon chapter mould image and sets
In standby memory.
Step 102, the chapter modulus block domain information for obtaining default polygon chapter mould image.
In the present embodiment, the chapter modulus block domain information for presetting polygon chapter mould image is 13 digital block domain informations.And
When obtaining the chapter modulus block domain information, comprising: carry out binaryzation to default polygon chapter mould image according to colouring information first
Processing, obtains binaryzation chapter mould image, then to binaryzation chapter mould image carry out it is positive and negative deposit, generation just depositing binaryzation chapter mould image with
Counter to deposit binaryzation chapter mould image, positive and negative image of depositing carries out storage of turning upside down to image.Such as: to deposit " 69 " this viscous when positive and negative
Even when number, instrument will cause the inaccuracy of the segmentation to image, therefore can be with effective solution by turning upside down positive and negative figure of depositing
To the problem of image segmentation inaccuracy.Finally, being counted with anti-binaryzation chapter mould image of depositing to just depositing binaryzation chapter mould image again
Block regional partition generates chapter modulus block domain information, when carrying out numeric area segmentation using momentum water droplet algorithm.
The chapter modulus block domain information obtained by this step is stored in device memory, facilitates the matching of numeric area.
Step 103 pre-processes polygon seal image to be identified, generates binaryzation seal image and seal position
Information;
In the present embodiment, polygon seal image to be identified is pre-processed, generates binaryzation seal image and seal
Location information, comprising: binary conversion treatment is carried out to polygon seal image to be identified, generates binaryzation seal image;Treat mirror
Other polygon seal image carries out coarse localization processing, generates seal location information.Polygon to be identified is printed in the present embodiment
Binaryzation is carried out to seal image using hsv color model algorithm when chapter image carries out binary conversion treatment.Due to wait reflect
Other polygon seal image only has blue and red two kinds of images, so binarization segmentation is carried out using hsv color model algorithm,
The value range of tone H simultaneously, saturation degree S, brightness V are determined all in accordance with the specific situation that identifies.
In the present embodiment, the seal location information of coarse localization polygon seal image to be identified includes: to extract described two
The largest connected region of the seal of value seal image, the largest connected region of seal are the seal location information.Due to print
The largest connected region of chapter is seal frame region, thus just being capable of position seal location information.
Step 104 zooms in and out binaryzation seal image, generates the binaryzation seal image of scaling.
Step 105 is split processing to the polygon seal image to be identified according to seal location information, generates
Rgb color mode seal image.
Step 106, the largest connected region for extracting the binaryzation seal image scaled, obtain polygon seal figure to be identified
The seal frame of picture;And the chapter mould frame of default polygon chapter mould image is obtained according to chapter modulus block domain information.
In the present embodiment, the chapter of chapter mould image can be sought by the chapter mould frame of the default polygon chapter mould image of acquirement
The true border width of mould frame.
Horizontal linear in step 107, extraction rgb color mode seal image.
Step 108, synthesized seal frame and binaryzation seal image generate synthesisization seal image, and obtain synthesisization
The mass center of seal image;
It, need to be by polygon seal image to be identified when identifying the polygon seal image true and false to be identified in the present embodiment
It is aligned with default polygon chapter mould image, i.e., it needs to be determined that a central point, will could carry out therebetween pixel matching,
The mass center of synthesisization seal image in this central point, that is, the present embodiment.
Horizontal linear and the chapter mould frame of horizontal direction are carried out angle automatching by step 109, obtain the inclination of horizontal direction
Angle;According to inclination angle, point rotates the rgb color mode seal image centered on mass center, generates the rotation print of angular registration
Chapter image;
In the present embodiment, the width for taking the chapter mould frame of default polygon chapter mould image there is also a need for, due to polygon seal
Frame be polygon, therefore when the straight line that extract on polygon seal frame to be identified just can determine inclining for seal image
Rake angle rotates polygon seal image to be identified further according to tilt angle, when rotation to tilt angle is 0, generates
The rotating seal image of angular registration.
Step 110 carries out binary conversion treatment to the rotating seal image of angular registration, generates binaryzation rotating seal figure
Picture.
In the present embodiment, when carrying out to the rotating seal image of angular registration and know processing, using hsv color model
Algorithm carries out binary conversion treatment to the rotating seal image of the angular registration.
