CN113378878B - Identity card counterfeit identification method and electronic equipment - Google Patents
Identity card counterfeit identification method and electronic equipment Download PDFInfo
- Publication number
- CN113378878B CN113378878B CN202110483713.8A CN202110483713A CN113378878B CN 113378878 B CN113378878 B CN 113378878B CN 202110483713 A CN202110483713 A CN 202110483713A CN 113378878 B CN113378878 B CN 113378878B
- Authority
- CN
- China
- Prior art keywords
- identity card
- hong kong
- image
- light condition
- kong identity
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Credit Cards Or The Like (AREA)
Abstract
The invention discloses an identity card authenticity identification method and electronic equipment, which are used for carrying out authenticity identification on a hong Kong identity card, wherein the identity card authenticity identification method comprises the following steps: acquiring images of a certificate to be authenticated under the conditions of white light, ultraviolet light and infrared light respectively; judging whether the certificate to be authenticated is a hong Kong identity card or not according to the cercis chinensis pattern at a preset position by using a back white light condition image of the certificate to be authenticated; judging a new version and an old version of the hong Kong identity card according to a characteristic window of a preset position by using a front white light condition image of the hong Kong identity card; aiming at the new version of hong Kong identity card, the authenticity of the new version of hong Kong identity card is identified by utilizing the image characteristics of the front and back surfaces of the new version of hong Kong identity card under various light source conditions; and aiming at the old version of hong Kong identity card, the authenticity of the old version of hong Kong identity card is identified by utilizing the image characteristics of the front and back surfaces of the old version of hong Kong identity card under various light source conditions. The invention can quickly and accurately identify the identity card of hong Kong and is universal for new and old identity cards.
Description
Technical Field
The invention belongs to the technical field of certificate authenticity identification, and particularly relates to an identity card authenticity identification method and electronic equipment.
Background
Identity documents relate to a large number of information safety, economic safety and personal safety problems in daily life, and a large number of false personal identity cards are used for illegal purposes in the market, so that the authenticity of the identity documents must be authenticated firstly when businesses are developed in sensitive fields such as finance, securities and the like. At present, deep learning algorithms are mainly adopted in the market for the certificate authenticity identification technology, the deep learning algorithms need a large number of training samples, and if the deep learning algorithms are applied to the hong Kong identity card authenticity identification technology, a large number of certificate samples are needed, and personal identity information can be seriously revealed. Moreover, the traditional manual identification of the hong Kong identity card is low in authenticity efficiency, so that an automatic, rapid and efficient identification system suitable for the hong Kong identity card is needed.
Disclosure of Invention
The invention provides an identity card authenticity identification method and electronic equipment, which can solve the problem that the information of an identity card is leaked possibly caused by a supervised learning authenticity identification method based on a large batch of training samples and can quickly and accurately identify the authenticity of a hong Kong identity card.
In order to achieve the technical purpose, the invention adopts the following technical scheme:
an identity card authenticity identification method is used for carrying out authenticity identification on a hong Kong identity card and comprises the following steps:
step 1, acquiring images of a certificate to be authenticated under the conditions of white light, ultraviolet light and infrared light respectively;
step 2, judging whether the certificate to be authenticated is a hong Kong identity card or not according to the bauhinia pattern at a preset position by using a back white light condition image of the certificate to be authenticated;
step 4, aiming at the new version of hong Kong identity card, the authenticity of the new version of hong Kong identity card is identified by utilizing the image characteristics of the front and back surfaces of the new version of hong Kong identity card under various light source conditions; aiming at the old version of hong Kong identity card, the authenticity of the old version of hong Kong identity card is identified by utilizing the image characteristics of the front and back surfaces of the old version of hong Kong identity card under various light source conditions;
the image of the hong Kong identity card under the white light, ultraviolet light and infrared light conditions respectively has different image characteristics, part of the preset image characteristics are taken as the counterfeit identification criterion, if the selected image characteristics meet the counterfeit identification requirement, the hongKong identity card is judged to pass the counterfeit identification, otherwise, the hongKong identity card is judged not to pass the counterfeit identification; wherein:
all image characteristics of the new identity card counterfeit identification include: transparent window features, bauhinia pattern features, feature circle patterns, landscape pattern features, guilloche pattern features;
all image characteristics of the old identity card counterfeit identification include: bauhinia pattern features, rectangular area feature patterns, line feature patterns, rainbow feature patterns.
In a more preferred technical scheme, the step 3 specifically comprises the following steps: and matching the characteristic window at the preset position in the white light condition image on the back surface of the certificate to be authenticated by using an edge characteristic template, wherein if the matching is successful, the hong Kong identity card is a new hong Kong identity card, and otherwise, the hong Kong identity card is an old hong Kong identity card.