After generating binaryzation rotating seal image, precise positioning need to be carried out to binaryzation rotating seal image, obtain two
The true center point position coordinates of value rotating seal image.When determining center position coordinate, it need to rotate and print in binaryzation
The centre of chapter image selects 20*20 zonule, and centered on each point, starts to select at the coordinate of [- 5,5]
The most pixel of match point, the true center point position coordinates as binaryzation rotating seal image.Histogram is recycled to throw
Shadow cuts binaryzation rotating seal image, selects the maximum multiple sides of projected area, cuts binaryzation further according to above-mentioned multiple sideline judges
Rotating seal image, generate the only cutting of reservation seal here is exactly the printed text image for only retaining seal in fact, and seal is printed
Texts and pictures picture and the default polygon chapter mould image carry out image alignment.
Step 111 carries out image information matching to binaryzation rotating seal image and default polygon chapter mould image, obtains
Maximum match parameter.
In the present embodiment, maximum match parameter includes: seal width maximum matching rate, seal pixel maximum matching rate, print
Chapter numeric area maximum matching rate and anti-counterfeit of seals line maximum matching rate.Wherein, judge the width of binaryzation rotating seal image
Whether difference between the width of default chapter mould image is more than percent the 2 of national standard width, if exceeding, then it is assumed that wait reflect
Other polygon seal image is unqualified, is accredited as dummy seal.
Step 112 judges whether maximum match parameter is greater than preset matching threshold, presets if maximum match parameter is greater than
Matching threshold, then identify that polygon seal to be identified is true seal, be otherwise dummy seal.
In the present embodiment, after obtaining maximum match parameter, judge whether maximum match parameter is greater than preset matching
Threshold value identifies that polygon seal to be identified is true seal if maximum match parameter is greater than preset matching threshold, is otherwise false
Seal.Specifically, the seal width maximum matching rate 0.2 in maximum match parameter, the matching of seal pixel maximum ought be met simultaneously
Rate is greater than 0.9, seal numeric area maximum matching rate greater than 0.85 and when anti-counterfeit of seals line maximum matching rate is greater than 0.9, then reflects
Other polygon seal image to be identified carries out step 113, is accredited as true seal by identifying;Otherwise can not by identify, into
Row step 11, is accredited as dummy seal.
A kind of seal true and false mirror method for distinguishing through this embodiment, realizing can be accurate when identifying polygon seal
When whether identification seal is true seal, the inspection accuracy of true seal is improved, and can print to undesirable dummy seal
The accurate rejection of chapter image improves the practicability of seal identification.
Fig. 2 is the structure drawing of device that a kind of seal true and false that one embodiment of the application provides identifies, and referring to fig. 2, seal is true
The device that puppet identifies includes: that the first acquisition module 201, second obtains module 202, preprocessing module 203, Zoom module 204, divides
Cut processing module 205, the first extraction module 206, the second extraction module 207, synthesis module 208, angular registration module 209, two
Value processing module 210, image information matching module 211, judgment module 212;
Wherein, second acquisition module 202 include: the first binaryzation submodule 2021, it is positive and negative deposit processing submodule 2012, number
Word divides submodule 2013;
Preprocessing module 203 includes: the second binaryzation submodule 2031, coarse localization module 2032.
First obtains module 201, is connected with the first binary conversion treatment submodule 2021 in the second acquisition module 202, uses
Two-value is carried out in acquisition polygon seal image to be identified and default polygon chapter mould image, and to default polygon chapter mould image
Change processing, obtains binaryzation chapter mould image.It is positive and negative deposit processing submodule 2022 respectively with the first binary conversion treatment submodule 2021,
Numeric area divides submodule 2023 and is connected, and for receiving binaryzation chapter mould image, and carries out to binaryzation chapter mould image positive and negative
Processing is deposited, generation is just depositing binaryzation chapter mould image and depositing binaryzation chapter mould image with counter, and is sent to numeric area segmentation submodule;
Numeric area divides submodule 2023, is connected with the first extraction module 206, for just depositing binaryzation chapter mould image and counter deposit two
Value chapter mould image carries out numeric area segmentation, generates chapter modulus block domain information, and chapter modulus block domain information is sent out
It send into the first extraction module 206.