In a more preferred technical scheme, the step 4 of identifying the new version of hong Kong identity card includes: matching the front white light condition image and the front infrared light condition image of the hong Kong identity card of the new edition by using a front window template; matching the back white light condition image and the back infrared light condition image of the hong Kong identity card of the new edition by using a back window template; if each matching is successful in the corresponding preset position, the counterfeit identification requirement is met; otherwise, the false identification requirement is not met.
In a more preferred technical scheme, the step 4 of identifying the new version of hong Kong identity card includes: carrying out template matching on a back white light condition image, a back ultraviolet light condition image and a back infrared light condition image of the hong Kong identity card of the new edition by using a preset bauhinia pattern template; if the back white light condition image and the back ultraviolet light condition image are matched with a preset bauhinia pattern template at preset positions, the authenticity identification requirement is met; otherwise, the false identification requirement is not met;
in a more preferred technical scheme, the step 4 of identifying the old version of hong Kong identity card includes: performing template matching on a back white light condition image, a back ultraviolet light condition image and a back infrared light condition image of the old hong Kong identity card by using a preset bauhinia pattern template; if the back white light condition image and the back ultraviolet light condition image are matched with a preset bauhinia pattern template at preset positions, the authenticity identification requirement is met; otherwise, the false identification requirement is not met.
In a more preferred technical scheme, the step 4 of identifying the new version of hong Kong identity card includes:
sequentially carrying out gray scale inversion, binarization processing and connected domain analysis on the front ultraviolet light condition image of the hong Kong identity card of the new edition to obtain the coordinate position, width, height and accumulated pixel value of the largest connected domain;
calculating the radius r of the characteristic circle according to the width and height of the maximum connected domain, and calculating the accumulated pixel value of the maximum connected domain and the area pi r of the characteristic circle based on the obtained radius r 2 Difference between Δ S 1 ；
If the coordinate position of the maximum connected domain, the radius of the characteristic circle and the area difference Delta S of the characteristic circle 1 If the values are all in accordance with the corresponding preset threshold values, the authenticity identification requirements are met; otherwise, the false identification requirement is not met;
in a more preferred technical scheme, the step 4 of identifying the old version of hong Kong identity card includes:
sequentially carrying out gray value inversion, binarization processing, corrosion and expansion morphological operation and connected domain analysis on a back ultraviolet light condition image of an old version hong Kong identity card to obtain the position, width, height and accumulated pixel value of a maximum connected domain;
calculating the characteristic rectangular area w h of the maximum connected domain according to the width w and the height h of the maximum connected domain, and then calculating the difference value delta S between the accumulated pixel value of the maximum connected domain and the characteristic rectangular area w h 2 ；
If the position, width, height and characteristic rectangular area difference value delta S of the maximum connected domain 2 If the values are all in accordance with the corresponding preset threshold values, the authenticity identification requirements are met; otherwise, the false identification requirement is not met.
In a more preferred technical scheme, the step 4 of identifying the new version of hong Kong identity card includes:
sequentially carrying out gray scale inversion, binarization processing, corrosion and expansion morphological processing on the back ultraviolet light condition image of the hong Kong identity card of the new edition;
carrying out XOR logic operation on a preset landscape image template and the image obtained in the last step, and then carrying out connected domain analysis on the obtained image;
if the accumulated value of the pixels of the maximum connected domain is smaller than the corresponding preset threshold value, the requirement of identifying the fake is met; otherwise, the false identification requirement is not met;
in a more preferred technical scheme, the step 4 of identifying the old version of hong Kong identity card includes:
sequentially carrying out gray scale inversion and binarization processing on the front ultraviolet light condition image of the old version of hong Kong identity card to obtain an image comprising four sub-regions;
counting the black pixel ratio in each sub-area to be P in turn i1 ,P i2 ,P i3 ,P i4 ；
Calculating the similarity between the image and the corresponding preset line template image based on the Euclidean distance, and if the similarity is smaller than the corresponding preset threshold value, conforming to the counterfeit identification requirement; otherwise, the false identification requirement is not met.
In a more preferred technical scheme, the step 4 of identifying the new version of hong Kong identity card includes:
sampling a guilloche area in a front white light condition image of the hong Kong identity card of the new edition to obtain a guilloche image;
converting the guilloche image from RGB color space to HSV color space, and counting the pixel proportion P with blue chromatic value ib And a pixel ratio P with a chromatic value of red ir ；
Calculating the similarity between the guilloche images and the corresponding preset guilloche template images based on the Euclidean distance, and if the similarity is smaller than a corresponding preset threshold value, conforming to the counterfeit identification requirement; otherwise, the false identification requirement is not met.