In the present embodiment, first obtain module 201, also with the second binary conversion treatment submodule in preprocessing module 203
2031 are connected, and the polygon seal image to be identified for will acquire is sent in the second binary conversion treatment submodule 2031.The
Two binaryzation submodules 2031 are connected with Zoom module 204, for that will carry out at binaryzation to polygon seal image to be identified
Reason generates binaryzation seal image, and binaryzation seal image is sent in Zoom module 204.And in the present embodiment, first
Module 201 is obtained, is also connected with coarse localization submodule 2032, the polygon seal image to be identified that will acquire is sent to roughly
In positioning submodule 2032;And coarse localization submodule 2032 is connected with dividing processing module 205, for polygon to be identified
The seal location information of seal image is positioned, and generates seal location information, then seal location information is sent to segmentation
Processing module 205.
In the present embodiment, Zoom module 204 is connected with the first extraction module 206, for the binaryzation seal image
It zooms in and out, generates the binaryzation seal image of scaling, and the binaryzation seal image of scaling is sent to the first extraction module
206。
In the present embodiment, dividing processing module 205 is connected with the second extraction module 207, for according to seal location information
Processing is split to polygon seal image to be identified, generates rgb color mode seal image, and rgb color mode is printed
Chapter image is sent in the second extraction module 207.
In the present embodiment, the first extraction module 206 is connected with synthesis module 208, for receiving the binaryzation seal of scaling
Image and extract scaling binaryzation seal image largest connected region, obtain the seal side of polygon seal image to be identified
Frame;And the chapter mould frame of default polygon chapter mould image is obtained according to chapter modulus block domain information, then seal frame is sent to
In synthesis module 208.
In the present embodiment, the second extraction module 207 is connected, for receiving rgb color with the angular registration module 209
Mode seal image, and the horizontal linear of rgb color mode seal image is extracted, and horizontal linear is sent to angular registration mould
In block 209.
In the present embodiment, synthesis module 208 is connected with the angular registration submodule 209, for synthesized seal frame with
Binaryzation seal image generates the seal image of synthesisization, and obtains the mass center of synthesisization seal image, and by synthesisization seal
The mass center of image is sent to angular registration submodule 209.
In the present embodiment, angular registration module 209 is connected with binary processing module 210, is used for horizontal linear and water
Square to chapter mould frame carry out angle automatching, obtain the inclination angle of horizontal direction, and according to inclination angle, the point centered on mass center
Rgb color mode seal image is rotated, generates the rotating seal image of angular registration, and by the rotating seal image of angular registration
It is sent to 210 module of binary conversion treatment.
In the present embodiment, binary processing module 210 is connected, for receiving angle with image information matching module 211
After the rotating seal image of registration, binary conversion treatment is carried out to the rotating seal image of angular registration, generates binaryzation rotation print
Chapter image, and binaryzation rotating seal image is sent to image information matching module 211.
In the present embodiment, image information matching module 211 is connected, for receiving binaryzation with the judgment module 212
Rotating seal image, and binaryzation rotating seal image and default polygon chapter mould image are subjected to image information matching, it obtains
Maximum match parameter, and maximum match parameter is sent in the judgment module 212.In the present embodiment, maximum match parameter
It include: seal width maximum matching rate, seal pixel maximum matching rate, seal numeric area maximum matching rate and anti-counterfeit of seals line
Maximum matching rate;And image information matching module 211, comprising:
(1) seal width matched sub-block, for binaryzation rotating seal image and default polygon chapter mould image into
The matching of row seal width, obtains seal width maximum matching rate;
(2) seal pixel matching submodule, for the binaryzation rotating seal image and the default polygon chapter
Mould image carries out seal pixel matching, obtains seal pixel maximum matching rate;
(3) seal digital coding Region Matching submodule, for binaryzation rotating seal image and default polygon chapter
Mould image carries out seal digital coding Region Matching, obtains silver-colored small cup numeric area maximum matching rate;
(4) anti-counterfeit of seals line Region Matching submodule, for binaryzation rotating seal image and the default polygon
Chapter mould image carries out anti-counterfeit of seals line Region Matching, obtains anti-counterfeit of seals line maximum matching rate
In the present embodiment, after obtaining maximum match parameter, judge whether maximum match parameter is greater than preset matching
Threshold value identifies that polygon seal to be identified is true seal if maximum match parameter is greater than preset matching threshold, is otherwise false
Seal.Specifically, the seal width maximum matching rate 0.2 in maximum match parameter, the matching of seal pixel maximum ought be met simultaneously
Rate is greater than 0.9, seal numeric area maximum matching rate greater than 0.85 and when anti-counterfeit of seals line maximum matching rate is greater than 0.9, then reflects
Other polygon seal image to be identified is accredited as true seal by identifying;Otherwise dummy seal can not be accredited as by identifying.