In a more preferred technical scheme, the step 4 of identifying the old version of hong Kong identity card includes:
sampling a rainbow area in a white light condition image on the front surface of an old hong Kong identity card to obtain a rainbow image;
converting the rainbow image from an RGB color space into an HSV color space, and counting the distribution range of chromatic values and saturation of adjacent areas; comparing the difference of the chroma values of the adjacent areas with the difference of the saturation;
if the distribution range and the difference value of the chromatic value and the saturation of the adjacent area are within the corresponding preset threshold value range, the requirement of counterfeit discrimination is met; otherwise, the false identification requirement is not met.
An electronic device comprising a memory and a processor, wherein the memory stores a computer program, and wherein the computer program, when executed by the processor, causes the processor to implement any one of the methods described above.
Advantageous effects
The invention is based on the image of the hong Kong ID card under the white light, ultraviolet light and infrared light conditions, which has different image characteristics, and uses part of the image as the false distinguishing characteristic, thereby being capable of quickly and accurately distinguishing the truth of the hong Kong ID card and being suitable for the hong Kong ID cards of the prior new version and the prior old version. In addition, the problem of identity card information leakage possibly caused by a supervised learning authentication method based on a large batch of training samples in the prior art can be solved.
Drawings
FIG. 1 is a block diagram of the identity card authentication system of hong Kong of the present invention;
FIG. 2 is a back view of a portion of a white light, ultraviolet light, infrared light scanned version of the hong Kong ID card of the present invention;
FIG. 3 is a line and section view of the present invention taken from a front image of an old version of hong Kong ID card scanned by ultraviolet light;
FIG. 4 is a view of the ultraviolet light scanned image of the face of a new version of hong Kong identity card of the present invention;
FIG. 5 is a binary, morphologically processed image of the back of an old version of hong Kong ID card scanned by ultraviolet light in accordance with the present invention.
Detailed Description
The following describes embodiments of the present invention in detail, which are developed based on the technical solutions of the present invention, and give detailed implementation manners and specific operation procedures to further explain the technical solutions of the present invention.
Referring to fig. 1, the identity card authentication method provided in this embodiment is used for authenticating a hong kong identity card, and includes the following steps:
step 1, obtaining images of a certificate to be authenticated under the conditions of white light, ultraviolet light and infrared light respectively.
And 2, judging whether the certificate to be authenticated is the hong Kong identity card or not according to the bauhinia pattern at the preset position by using the back white light condition image of the certificate to be authenticated.
Before the hong Kong identity card is authenticated, whether the identity card to be authenticated is the hong Kong identity card or not is firstly confirmed. In the middle-upper position of back images in new and old versions of hong Kong ID card scanned by white light, the characteristics of the bauhinia pattern exist, and whether the certificate to be authenticated corresponding to the input image is the hong Kong ID card can be judged by detecting the bauhinia pattern. And (3) importing a pre-stored redbud pattern template to match with the image on the back side of the identity card, if the upper position in the image can be accurately matched successfully, the identity card is determined to be the hong Kong identity card, and the redbud pattern in the middle position on the back side of the hong Kong identity card of the new edition can be seen in the first graph in FIG. 2.
And 3, judging the new and old versions of the hong Kong identity card according to a characteristic window at a preset position by using the front white light condition image of the hong Kong identity card.
Because the new-version identification card and the old-version identification card of hong Kong have different image characteristics, before the identification card is identified, whether the hong Kong identification card is the new-version identification card or the old-version identification card needs to be confirmed.
The new version of hong Kong identity card has transparent window characteristics, so a preset window edge characteristic template can be used for respectively carrying out template matching on the front side of the identity card scanned by white light, when the input picture can be successfully matched with the edge characteristic window template, the input picture is judged to be the hong Kong identity card of the new version, and otherwise, the input picture is judged to be the hong Kong identity card of the old version. The presence of the window feature pattern for the new version of hong kong id card can be seen in fig. 2 (a) and (b).
Step 4, aiming at the new version of hong Kong identity card, the authenticity of the new version of hong Kong identity card is identified by utilizing the image characteristics of the front surface and the back surface under various light source conditions; and aiming at the old version of hong Kong identity card, the authenticity of the old version of hong Kong identity card is identified by utilizing the image characteristics of the front and back surfaces of the old version of hong Kong identity card under various light source conditions.
The images of the hong Kong identity card under the white light, ultraviolet light and infrared light conditions respectively have different image characteristics, so that the identity card to be authenticated can be authenticated by utilizing any one or more image characteristics. In different embodiments, a part of the image features may be preset as a counterfeit identification criterion, and if selected, if each preset image feature meets the counterfeit identification requirement, the image features are considered to pass the counterfeit identification, otherwise, the image features are considered to fail the counterfeit identification. In this embodiment, if all image features meet the authentication requirement, the authentication may be considered to be passed, otherwise, the authentication may not be passed, and a specific embodiment is shown in fig. 1, which includes:
first, new edition identity card counterfeit identification
1. Transparent window feature pattern
A transparent window characteristic pattern exists in the front white light condition image and the front infrared light condition image of the hong Kong identity card of the new edition; and the back white light condition image and the back infrared light condition image of the hong Kong identity card of the new edition have another transparent window characteristic pattern, and the shape of the transparent window characteristic pattern is different from that of the front transparent window characteristic pattern. Therefore, two different window templates can be used to match the front and back pictures scanned by two light sources respectively: matching the front white light condition image and the front infrared light condition image of the hong Kong identity card of the new edition by using a front window template; matching the back white light condition image and the back infrared light condition image of the hong Kong identity card of the new edition by using a back window template; if each matching is successful in the corresponding preset position, the counterfeit identification requirement of the characteristic pattern of the transparent window is met; otherwise, the anti-counterfeiting requirement of the transparent window characteristic pattern is not met.
2. Patterned features of bauhinia
In a new hong Kong identity card picture, the Chinese redbud patterns exist at the same positions of the white light and ultraviolet light scanning back pictures, but the patterns do not exist at the same positions of the infrared scanning images, and the preset Chinese redbud pattern template is used for carrying out template matching on the back scanning pictures of three light sources: carrying out template matching on a back white light condition image, a back ultraviolet light condition image and a back infrared light condition image of the hong Kong identity card of the new edition by using a preset bauhinia pattern template; if the back white light condition image and the back ultraviolet light condition image are matched with a preset bauhinia pattern template at preset positions, the authenticity identification requirement of the bauhinia pattern is met; otherwise, the false identification requirement of the bauhinia pattern is not met.
3. Characteristic circle pattern
In the new version of the hong Kong identity card ultraviolet scanning front picture, a bright circular area with a fixed size exists, so that a characteristic circle formed by the bright circular area can be used as a fake identification image characteristic, and the fake identification process of the characteristic circle pattern is as follows:
(1) firstly, carrying out gray value inversion on a back ultraviolet light condition image of a new version of hong Kong identity card, then carrying out binarization processing on the inverted image, extracting a circular area, and carrying out corrosion processing on the binary image to remove character information of the bright circular area because file information exists in the extracted circular area, and then carrying out expanded image morphological processing on the picture to recover the original size of the circular area to obtain a relatively clean circular area, as shown in fig. 4. Finally, carrying out connected domain analysis on the image, and determining the coordinate position, the width w, the height h and the accumulated pixel value S of the maximum connected domain circle ；
(2) Computing features from width and height of maximum connected domainRadius of circle
Calculating the accumulated pixel value S of the maximum connected domain based on the obtained radius r circle And characteristic circle area pi r 2 Difference between Δ S 1 ；
(3) If the coordinate position of the maximum connected domain, the radius of the characteristic circle and the area difference Delta S of the characteristic circle 1 If the true values are all in accordance with the corresponding preset threshold values, the false distinguishing requirements of the characteristic circle are met; otherwise, the false distinguishing requirement of the characteristic circle is not met.
Since the image has no other feature objects except for the feature circle, the accumulated pixel values of connected components existing at other positions except the maximum connected component are strictly limited to be less than the preset threshold.
4. Landscape pattern feature
In the new version of hong Kong identity card ultraviolet back scanning picture, a bright landscape pattern characteristic area exists, so the bright landscape pattern characteristic area can be used as a counterfeit identification image characteristic, and the counterfeit identification process based on the bright landscape pattern characteristic area is as follows:
(1) firstly, carrying out gray value reversal on a back ultraviolet light condition image of a new hong Kong identity card, then carrying out binarization processing on the reversed image, and extracting a landscape area, wherein the pattern area has more details, so that the binary image needs to be firstly corroded and then subjected to expanded image morphological processing;
(2) carrying out XOR logic operation on a preset landscape image template and the image obtained in the last step, and then carrying out connected domain analysis on the obtained image;
(3) if the accumulated value of the pixels of the maximum connected domain is smaller than the corresponding preset threshold value, the false distinguishing requirement of the landscape pattern characteristic is met; otherwise, the false distinguishing requirement of the landscape pattern characteristic is not met;
5. torsion cord pattern features
The red fine grid lines and the blue coarse grid lines of the image of the hong Kong identity card front face scanned by white light are wound to form a button cable image, so that a torsion cable pattern can be used as a fake identification image characteristic, and the fake identification process based on the torsion cable pattern is as follows:
sampling a guilloche area in a front white light condition image of the hong Kong identity card of the new edition to obtain a guilloche image;
converting the guilloche image from RGB color space to HSV color space, and counting the pixel proportion P with blue chromatic value ib And a pixel ratio P with a chromatic value of red ir ；
Calculating the similarity between the guilloche images and the corresponding preset guilloche template images based on the Euclidean distance, and if the similarity is smaller than a corresponding preset threshold value, conforming to the counterfeit identification requirement; otherwise, the false identification requirement is not met;
wherein, the calculation formula of the Euclidean distance is as follows:
in the formula, P tb And P ir Pixel ratios of blue and red chrominance values in the guilloche template image, respectively.