The device that a kind of seal true and false through this embodiment identifies can be more convenient and accurately realize what seal identified
Overall Steps have higher practicability, improve the accuracy for identifying true seal, and can print to undesirable dummy seal
The accurate rejection of chapter image improves the practicability of seal identification.
It is understood that same or similar part can mutually refer in the various embodiments described above, in some detailed implementations
The content of example explanation may refer to the same or similar content in other embodiments.
It should be noted that in the description of the present invention, term " first ", " second " etc. are used for description purposes only, without
It can be interpreted as indication or suggestion relative importance.In addition, in the description of the present invention, unless otherwise indicated, the meaning of " multiple "
Refer at least six.
Any process described otherwise above or method description are construed as in flow chart or herein, and expression includes
It is one or more for realizing specific logical function or process the step of executable instruction code module, segment or portion
Point, and the range of the preferred embodiment of the present invention includes other realization, wherein can not press shown or discussed suitable
Sequence, including according to related function by it is basic simultaneously in the way of or in the opposite order, Lai Zhihang function, this should be of the invention
Embodiment person of ordinary skill in the field understood.
Those skilled in the art are understood that realize all or part of step that above-described embodiment method carries
It suddenly is that relevant hardware can be instructed to complete by program, the program can store in a kind of computer-readable storage medium
In matter, which when being executed, includes the steps that one or a combination set of embodiment of the method.
In the description of this specification, reference term " one embodiment ", " some embodiments ", " example ", " specifically show
The description of example " or " some examples " etc. means specific features, structure, material or spy described in conjunction with this embodiment or example
Point is included at least one embodiment or example of the invention.In the present specification, schematic expression of the above terms are not
Centainly refer to identical embodiment or example.Moreover, particular features, structures, materials, or characteristics described can be any
One or more embodiment or examples in can be combined in any suitable manner.
Although the embodiments of the present invention has been shown and described above, it is to be understood that above-described embodiment is example
Property, it is not considered as limiting the invention, those skilled in the art within the scope of the invention can be to above-mentioned
Embodiment is changed, modifies, replacement and variant.
Claims (10)
1. a kind of seal distinguishing method between true and false characterized by comprising
Obtain polygon seal image to be identified and default polygon chapter mould image;
Obtain the chapter modulus block domain information of the default polygon chapter mould image;
The polygon seal image to be identified is pre-processed, binaryzation seal image and seal location information are generated;
The binaryzation seal image is zoomed in and out, the binaryzation seal image of scaling is generated;
Processing is split to the polygon seal image to be identified according to the seal location information, generates rgb color mould
Formula seal image;
The largest connected region for extracting the binaryzation seal image of the scaling obtains the polygon seal image to be identified
Seal frame;And the chapter mould frame of the default polygon chapter mould image is obtained according to chapter modulus block domain information;
Extract the horizontal linear in the rgb color mode seal image;
The seal frame and the binaryzation seal image are synthesized, generates synthesisization seal image, and obtain the synthesisization
The mass center of seal image;
The horizontal linear and the chapter mould frame of horizontal direction are subjected to angle automatching, obtain the inclination angle of horizontal direction;
According to the inclination angle, point rotates the rgb color mode seal image centered on the mass center, generates the rotation of angular registration
Turn seal image;
Binary conversion treatment is carried out to the rotating seal image of the angular registration, generates binaryzation rotating seal image;
Image information matching is carried out to the binaryzation rotating seal image and the default polygon chapter mould image, obtains maximum
Match parameter;
Judge whether the maximum match parameter is greater than preset matching threshold, if the maximum match parameter is greater than described preset
Matching threshold, then identify that the polygon seal to be identified is true seal, be otherwise dummy seal.
2. a kind of seal distinguishing method between true and false according to claim 1, which is characterized in that described to obtain default polygon chapter
The chapter modulus block domain information of mould image, comprising:
Binary conversion treatment is carried out to the default polygon chapter mould image, obtains binaryzation chapter mould image;
Positive and negative to binaryzation chapter mould image progress to deposit processing, generation is just depositing binaryzation chapter mould image and is depositing binaryzation Zhang Mo with counter
Image;
Binaryzation chapter mould image is just being deposited and anti-binaryzation chapter mould image of depositing carries out the numeric area segmentation generation Zhang Mo to described
Numeric area information.