Identification of old and used identity cards
1. Patterned features of bauhinia
The same China redbud pattern as the Chinese redbud pattern on the back of the new hong Kong identity card, and the same China redbud pattern as the Chinese hong Kong identity card on the back of the old hong Kong identity card, therefore, the same method for identifying the fake of the new hong Kong identity card based on the Chinese redbud pattern can be adopted, and the fake of the old hong Kong identity card based on the Chinese redbud pattern is identified: performing template matching on a back white light condition image, a back ultraviolet light condition image and a back infrared light condition image of the old hong Kong identity card by using a preset bauhinia pattern template; if the back white light condition image and the back ultraviolet light condition image are matched with a preset cercis chinensis pattern template at a preset position, the authenticity identification requirement of the cercis chinensis pattern is met; otherwise, the false identification requirement of the bauhinia pattern is not met.
2. Rectangular region feature pattern
In the old version of ultraviolet back scanning picture, a bright rectangular area exists, so that the bright rectangular area characteristic pattern can be used as a fake identification image characteristic, and the fake identification process based on the bright rectangular area characteristic pattern is as follows:
(1) carrying out gray value inversion on a back ultraviolet light condition image of an old version of hong Kong identity card, then carrying out binarization processing on the inverted image, and in the process of extracting a rectangular area, considering that the rectangular area has file information, the binary image needs to be firstly corroded, so that the character information of the rectangular area can be removed, and then carrying out expanded image morphological processing on the picture, so that the original size of the rectangular area can be recovered, and a clean rectangular area can be obtained; finally, the obtained image is analyzed in a connected domain mode to obtain the position, the width w, the height h and the accumulated pixel value S of the maximum connected domain rec ；
(2) Calculating the characteristic rectangular area w x h of the maximum connected domain according to the width w and the height h of the maximum connected domain, and then calculating the accumulated pixel value S of the maximum connected domain rec Difference value deltaS between characteristic rectangular area w x h 2 ；
(3) If the position, width, height and characteristic rectangular area difference value delta S of the maximum connected domain 2 If the detected values are consistent with the corresponding preset threshold values, the false distinguishing requirements based on the rectangular region characteristic patterns are met; otherwise, the false identification requirement based on the characteristic pattern of the rectangular area is not met.
The size of the width w and height h should theoretically be equal to the actual size of the picture rectangular pattern, i.e. (w-w) T ）＜w threshold ，(h-h T ）＜h threshold Wherein w is T 、h T Width and height of the actual rectangular pattern for the image, w threshold 、h threshold The threshold value of the difference value between the width and the height of the back ultraviolet condition image and the rectangular pattern of the template picture is obtained. Since there is no other pattern except the rectangular feature pattern in the back side ultraviolet light condition image, the accumulated pixel values of connected components existing at other positions except the maximum connected component are strictly limited within a preset range.
3. Line feature pattern
The front ultraviolet light condition image of the old hong Kong identity card has a plurality of different bright lines in different areas, so that the bright line characteristic pattern can be used as a counterfeit identification image characteristic, and the counterfeit identification process based on the bright line characteristic pattern is as follows:
(1) in order to achieve a better line extraction effect, firstly, carrying out gray value inversion on the picture, then carrying out binarization processing, and dividing the binarized image into four different sub-regions;
(2) counting the black pixel ratio in each sub-area to be P in turn i1 ,P i2 ,P i3 ,P i4 ；
(3) Calculating the similarity between the image and the corresponding preset line template image based on the Euclidean distance, and if the similarity is smaller than the corresponding preset threshold value, conforming to the counterfeit identification requirement based on the line feature pattern; otherwise, the false identification requirement based on the line characteristic pattern is not met.