3. a kind of seal distinguishing method between true and false according to claim 1, which is characterized in that wait reflect described in the coarse localization
The seal location information of other polygon seal image, comprising: the largest connected region of seal of the binaryzation seal image is extracted,
The largest connected region of seal is the seal location information.
4. a kind of seal distinguishing method between true and false according to claim 1, which is characterized in that described to described to be identified polygon
Shape seal image is pre-processed, and binaryzation seal image and seal location information are generated, comprising:
Binary conversion treatment is carried out to the polygon seal image to be identified, generates binaryzation seal image;
Coarse localization processing is carried out to the polygon seal image to be identified, generates seal location information.
5. a kind of seal distinguishing method between true and false according to claim 1, which is characterized in that further include: to the binaryzation
Rotating seal image is accurately positioned, and the true center point position coordinates of the binaryzation rotating seal image are obtained;
The each side for cutting the binaryzation rotating seal image, obtains the printed text image of seal, and by the seal impression figure
As carrying out image alignment with the default polygon chapter mould image.
6. a kind of seal distinguishing method between true and false according to claim 1, which is characterized in that the maximum match parameter packet
Include: seal width maximum matching rate, seal pixel maximum matching rate, seal numeric area maximum matching rate and anti-counterfeit of seals line are most
Big matching rate;
It is described that image information matching is carried out to the binaryzation rotating seal image and the default polygon chapter mould image, it obtains
Maximum match parameter, comprising:
The matching of seal width is carried out to the binaryzation rotating seal image and the default polygon chapter mould image, is obtained described
Seal width maximum matching rate;
Seal pixel matching is carried out to the binaryzation rotating seal image and the default polygon chapter mould image, is obtained described
Seal pixel maximum matching rate;
Seal digital coding Region Matching is carried out to the binaryzation rotating seal image and the default polygon chapter mould image,
Obtain the seal numeric area maximum matching rate;
Anti-counterfeit of seals line Region Matching is carried out to the binaryzation rotating seal image and the default polygon chapter mould image, is obtained
To the anti-counterfeit of seals line maximum matching rate.
7. a kind of seal true and false identification device characterized by comprising
First obtains module, for obtaining polygon seal image to be identified and default polygon chapter mould image;
Second obtains module, for extracting the chapter modulus block domain information of the default polygon chapter mould image;
Preprocessing module, for being pre-processed to the polygon seal image to be identified, generate binaryzation seal image with
Seal location information;
Zoom module generates the binaryzation seal image of scaling for zooming in and out to the binaryzation seal image;
Dividing processing module, for being split place to the polygon seal image to be identified according to the seal location information
Reason generates rgb color mode seal image;
First extraction module, the largest connected region of the binaryzation seal image for extracting the scaling obtain described wait reflect
The seal frame of other polygon seal image;And the default polygon chapter mould figure is obtained according to chapter modulus block domain information
The chapter mould frame of picture;
Second extraction module, for extracting the horizontal linear of the rgb color mode seal image;
Synthesis module generates the seal image of synthesisization for synthesizing the seal frame and the binaryzation seal image, and
Obtain the mass center of the synthesisization seal image;
Angular registration module is obtained for the chapter mould frame of the horizontal linear and horizontal direction to be carried out angle automatching
The inclination angle of horizontal direction, and according to the inclination angle, point rotates the rgb color mode seal figure centered on the mass center
Picture generates the rotating seal image of angular registration;
Binary processing module carries out binary conversion treatment for the rotating seal image to the angular registration, generates binaryzation
Rotating seal image;
Image information matching module, for being carried out to the binaryzation rotating seal image and the default polygon chapter mould image
Image information matching, obtains maximum match parameter;
Judgment module, for judging whether the maximum match parameter is greater than preset matching threshold, if the maximum matching ginseng
Number is greater than the preset matching threshold, then identifies that the polygon seal to be identified is true seal, be otherwise dummy seal.
8. a kind of seal true and false identification device according to claim 7, which is characterized in that described second obtains module packet
Include: the first binary conversion treatment submodule, positive and negative to deposit processing submodule, numeric area divides submodule;
The first binary conversion treatment submodule is obtained for carrying out binary conversion treatment to the default polygon chapter mould image
Binaryzation chapter mould image;
It is described it is positive and negative deposit processing submodule, for the binaryzation chapter mould image carry out it is positive and negative deposit processing, generate and just deposit two-value
Change chapter mould image and deposits binaryzation chapter mould image with counter;
The numeric area divides submodule, for it is described just depositing binaryzation chapter mould image and it is counter deposit binaryzation chapter mould image into
Line number block regional partition generates chapter modulus block domain information.
9. a kind of seal true and false identification device according to claim 1, which is characterized in that the preprocessing module includes:
Second binary conversion treatment submodule and coarse localization submodule;
The second binary conversion treatment submodule, it is raw for carrying out binary conversion treatment to the polygon seal image to be identified
At binaryzation seal image;
The coarse localization submodule is positioned for the seal location information to the polygon seal image to be identified,
And generate seal location information.
10. a kind of seal true and false identification device according to claim 1, which is characterized in that the maximum match parameter packet
Include: seal width maximum matching rate, seal pixel maximum matching rate, seal numeric area maximum matching rate and anti-counterfeit of seals line are most
Big matching rate;
Described image information matches module, comprising:
Seal width matched sub-block, for the binaryzation rotating seal image and the default polygon chapter mould image into
The matching of row seal width, obtains the seal width maximum matching rate;
Seal pixel matching submodule, for the binaryzation rotating seal image and the default polygon chapter mould image into
Row seal pixel matching obtains the seal pixel maximum matching rate;
Seal digital coding Region Matching submodule, for the binaryzation rotating seal image and the default polygon chapter
Mould image carries out seal digital coding Region Matching, obtains the seal numeric area maximum matching rate;
Anti-counterfeit of seals line Region Matching submodule, for the binaryzation rotating seal image and the default polygon Zhang Mo
Image carries out anti-counterfeit of seals line Region Matching, obtains the anti-counterfeit of seals line maximum matching rate.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910255966.2A CN109993739B (en) | 2019-03-29 | 2019-03-29 | Seal authenticity identification method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910255966.2A CN109993739B (en) | 2019-03-29 | 2019-03-29 | Seal authenticity identification method and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109993739A true CN109993739A (en) | 2019-07-09 |
CN109993739B CN109993739B (en) | 2023-04-28 |
Family
ID=67131871
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910255966.2A Active CN109993739B (en) | 2019-03-29 | 2019-03-29 | Seal authenticity identification method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109993739B (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110414525A (en) * | 2019-07-31 | 2019-11-05 | 新华三信息安全技术有限公司 | A kind of character identifying method and device |
CN112883974A (en) * | 2021-05-06 | 2021-06-01 | 江西省江咨金发数据科技发展有限公司 | Electronic letter identification system based on image verification |
CN113486901A (en) * | 2021-06-23 | 2021-10-08 | 内江师范学院 | Novel high-imitation round seal identification method |
CN114419655A (en) * | 2022-03-31 | 2022-04-29 | 北京惠朗时代科技有限公司 | Seal verification method, seal verification device, seal verification medium and seal verification equipment |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2005064524A1 (en) * | 2003-12-30 | 2005-07-14 | Ruizhen Liu | Digital seal system |
CN101008985A (en) * | 2007-01-18 | 2007-08-01 | 章毅 | Seal identification system and controlling method thereof |
WO2010126269A2 (en) * | 2009-04-27 | 2010-11-04 | 대한민국(관리부서:행정안전부 국립과학수사연구소장) | Method and recording medium for seal forgery inspection |
CN102136908A (en) * | 2010-01-22 | 2011-07-27 | 张胤微 | Method for stamping electronic seal by using stamp and stamping device realized by same |
CN108509992A (en) * | 2017-12-27 | 2018-09-07 | 广西新章物电防伪技术开发有限公司 | A kind of seal impression Weigh sensor method for anti-counterfeit based on machine learning techniques |