Wherein, the calculation formula of the Euclidean distance is as follows:
in the formula, P t1 ,P t2 ,P t3 ,P t4 The ratio of the black pixels in the 4 sub-regions of the preset line template image is respectively.
4. Rainbow characteristic pattern
In the printing process of the front ultraviolet light condition image of the old hong Kong identity card, the process of color gradual change appears, which is called rainbow printing. Therefore, the rainbow characteristic pattern can be used as a counterfeit identification image characteristic, and the counterfeit identification process based on the rainbow characteristic pattern is as follows:
(1) sampling a rainbow area in a white light condition image on the front surface of an old hong Kong identity card to obtain a rainbow image;
(2) converting the rainbow image from an RGB color space into an HSV color space, and counting the distribution range of chromatic values and saturation of adjacent areas; comparing the difference of the chroma values of the adjacent areas with the difference of the saturation;
(3) if the distribution range and the difference value of the chromatic value and the saturation of the adjacent area are in the corresponding preset threshold range, the false distinguishing requirement based on the rainbow characteristic pattern is met; otherwise, the requirement of false identification based on the rainbow characteristic pattern is not met.
The above embodiments are preferred embodiments of the present application, and those skilled in the art can make various changes or modifications without departing from the general concept of the present application, and such changes or modifications should fall within the scope of the claims of the present application.
Claims (9)
1. An identity card authenticity identification method is characterized by being used for authenticating a hong Kong identity card and comprising the following steps:
step 1, acquiring images of a certificate to be authenticated under the conditions of white light, ultraviolet light and infrared light respectively;
step 2, judging whether the certificate to be authenticated is a hong Kong identity card or not according to a redbud pattern at a preset position by using a back white light condition image of the certificate to be authenticated;
step 3, judging a new version and an old version of the hong Kong identity card according to a characteristic window of a preset position by using a front white light condition image of the hong Kong identity card;
step 4, aiming at the new version of hong Kong identity card, the authenticity of the new version of hong Kong identity card is identified by utilizing the image characteristics of the front and back surfaces of the new version of hong Kong identity card under various light source conditions; aiming at the old version of hong Kong identity card, the authenticity of the old version of hong Kong identity card is identified by utilizing the image characteristics of the front and back surfaces of the old version of hong Kong identity card under various light source conditions;
the image of the hong Kong identity card under the white light, ultraviolet light and infrared light conditions respectively has different image characteristics, part of the preset image characteristics are taken as the counterfeit identification criterion, if the selected image characteristics meet the counterfeit identification requirement, the hongKong identity card is judged to pass the counterfeit identification, otherwise, the hongKong identity card is judged not to pass the counterfeit identification; wherein:
all image characteristics of the new identity card counterfeit identification include: transparent window features, bauhinia pattern features, feature circle patterns, landscape pattern features, guilloche pattern features;
all image characteristics of the old identity card counterfeit identification include: bauhinia pattern features, rectangular area feature patterns, line feature patterns, rainbow feature patterns.
2. The identity card authentication method according to claim 1, wherein the step 3 is specifically: and matching the characteristic windows at preset positions in the white light condition image on the back of the certificate to be authenticated by using an edge characteristic template, wherein if the matching is successful, the hong Kong identity card is a new version of hong Kong identity card, and otherwise, the hong Kong identity card is an old version of hong Kong identity card.
3. The method as claimed in claim 1, wherein the step 4 of verifying the identity of hong Kong identity card includes: matching the front white light condition image and the front infrared light condition image of the hong Kong identity card of the new edition by using a front window template; matching the back white light condition image and the back infrared light condition image of the hong Kong identity card of the new edition by using a back window template; if each matching is successful in the corresponding preset position, the counterfeit identification requirement is met; otherwise, the false identification requirement is not met.
4. The method as claimed in claim 1, wherein the step 4 of verifying the identity of hong Kong identity card includes: carrying out template matching on a back white light condition image, a back ultraviolet light condition image and a back infrared light condition image of the hong Kong identity card of the new edition by using a preset bauhinia pattern template; if the back white light condition image and the back ultraviolet light condition image are matched with a preset bauhinia pattern template at preset positions, the authenticity identification requirement is met; otherwise, the product does not meet the counterfeit identification requirement;
and/or;
the step 4 of identifying the old version of hong Kong identity card includes the following steps: performing template matching on a back white light condition image, a back ultraviolet light condition image and a back infrared light condition image of the old hong Kong identity card by using a preset bauhinia pattern template; if the back white light condition image and the back ultraviolet light condition image are matched with a preset bauhinia pattern template at preset positions, the authenticity identification requirement is met; otherwise, the false identification requirement is not met.