-
2019
- 2019-03-29 CN CN201910255966.2A patent/CN109993739B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2005064524A1 (en) * | 2003-12-30 | 2005-07-14 | Ruizhen Liu | Digital seal system |
CN101008985A (en) * | 2007-01-18 | 2007-08-01 | 章毅 | Seal identification system and controlling method thereof |
WO2010126269A2 (en) * | 2009-04-27 | 2010-11-04 | 대한민국(관리부서:행정안전부 국립과학수사연구소장) | Method and recording medium for seal forgery inspection |
CN102136908A (en) * | 2010-01-22 | 2011-07-27 | 张胤微 | Method for stamping electronic seal by using stamp and stamping device realized by same |
CN108509992A (en) * | 2017-12-27 | 2018-09-07 | 广西新章物电防伪技术开发有限公司 | A kind of seal impression Weigh sensor method for anti-counterfeit based on machine learning techniques |
Non-Patent Citations (1)
Title |
---|
杨敏等: "印章与签名分离的图像处理算法研究", 《警察技术》 * |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110414525A (en) * | 2019-07-31 | 2019-11-05 | 新华三信息安全技术有限公司 | A kind of character identifying method and device |
CN110414525B (en) * | 2019-07-31 | 2022-02-25 | 新华三信息安全技术有限公司 | Character recognition method and device |
CN112883974A (en) * | 2021-05-06 | 2021-06-01 | 江西省江咨金发数据科技发展有限公司 | Electronic letter identification system based on image verification |
CN113486901A (en) * | 2021-06-23 | 2021-10-08 | 内江师范学院 | Novel high-imitation round seal identification method |
CN113486901B (en) * | 2021-06-23 | 2023-05-02 | 内江师范学院 | Novel high imitation round seal identification method |
CN114419655A (en) * | 2022-03-31 | 2022-04-29 | 北京惠朗时代科技有限公司 | Seal verification method, seal verification device, seal verification medium and seal verification equipment |
Also Published As
Publication number | Publication date |
---|---|
CN109993739B (en) | 2023-04-28 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109993739A (en) | A kind of seal distinguishing method between true and false and device | |
US10304166B2 (en) | Eye beautification under inaccurate localization | |
US20050129326A1 (en) | Image processing apparatus and print system | |
JP4335565B2 (en) | Method and apparatus for detecting and / or tracking one or more color regions in an image or sequence of images | |
CN105956515B (en) | Solid-EO-1 hyperion face identification method based on pole imaging | |
CN105825494B (en) | A kind of image processing method and mobile terminal | |
CN102930265B (en) | A kind of many I.D.s scan method and device | |
CN107705248A (en) | Image processing method, device, electronic equipment and computer-readable recording medium | |
CN102156849B (en) | Reading device and reading method of two-dimensional bar code marked on metal cylindrical surface directly | |
CN107358627B (en) | Fruit size detection method based on Kinect camera | |
TW200818861A (en) | Color scanning to enhance bitonal image | |
CN104299011A (en) | Skin type and skin problem identification and detection method based on facial image identification | |
CN106020745B (en) | 3D printing path generating method and device based on recognition of face | |
CN108133216A (en) | The charactron Recognition of Reading method that achievable decimal point based on machine vision is read | |
CN104517101A (en) | Game poker card recognition method based on pixel square difference matching | |
CN110378351A (en) | Seal discrimination method and device | |
CN109767436A (en) | A kind of method and device that the seal true and false identifies | |
CN102073872A (en) | Image-based method for identifying shape of parasite egg | |
CN110188590B (en) | Face shape distinguishing method based on three-dimensional face model | |
CN115082776A (en) | Electric energy meter automatic detection system and method based on image recognition | |
CN109583330A (en) | A kind of pore detection method for human face photo | |
CN110334647A (en) | A kind of parameter format method based on image recognition | |
CN109766850A (en) | Fingerprint image matching method based on Fusion Features | |
CN110335406B (en) | Multimedia glasses type portable currency detector | |
CN109871929A (en) | A kind of identity ring, recognition methods, identification device and identifying system |
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: 20230413 Address after: 518000 419, Building 4, Software Industry Base, No. 19, 17 and 18, Haitian 1st Road, Binhai Community, Yuehai Street, Nanshan District, Shenzhen, Guangdong Province Applicant after: Shenzhen Anyin Technology Co.,Ltd. Address before: 518000 whole building of Qilin road production center, Nanshan District, Shenzhen City, Guangdong Province Applicant before: SHENZHEN CHUANGYE SIGNET INDUSTRIAL CO.,LTD. |
|
TA01 | Transfer of patent application right | ||
GR01 | Patent grant | ||
GR01 | Patent grant |