5. The method as claimed in claim 1, wherein the step 4 of verifying the identity of hong Kong identity card includes:
sequentially carrying out gray scale inversion, binarization processing and connected domain analysis on the front ultraviolet light condition image of the hong Kong identity card of the new edition to obtain the coordinate position, width, height and accumulated pixel value of the largest connected domain;
calculating the radius r of the characteristic circle according to the width and height of the maximum connected domain, and calculating the accumulated pixel value of the maximum connected domain and the area pi r of the characteristic circle based on the obtained radius r 2 Difference between Δ S 1 ；
If the coordinate position of the maximum connected domain, the radius of the characteristic circle and the area difference Delta S of the characteristic circle 1 If the values are all in accordance with the corresponding preset threshold values, the authenticity identification requirements are met; otherwise, the false identification requirement is not met;
and/or;
the step 4 of identifying the old version of hong Kong identity card includes the following steps:
sequentially carrying out gray value inversion, binarization processing, corrosion and expansion morphological operation and connected domain analysis on a back ultraviolet light condition image of an old version hong Kong identity card to obtain the position, width, height and accumulated pixel value of a maximum connected domain;
calculating the characteristic rectangular area w h of the maximum connected domain according to the width w and the height h of the maximum connected domain, and then calculating the difference value delta S between the accumulated pixel value of the maximum connected domain and the characteristic rectangular area w h 2 ；
If the position, width, height and characteristic rectangular area difference value delta S of the maximum connected domain 2 If the data are in accordance with the corresponding preset threshold values, the data are in accordance with the counterfeit identification requirement; otherwise, the false identification requirement is not met.
6. The method as claimed in claim 1, wherein the step 4 of verifying the identity of hong Kong identity card includes:
sequentially carrying out gray scale inversion, binarization processing, corrosion and expansion morphological processing on the back ultraviolet light condition image of the hong Kong identity card of the new edition;
carrying out XOR logic operation on a preset landscape image template and the image obtained in the last step, and then carrying out connected domain analysis on the obtained image;
if the accumulated value of the pixels of the maximum connected domain is smaller than the corresponding preset threshold value, the requirement of identifying the fake is met; otherwise, the false identification requirement is not met;
and/or;
the step 4 of identifying the old version of hong Kong identity card includes the following steps:
sequentially carrying out gray scale inversion and binarization processing on the front ultraviolet light condition image of the old version of hong Kong identity card to obtain an image comprising four sub-regions;
counting the black pixel ratio in each sub-area to be P in turn i1 ,P i2 ,P i3 ,P i4 ；
Calculating the similarity between the image and the corresponding preset line template image based on the Euclidean distance, and if the similarity is smaller than the corresponding preset threshold value, conforming to the counterfeit identification requirement; otherwise, the false identification requirement is not met.
7. The method as claimed in claim 1, wherein the step 4 of verifying the identity of hong Kong identity card includes:
sampling a guilloche area in a front white light condition image of the hong Kong identity card of the new edition to obtain a guilloche image;
converting the guilloche image from RGB color space to HSV color space, and counting the pixel proportion P of blue chromatic value ib And a pixel ratio P with a chromatic value of red ir ；
Calculating the similarity between the guilloche images and the corresponding preset guilloche template images based on the Euclidean distance, and if the similarity is smaller than a corresponding preset threshold value, conforming to the counterfeit identification requirement; otherwise, the product does not meet the authenticity identification requirement.
8. The method of claim 1, wherein the step 4 of authenticating the old version of hong Kong identity card comprises:
sampling a rainbow area in a white light condition image on the front surface of an old version of hong Kong identity card to obtain a rainbow image;
converting the rainbow image from an RGB color space into an HSV color space, and counting the distribution range of chromatic values and saturation of adjacent areas; comparing the difference of the chroma values of the adjacent areas with the difference of the saturation;
if the distribution range and the difference value of the chromatic value and the saturation of the adjacent area are within the corresponding preset threshold value range, the requirement of counterfeit discrimination is met; otherwise, the false identification requirement is not met.
9. An electronic device comprising a memory and a processor, the memory having stored therein a computer program, wherein the computer program, when executed by the processor, causes the processor to implement the method of any of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110483713.8A CN113378878B (en) | 2021-04-30 | 2021-04-30 | Identity card counterfeit identification method and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110483713.8A CN113378878B (en) | 2021-04-30 | 2021-04-30 | Identity card counterfeit identification method and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113378878A CN113378878A (en) | 2021-09-10 |
| CN113378878B true CN113378878B (en) | 2022-08-16 |
Family
ID=77570453
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110483713.8A Active CN113378878B (en) | 2021-04-30 | 2021-04-30 | Identity card counterfeit identification method and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113378878B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109684987A (en) * | 2018-12-19 | 2019-04-26 | 南京华科和鼎信息科技有限公司 | A kind of authentication system and method based on certificate |
| CN110895693A (en) * | 2019-09-12 | 2020-03-20 | 华中科技大学 | Authentication method and authentication system for anti-counterfeiting information of certificate |
| CN111986378A (en) * | 2020-07-30 | 2020-11-24 | 湖南长城信息金融设备有限责任公司 | Bill color fiber yarn detection method and system |
| CN112017352A (en) * | 2020-09-03 | 2020-12-01 | 平安科技（深圳）有限公司 | Certificate authentication method, device, equipment and readable storage medium |
| CN112200136A (en) * | 2020-10-29 | 2021-01-08 | 腾讯科技（深圳）有限公司 | Certificate authenticity identification method and device, computer readable medium and electronic equipment |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB201411295D0 (en) * | 2014-06-25 | 2014-08-06 | Idscan Biometrics Ltd | Identity document digital image correction method, system and computer program |
-
2021
- 2021-04-30 CN CN202110483713.8A patent/CN113378878B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109684987A (en) * | 2018-12-19 | 2019-04-26 | 南京华科和鼎信息科技有限公司 | A kind of authentication system and method based on certificate |
| CN110895693A (en) * | 2019-09-12 | 2020-03-20 | 华中科技大学 | Authentication method and authentication system for anti-counterfeiting information of certificate |
| CN111986378A (en) * | 2020-07-30 | 2020-11-24 | 湖南长城信息金融设备有限责任公司 | Bill color fiber yarn detection method and system |
| CN112017352A (en) * | 2020-09-03 | 2020-12-01 | 平安科技（深圳）有限公司 | Certificate authentication method, device, equipment and readable storage medium |
| CN112200136A (en) * | 2020-10-29 | 2021-01-08 | 腾讯科技（深圳）有限公司 | Certificate authenticity identification method and device, computer readable medium and electronic equipment |
Non-Patent Citations (2)
| Title |
|---|
| 面向二代身份证的图文识别;洪雅敏;《兰州工业学院学报》;20130630;全文 * |
| 香港12月份起更换新版智能身份证，九大防伪特征集聚;苏州印象镭射;《https://www.gzlaser.com/news/980.html》;20181126;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113378878A (en) | 2021-09-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11062163B2 (en) | Iterative recognition-guided thresholding and data extraction | |
| US11302109B2 (en) | Range and/or polarity-based thresholding for improved data extraction | |
| CA2658357C (en) | Document authentication using template matching with fast masked normalized cross-correlation | |
| US6901155B2 (en) | Wavelet-enhanced automated fingerprint identification system | |
| JP6080259B2 (en) | Character cutting device and character cutting method | |
| Shen et al. | Improving OCR performance with background image elimination | |
| US9349237B2 (en) | Method of authenticating a printed document | |
| CN109409158B (en) | Anti-counterfeiting method based on two-dimensional code edge roughness | |
| Deep Kaur et al. | An analysis of image forgery detection techniques | |
| CN111986378B (en) | Bill color fiber yarn detection method and system | |
| Abramova et al. | Detecting copy–move forgeries in scanned text documents | |
| CN104102907A (en) | LBP (Local Binary Pattern) face recognition method for eliminating illumination evenness | |
| Rajan et al. | An extensive study on currency recognition system using image processing | |
| CN113378878B (en) | Identity card counterfeit identification method and electronic equipment | |
| KR100726473B1 (en) | Apparatus for classifying an image and method therefor | |
| Nandanwar et al. | DCT-phase statistics for forged IMEI numbers and air ticket detection | |
| Chapel et al. | Authentication of holograms with mixed patterns by direct LBP comparison | |
| Xu et al. | Coin recognition method based on SIFT algorithm | |
| Forczmański et al. | Low-level image features for stamps detection and classification | |
| Ali et al. | Detection and extraction of pantograph region from bank cheque images | |
| Tang et al. | A novel similar background components connection algorithm for colorful text detection in natural images | |
| Savakar et al. | A novel forgery detection technique for document images | |
| Silva-Mata¹ | Rubén Sánchez-Rivero1, 2, Yasmany Febles-Espinosa¹, 2, Francisco José Silva-Mata¹, 2, and Annette Morales-González¹, 2 () 9 1 Advanced Technologies Application Center, CENATAV, Havana, Cuba {rsanchez, fjsilva, amorales}@ cenatav. co. cu | |
| Chethan et al. | Image dewarping and text extraction from mobile captured distinct documents | |
| Singh et al. | Fingerprint recognition |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |