CN106780953B - Paper money counterfeit distinguishing method and system based on double-crown-word number - Google Patents
Paper money counterfeit distinguishing method and system based on double-crown-word number Download PDFInfo
- Publication number
- CN106780953B CN106780953B CN201710026017.8A CN201710026017A CN106780953B CN 106780953 B CN106780953 B CN 106780953B CN 201710026017 A CN201710026017 A CN 201710026017A CN 106780953 B CN106780953 B CN 106780953B
- Authority
- CN
- China
- Prior art keywords
- word number
- crown word
- crown
- projection
- image
- 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
- G07—CHECKING-DEVICES
- G07D—HANDLING OF COINS OR VALUABLE PAPERS, e.g. TESTING, SORTING BY DENOMINATIONS, COUNTING, DISPENSING, CHANGING OR DEPOSITING
- G07D7/00—Testing specially adapted to determine the identity or genuineness of valuable papers or for segregating those which are unacceptable, e.g. banknotes that are alien to a currency
- G07D7/004—Testing specially adapted to determine the identity or genuineness of valuable papers or for segregating those which are unacceptable, e.g. banknotes that are alien to a currency using digital security elements, e.g. information coded on a magnetic thread or strip
- G07D7/0047—Testing specially adapted to determine the identity or genuineness of valuable papers or for segregating those which are unacceptable, e.g. banknotes that are alien to a currency using digital security elements, e.g. information coded on a magnetic thread or strip using checkcodes, e.g. coded numbers derived from serial number and denomination
Abstract
The invention relates to a paper money counterfeit discriminating technology, in particular to a paper money counterfeit discriminating method for self-checking through a special double-crown-word number of paper money and a system for realizing the method, which comprises the steps of firstly obtaining a paper money image signal; then, image preprocessing is carried out, including bilinear interpolation and binarization; then equally dividing the character into four areas, respectively counting the black dot proportion, and comparing the similarity of the horizontal black dot proportion and the vertical black dot proportion; then, carrying out X-direction projection and Y-direction projection in the transverse direction and the longitudinal direction respectively, and counting projection energy; and finally, the final coupling degree judgment is carried out by combining the black dot proportion similarity and the projection energy similarity, and whether the transverse character is consistent with the longitudinal character or not is judged to efficiently identify the counterfeit money. The method has the advantages of high accuracy, concise algorithm implementation, less calculation amount, suitability for embedded financial instruments and capability of effectively improving the authenticity identification performance of the financial instruments.
Description
Technical Field
The invention relates to a paper money counterfeit distinguishing technology, in particular to a paper money counterfeit distinguishing method for self-checking through a special double-crown-word number of paper money and a system for realizing the method.
Background
The currency serves as a value symbol and a value exchange medium and plays an irreplaceable important role in commodity economy, and plays a role in regulating, promoting and stabilizing national economy.
In recent years, the Chinese economy continues to develop at a high speed, social wealth is rapidly accumulated, and the phenomenon of counterfeit money preparation and selling is rapidly spread. The circulation of counterfeit money influences the market law, disturbs the social and economic order, and damages the credit of one national currency and the interests of the social public. Therefore, the counterfeit money is not listed as the object of payment and striking in all countries of the world, and the anti-counterfeit money has become a worldwide topic.
The types of counterfeit money are divided into two types: counterfeit money, and altered money. The counterfeit money has larger difference with the genuine money due to the characteristics of the material, the printing ink and the printing manufacturing process of the counterfeit money, so the current financial machines and tools can be basically and effectively identified, the changed money is the paper money changed by adopting illegal means such as uncovering the front surface or the back surface of the ticket, cutting, splicing and changing the denomination and the like, the common 'changed money' on the market mainly comprises spliced money, one genuine money is divided into two parts, and simultaneously, half of genuine money and half of counterfeit money are spliced together by sticking and complementing; the method is characterized in that small-denomination real coins are cut, color-changing fluorescent numbers on the real coins are cut, and then the color-changing fluorescent numbers are adhered and repaired on the same position of large-denomination counterfeit coins, so that the financial machines and tools are easy to be confused, and how to effectively identify the counterfeit money by the financial machines and tools is a great technical problem to be solved in the financial machine tool industry.
At present, paper money of many countries has double-crown numbers, taking the RMB as an example, 100 Yuan of the RMB 2015 edition and 100 Yuan and 50 Yuan of the 1999 edition have double-crown numbers, namely, transverse crown numbers and longitudinal crown numbers, and the crown numbers of two crown number areas are consistent. The counterfeit money can be detected better by comparing whether the left and right crown word numbers are consistent or not through the counterfeit identification technology.
The existing crown word number extraction technology obtains the crown word number of the paper currency by obtaining a crown word number image of the paper currency and extracting features, if double crown word number identification is carried out in a conventional mode, OCR (optical character Recognition) extraction of the crown word number is required to be carried out twice respectively, and then comparison of whether characters are the same or not is carried out, so that the technical problems that the efficiency is low, the calculation amount is large, and the crown word number cannot be identified after unified string change exist in the prior art.
Disclosure of Invention
The invention aims to solve the technical problem of low processing speed caused by large operation amount in the prior art, and provides a method for confirming whether transverse and longitudinal crown word numbers are consistent or not by extracting and comparing character image characteristics of the transverse crown word numbers and the longitudinal crown word numbers so as to confirm whether the transverse and longitudinal crown word numbers are changed into coins or not.
The paper currency identification method based on the double-crown-word number specifically comprises the following processing steps:
step 1: collecting image information of the paper money, and respectively extracting a first crown word number area and a second crown word number area;
step 2: respectively carrying out size normalization on the first crown word number region and the second crown word number region, and carrying out binarization processing to form a first crown word number binarization image and a second crown word number binarization image;
and step 3: respectively carrying out identical at least 2n equal partition on the first crown word number binary image and the second crown word number binary image, wherein n is a natural number, calculating the proportion of black points in each partition area, carrying out absolute difference statistics on the proportion of the black points in each corresponding area of the first crown word number and the second crown word number in sequence, summing all the differences, and recording the sum as a characteristic T1;
and 4, step 4: respectively carrying out black point projection in the X direction and the Y direction on the first crown word number image and the second crown word number image, then respectively calculating projection energy in the X direction and the Y direction, and carrying out absolute difference on the projection energy in the X direction of the first crown word number and the projection energy in the Y direction corresponding to the second crown word number, and marking as a characteristic T2; carrying out absolute difference on the projection energy in the Y direction of the first crown word number and the projection energy in the X direction corresponding to the second crown word number, and recording as a characteristic T3;
and 5: and carrying out coupling degree statistics on the characteristic T1, the characteristic T2 and the characteristic T3, and judging whether the first crown word number is consistent with the second crown word number according to the coupling degree value so as to obtain the counterfeit identification result of whether the corresponding paper money is changed.
Preferably, the step 1: collecting image information of paper money, and respectively extracting a first crown word number area and a second crown word number area, specifically:
step 1.1: collecting image information of the paper money, and respectively extracting a first crown word number area and a second crown word number area;
step 1.2: character region extraction is respectively carried out on the first crown region and the second crown region, and the first crown region and the second crown region are respectively marked as horizon char (i) (i 1.. N) and Verticala char (i) (i 1.. N), wherein N represents the number of the crown containing characters.
Further, the step 2: respectively carrying out size normalization on the first crown word number region and the second crown word number region, and carrying out binarization processing to form a first crown word number binarization image and a second crown word number binarization image, specifically:
step 2.1: the character regions horizon Char (i) (i ═ 1 … N) and vertical Char (i) (i ═ 1 … N) are respectively subjected to size normalization, forming character normalized images H2Char (i) (i ═ 1.. N) and V2Char (i) (i ═ 1.. N), where N denotes the number of the crown numbers containing the characters.
Step 2.2: respectively binarizing the character normalized image H2Char (i) (i 1.. N) and the character normalized image V2Char (i) (i 1.. N) to form a character binarized image H3Char (i) (i 1.. 1 … N) and a character binarized image V3Char (i) (i 1.. N);
further, the step 3: respectively carrying out identical at least 2n equal partition on the first crown word number binary image and the second crown word number binary image, wherein n is a natural number, calculating the proportion of black points in each partition area, carrying out absolute difference statistics on the proportion of the black points in each corresponding area of the first crown word number and the second crown word number in sequence, summing all the differences, and recording the sum as a characteristic T1, wherein the method specifically comprises the following steps:
step 3.1: respectively carrying out identical at least 2N equal division on the character binarization image H3Char (i) (i is 1.. N) and the character binarization image V3Char (i) (i is 1 … N), wherein j represents the number of divided blocks and N is a natural number, and the division is marked as H3Charpart (i, j) (j is 1.. 2N) and V3Charpart (i, j) (j is 1.. 2N);
step 3.2: respectively counting proportion values of black points in H3CharPart (i, j) (j is 1.. 2n) and V3CharPart (i, j) (j is 1.. 2n) blocks, and respectively marking the proportion values as H3CharBlackPointRatio (i, j) (j is 1.. 2n) and V3CharBlackPointRatio (i, j) (j is 1.. 2n), wherein j represents the number of the divided blocks, and n is a natural number;
step 3.3: absolute difference statistics were performed on the comparative example values H3CharBlackPointRatio (i, j) (j 1.. 2n) and V3CharBlackPointRatio (i, j) (j 1.. 2n), specifically:
the absolute difference value and the CharBlackRatioSameLevel (i) form a feature T1.
Further, the step 4: respectively carrying out black point projection in the X direction and the Y direction on the first crown word number image and the second crown word number image, then respectively calculating projection energy in the X direction and the Y direction, and carrying out absolute difference on the projection energy in the X direction of the first crown word number and the projection energy in the Y direction corresponding to the second crown word number, and marking as a characteristic T2; the absolute difference between the projection energy in the Y direction of the first crown word number and the projection energy in the X direction corresponding to the second crown word number is recorded as a feature T3, which specifically includes:
step 4.1: performing X-direction and Y-direction black point projections on the partitions H3CharPart (i, j) (j 1.. 2n) and V3CharPart (i, j) (j 1.. 2n), respectively, the first crown X-direction and Y-direction projections forming projection markers H3CharShadowX (i) (i 1.. 2n) and H3 charshady (i) (i 1.. 2n), respectively, and the second crown X-direction and Y-direction projections forming projection markers V3CharShadowX (i) (i 1.. 2n) and V3 charshady (i) (i 1.. 2n), respectively;
step 4.2: obtaining a projection energy H3 charshardowxpower (i) (i ═ 1.. 2n) and H3 charshardowxpower (i) (i ═ 1.. 2n) of a first prefix X-direction projection H3 charshardow X (i) (i ═ 1.. 2n) and a Y-direction projection H3 chardow Y (i) (i ═ 1.. 2n) according to the ShadowPower ═ Σ k ═ f (k), where k denotes the signal point position and f (k) denotes the signal point amplitude; the projection energies of the second prefix X-direction projection V3 charshardowx (i) (i ═ 1 … 2n) and the Y-direction projection V3 charshardowy (i) (i ═ 1.. 2n) are denoted V3 charshardowxpower (i) (i ═ 1.. 2n), V3 charshardowypower (i) (i ═ 1.. 2n), respectively;
step 4.3: absolute difference is carried out between the projection energy H3CharShadowXpower (i) (i is 1.. 2n) in the X direction of the first crown word and the projection energy V3CharShadowYpower (i) (i is 1.. 2n) in the Y direction corresponding to the second crown word, and the absolute difference is recorded as a characteristic T2; the projection energy H3CharShadowYpower (i) (i 1.. 2n) in the Y direction of the first prefix number and the projection energy V3CharShadowXpower (i) (i 1.. 2n) in the X direction corresponding to the second prefix number are absolute-differenced and are denoted as a feature T3.
Further, the step 5: carrying out coupling degree statistics on the characteristic T1, the characteristic T2 and the characteristic T3, and judging whether the first crown word number is consistent with the second crown word number according to the coupling degree value so as to obtain an authenticity identification result of whether the corresponding paper money is changed, wherein the method specifically comprises the following steps:
step 5.1: the coupling statistics were performed according to the following formula:
FinalSameLevel(i)＝coef1*T1+coef1*T2+coef1*T3
wherein coef1+ coef2+ coef3 is 1;
step 5.2, judging the coupling degree value according to the following conditions,
when FinalJudge (i) is 1, the first crown number is different from the second crown number, and when FinalJudge (i) is 0, the two crown numbers are the same;
and 5.3, finally giving the counterfeit identification result of whether the paper money is changed.
The invention also aims to provide a double-crown-number-based paper currency counterfeit detection system, which specifically comprises:
the image acquisition unit is used for acquiring image information of the paper money and respectively extracting a first crown word number area and a second crown word number area;
the image processing unit is used for carrying out size normalization on the first crown word number region and the second crown word number region and carrying out binarization processing to form a first crown word number binarization image and a second crown word number binarization image;
an image blocking and black point counting unit, which is used for carrying out the same at least 2n equal partition on the first crown word number binary image and the second crown word number binary image, wherein n is a natural number, calculating the proportion of black points in each partition area, carrying out absolute difference value counting on the proportion of the black points in each corresponding area of the first crown word number and the second crown word number in sequence, summing all the difference values, and recording the sum value as a characteristic T1;
an image black point projection and energy difference value calculation unit, which is used for carrying out black point projection in the X direction and the Y direction on the first crown word number image and the second crown word number image, then calculating projection energy in the X direction and the Y direction respectively, and carrying out absolute difference on the projection energy in the X direction of the first crown word number and the projection energy in the Y direction corresponding to the second crown word number, and marking as a characteristic T2; carrying out absolute difference on the projection energy in the Y direction of the first crown word number and the projection energy in the X direction corresponding to the second crown word number, and recording as a characteristic T3;
and the characteristic coupling degree judging unit is used for carrying out coupling degree statistics on the characteristic T1, the characteristic T2 and the characteristic T3 and judging whether the first crown word number is consistent with the second crown word number according to the coupling degree so as to obtain a counterfeit identification result of whether the corresponding paper money is changed.
Preferably, the image capturing unit further includes a character region extracting unit, configured to extract a character region from the first crown word number region and the second crown word number region.
Compared with the prior art, the technical scheme has the following beneficial effects:
respectively counting black point occupation ratios of the first crown word number binary image and the second crown word number binary image, and comparing the horizontal and vertical black point occupation ratio similarity to form a first judgment dimension; then respectively carrying out X-direction projection and Y-direction projection on the first crown word number binary image and the second crown word number binary image, and counting projection energy; and finally, combining the black point proportion similarity and the projection energy similarity to form a second judgment dimension and a third judgment dimension, and finally, judging the final coupling degree according to the first judgment dimension, the second judgment dimension and the third judgment dimension so as to judge whether the first crown word number is consistent with the second crown word number. The method has the advantages of high accuracy, simple algorithm implementation, less calculation amount, suitability for embedded financial instruments and capability of effectively improving the authenticity identification performance of the financial instruments.
Drawings
FIG. 1 is a schematic diagram of a paper currency identification system based on double-crown-number feature according to the present invention;
FIG. 2 is a schematic diagram of the optimized functional units of the paper currency identification system shown in FIG. 1;
FIG. 3 is a processing flow chart of a paper currency counterfeit identification method based on double-crown-number characteristics provided by the invention.
DETAILED DESCRIPTION OF EMBODIMENT (S) OF INVENTION
The embodiments described below are only a part of embodiments, but not all embodiments, of the double-serial-number-based banknote authentication method and system provided by the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, the present invention provides a paper currency identification system based on double-crown-number feature, which specifically includes: the image acquisition unit 1 is used for acquiring image information of the paper money and respectively extracting a first crown word number area and a second crown word number area;
the image processing unit 2 is used for carrying out size normalization on the first crown word number region and the second crown word number region and carrying out binarization processing to form a first crown word number binarization image and a second crown word number binarization image;
an image blocking and black point counting unit 3, configured to perform the same at least 2n equal partition on the first crown word number binarized image and the second crown word number binarized image, where n is a natural number, and on the premise of ensuring precision and efficiency, generally n is 2, which may meet the requirement, and certainly the larger the value of n is, the higher the accuracy of false identification is, but the larger the computation amount is, the processing efficiency is reduced, and the proportion of black points in each partitioned area is calculated, and the absolute difference value counting is performed on the proportion of black points in each corresponding area of the first crown word number and the second crown word number in sequence, and the sum value is recorded as a feature T1;
an image black point projection and energy difference value calculation unit 4, configured to perform black point projection in the X direction and the Y direction on the first crown word number binarized image and the second crown word number binarized image, then calculate projection energy in the X direction and the Y direction, respectively, and perform an absolute difference between the projection energy in the X direction of the first crown word number and the projection energy in the Y direction corresponding to the second crown word number, which is denoted as a feature T2; carrying out absolute difference on the projection energy in the Y direction of the first crown word number and the projection energy in the X direction corresponding to the second crown word number, and recording as a characteristic T3;
and the characteristic coupling degree judging unit 5 is used for carrying out coupling degree statistics on the characteristic T1, the characteristic T2 and the characteristic T3 and judging whether the first crown word number is consistent with the second crown word number according to the coupling degree so as to obtain a counterfeit identification result of whether the corresponding paper money is changed.
Referring to fig. 2, preferably, the image capturing unit further includes a character region extracting unit 11 for extracting character regions of the first crown word number region and the second crown word number region. Correspondingly, the image processing unit 2 is configured to perform size normalization on the character region of the first crown word size region and the character region of the second crown word size region, and perform binarization processing to form a first crown word size character region binarized image and a second crown word size character region binarized image;
the image blocking and black point counting unit 3 is configured to equally divide the first crown word number character region binarized image and the second crown word number character region binarized image into at least 2n equal partitions, where n is a natural number, and on the premise of ensuring accuracy and efficiency, generally n is 2, which may meet the requirement, and certainly the larger the value of n is, the higher the accuracy of false identification is, but the larger the computation amount is, the processing efficiency is reduced, and the proportion of black points in each partition region is calculated, and the absolute difference value counting is performed on the proportion of black points in each corresponding region in the first crown word number character region and the second crown word number character region in sequence, and the sum value is recorded as a feature T1;
the image black point projection and energy difference value calculation unit 4 is configured to perform black point projection in the X direction and the Y direction on the first crown word size character region binarized image and the second crown word size character region binarized image, then calculate projection energy in the X direction and the Y direction, and perform an absolute difference between the projection energy in the X direction of the first crown word size character region and the projection energy in the Y direction corresponding to the second crown word size character region, and record the absolute difference as a feature T2; carrying out absolute difference on the projection energy in the Y direction of the character area of the first prefix number and the projection energy in the X direction corresponding to the character area of the second prefix number, and recording as a characteristic T3;
the characteristic coupling degree judging unit 5 is configured to perform coupling degree statistics on the characteristic T1, the characteristic T2 and the characteristic T3, and judge whether the first crown word number and the second crown word number are consistent according to the coupling degree value, so as to obtain a counterfeit discrimination result of whether the corresponding banknote is a counterfeit banknote.
The system extracts the characteristics of the transverse crown word number and the longitudinal crown word number on the paper money and then compares the transverse characteristics with the longitudinal characteristics to determine whether the transverse crown word number and the longitudinal crown word number are consistent or not, so that the effect of detecting variable coinage is achieved. The method is high in accuracy and convenient and fast, and can effectively improve the counterfeit distinguishing performance of the financial instruments.
Referring to fig. 3, the method for identifying counterfeit paper money based on the double-crown-number feature according to the embodiment of the present invention specifically includes the following processing steps:
s1: collecting image information of the paper money, and respectively extracting a first crown word number area and a second crown word number area; taking RMB as an example: respectively extracting a transverse crown word number area and a longitudinal crown word number area on the paper money, and respectively marking as a crown word number A and a crown word number B;
s2: respectively carrying out size normalization on the first crown word number region and the second crown word number region, and carrying out binarization processing to form a first crown word number binarization image and a second crown word number binarization image; taking RMB as an example: respectively cutting and extracting N characters of the crown word number A and the crown word number B, wherein N is equal to 10, carrying out size normalization on the characters of the crown word number A and the corresponding characters of the crown word number B, and then carrying out binarization;
s3: respectively carrying out identical at least 2n equal partition on the first crown word number binary image and the second crown word number binary image, wherein n is a natural number, on the premise of ensuring the precision and the efficiency, the requirement can be met generally when n is 2, the higher the value of n is, the higher the authenticity identification accuracy is, but the calculation amount is increased, the processing efficiency is reduced, the proportion of black points in each partition area is calculated, absolute difference value statistics is carried out on the proportion of the black points in each corresponding area of the first crown word number and the second crown word number in sequence, all the difference values are summed, and the sum value is marked as a characteristic T1; taking RMB as an example: equally dividing each character image of the crown word number A into 4 regions, calculating the proportion of black points of each divided region in the whole character, and counting the proportion of the black points of the crown word number B according to the same method; then carrying out absolute difference statistics on black point proportions of 4 areas of corresponding characters of the crown word number A and the crown word number B in sequence, summing the 4 difference values, and recording the summed value as a characteristic T1;
s4: respectively carrying out black point projection in the X direction and the Y direction on the first crown word number binary image and the second crown word number binary image, then respectively calculating projection energy in the X direction and the Y direction, and carrying out absolute difference on the projection energy in the X direction of the first crown word number and the projection energy in the Y direction corresponding to the second crown word number, and marking as a characteristic T2; carrying out absolute difference on the projection energy in the Y direction of the first crown word number and the projection energy in the X direction corresponding to the second crown word number, and recording as a characteristic T3; taking RMB as an example: carrying out black point projection in the X direction and the Y direction on each character of the crown word number A, then respectively calculating projection energy in the X direction and the Y direction, and calculating the projection energy in the X direction and the projection energy in the Y direction by the crown word number B according to the same method; carrying out absolute difference on the projection energy of each character X direction of the crown word number A and the projection energy of the corresponding character Y direction of the crown word number B, and marking as a characteristic T2; carrying out absolute difference on the projection energy of each character Y direction of the crown word number A and the projection energy of the crown word number B corresponding to the character X direction, and marking as a characteristic T3;
s5: and (3) carrying out coupling degree statistics on the characteristic T1, the characteristic T2 and the characteristic T3, and judging whether the crown word number A is consistent with the crown word number B according to the coupling degree value so as to obtain an authenticity identification result of whether the corresponding paper currency is changed into currency, taking the RMB as an example: and (4) judging the coupling degree of the features 1, 2 and 3, and judging whether the corresponding characters of the crown word number A and the crown word number B are the same or not, so as to judge whether the crown word number A is consistent with the crown word number B or not.
The processing steps of the method can be used for operating the characteristic T1, the characteristic T2 and the characteristic T3 one by one or a plurality of parallel processing, and only can be expressed in sequence due to the limitation of character description, but in technical realization, a computer can be selected to be used for operating the characteristic T1, the characteristic T2 and the characteristic T3 one by one or a plurality of parallel processing according to the operation performance.
Preferably, the banknote authentication method based on the double-crown-number feature can be further realized by the following steps:
s1.1: collecting image information of the paper money, and respectively extracting a first crown word number area and a second crown word number area;
s1.2: extracting character regions of the first crown region and the second crown region respectively, wherein the character regions are marked as horizon charr (i) (i is 1.. N) and Verticalchar (i) (i is 1.. N), and N represents the number of characters contained in the crown; taking RMB as an example: n is 10.
S2.1: size normalization is performed on the character regions horizon Char (i) (i 1.. N) and vertical Char (i) (i 1.. N), respectively, to form character normalized images H2Char (i) (i 1.. N) and V2Char (i) (i 1.. N), where N represents the number of characters contained in the prefix number.
S2.2: respectively carrying out binarization processing on the character normalized image H2Char (i) (i 1.. N) and the V2Char (i) (i 1.. N) to form a character binarized image H3Char (i) (i 1.. N) and a V3Char (i) (i 1.. N);
s3.1: the method comprises the steps of performing identical at least 2N equal partitioning on a character binarization image H3Char (i) (i is 1.. N) and a character binarization image V3Char (i) (i is 1.. N), wherein N is a natural number, wherein N is 2, namely performing 4 equal partitioning on each character binarization image, wherein the image size of the H3Char (i) and the V3Char (i) is R.S, wherein i represents an ith character, and R and S represent a character height and a character width, dividing the H3Char (i) and the V3Char (i) into 4 small blocks of R/2S/2, and marking the small blocks as H3CharPart (i, j) (j is 1.. 2N) and V3Char Part (i, j) (j is 1.. 2N), wherein j represents a divided block number of 2, N;
s3.2: respectively counting proportion values of black points in H3CharPart (i, j) (j is 1.. 2n) and V3CharPart (i, j) (j is 1.. 2n) blocks, and respectively marking the proportion values as H3CharBlackPointRatio (i, j) (j is 1.. 2n) and V3CharBlackPointRatio (i, j) (j is 1.. 2n), wherein j represents the number of the divided blocks, and n is a natural number;
s3.3: absolute difference statistics were performed on the comparative example values H3CharBlackPointRatio (i, j) (j 1.. 2n) and V3CharBlackPointRatio (i, j) (j 1.. 2n), specifically:
the absolute difference value and the CharBlackRatioSameLevel (i) form a feature T1.
S4.1: performing X-direction and Y-direction black point projections on each of the character blocks H3CharPart (i, j) (j 1.. 2n) and V3CharPart (i, j) (j 1.. 2n), respectively, the first crown X-direction and Y-direction projections forming projection marks H3 charsharowx (i) (i 1.. 2n) and H3 charsharowy (i) (i 1.. 2n), respectively, and the second crown X-direction and Y-direction projections forming projection marks V3 charsharowx (i) (i 1.. 2n) and V3 charsharowy (i) (i 1.. 2n), respectively;
s4.2: obtaining a projection energy H3 charshardowxpower (i) (i ═ 1.. 2n) and H3 charshardowxpower (i) (i ═ 1.. 2n) of a first prefix X-direction projection H3 charshardow X (i) (i ═ 1.. 2n) and a Y-direction projection H3 chardow Y (i) (i ═ 1.. 2n) according to the ShadowPower ═ Σ k ═ f (k), where k denotes the signal point position and f (k) denotes the signal point amplitude; the projection energies of the second prefix X-direction projection V3 charshardowx (i) (i 1.. 2n) and the Y-direction projection V3 charshardowy (i) (i 1.. 2n) are denoted as V3 charshardowxpower (i) (i 1.. 2n), V3 charshardowypower (i) (i 1.. 2n), respectively;
s4.3: absolute difference is carried out between the projection energy H3CharShadowXpower (i) (i is 1.. 2n) in the X direction of the first crown word and the projection energy V3CharShadowYpower (i) (i is 1.. 2n) in the Y direction corresponding to the second crown word, and the absolute difference is recorded as a characteristic T2; the projection energy H3CharShadowYpower (i) (i 1.. 2n) in the Y direction of the first prefix number and the projection energy V3CharShadowXpower (i) (i 1.. 2n) in the X direction corresponding to the second prefix number are absolute-differenced and are denoted as a feature T3.
S5.1: the coupling statistics were performed according to the following formula:
FinalSameLevel(i)＝coef1*T1+coef1*T2+coef1*T3
wherein coef1+ coef2+ coef3 is 1;
s5.2, judging the coupling degree value according to the following conditions,
when FinalJudge (i) is 1, the first crown number is different from the second crown number, and when FinalJudge (i) is 0, the two crown numbers are the same;
and S5.3, finally giving the counterfeit identification result of whether the paper money is changed.
In the embodiment of the invention, firstly, a paper money image signal is obtained; then, image preprocessing is carried out, including bilinear interpolation and binarization; then equally dividing the character into four areas, respectively counting the black dot proportion, and comparing the similarity of the horizontal black dot proportion and the vertical black dot proportion; then, carrying out X-direction projection and Y-direction projection in the transverse direction and the longitudinal direction respectively, and counting projection energy; and finally, judging the final coupling degree by combining the black dot proportion similarity and the projection energy similarity, and judging whether the horizontal character is consistent with the vertical character. The method has the advantages of high accuracy, concise algorithm implementation, less calculation amount, suitability for embedded financial instruments and capability of effectively improving the authenticity identification performance of the financial instruments.
The above disclosure is only one preferred embodiment of the present invention, and certainly should not be taken as limiting the scope of the invention, which is defined by the claims and their equivalents.
Claims (8)
1. A paper money counterfeit distinguishing method based on double-crown-word numbers comprises the following steps:
step 1: collecting image information of the paper money, and respectively extracting a first crown word number area and a second crown word number area;
step 2: respectively carrying out size normalization on the first crown word number region and the second crown word number region, and carrying out binarization processing to form a first crown word number binarization image and a second crown word number binarization image;
and step 3: respectively carrying out identical at least 2n equal partition on the first crown word number binary image and the second crown word number binary image, wherein n is a natural number, calculating the proportion of black points in each partition area, carrying out absolute difference statistics on the proportion of the black points in each corresponding area of the first crown word number and the second crown word number in sequence, summing all the differences, and recording the sum as a characteristic T1;
and 4, step 4: respectively carrying out black point projection in the X direction and the Y direction on the first crown word number binary image and the second crown word number binary image, then respectively calculating projection energy in the X direction and the Y direction, and carrying out absolute difference on the projection energy in the X direction of the first crown word number and the projection energy in the Y direction corresponding to the second crown word number, and marking as a characteristic T2; carrying out absolute difference on the projection energy in the Y direction of the first crown word number and the projection energy in the X direction corresponding to the second crown word number, and recording as a characteristic T3;
and 5: and carrying out coupling degree statistics on the characteristic T1, the characteristic T2 and the characteristic T3, and judging whether the first crown word number is consistent with the second crown word number according to the coupling degree value so as to obtain the counterfeit identification result of whether the corresponding paper money is changed.
2. The double-crown-number-based paper money authentication method according to claim 1, wherein the step 1: collecting image information of paper money, and respectively extracting a first crown word number area and a second crown word number area, specifically:
step 1.1: collecting image information of the paper money, and respectively extracting a first crown word number area and a second crown word number area;
step 1.2: character region extraction is respectively carried out on the first crown region and the second crown region, and the first crown region and the second crown region are respectively marked as horizon char (i) (i 1.. N) and Verticala char (i) (i 1.. N), wherein N represents the number of the crown containing characters.
3. The double-crown-number-based paper money discriminating method according to claim 2, wherein the step 2: respectively carrying out size normalization on the first crown word number region and the second crown word number region, and carrying out binarization processing to form a first crown word number binarization image and a second crown word number binarization image, specifically:
step 2.1: respectively carrying out size normalization on the character regions, namely the horizotalchar (i () i ═ 1.. N) and the Verticalchar (i () i ═ 1.. N), so as to form character normalized images H2Char (i () i ═ 1.. N) and V2Char (i () i ═ 1.. N), wherein N represents the number of the crown numbers containing the characters;
step 2.2: binarizing the character normalized image H2Char (i () i ═ 1.. N) and V2Char (i () i ═ 1.. N), respectively, to form a character binarized image H3Char (i () i ═ 1.. N) and V3Char (i () i ═ 1.. N).
4. The double-crown-number-based paper money discriminating method according to claim 3, wherein the step 3: respectively carrying out identical at least 2n equal partition on the first crown word number binary image and the second crown word number binary image, wherein n is a natural number, calculating the proportion of black points in each partition area, carrying out absolute difference statistics on the proportion of the black points in each corresponding area of the first crown word number and the second crown word number in sequence, summing all the differences, and recording the sum as a characteristic T1, wherein the method specifically comprises the following steps:
step 3.1: respectively carrying out identical at least 2N equal division on a character binarization image H3Char (i) (i is 1.. N) and a character binarization image V3Char (i) (i is 1.. N), wherein j represents the number of divided blocks and N is a natural number, and marked as H3Charpart (i, j) (j is 1.. 2N) and V3Charpart (i, j) (j is 1.. 2N);
step 3.2: respectively counting proportion values of black points in H3CharPart (i, j) (j is 1.. 2n) and V3CharPart (i, j) (j is 1.. 2n) blocks, and respectively marking the proportion values as H3CharBlackPointRatio (i, j) (j is 1.. 2n) and V3CharBlackPointRatio (i, j) (j is 1.. 2n), wherein j represents the number of the divided blocks, and n is a natural number;
step 3.3: absolute difference statistics were performed on comparative example values H3CharBlackPointRatio (i, j) (j 1.. 2n) and V3CharBlackPointRatio (i, j) (j 1.. 2n), specifically:
the absolute difference value and the CharBlackRatioSameLevel (i) form a feature T1.
5. The double-crown-number-based paper money discriminating method according to claim 4, wherein the step 4: respectively carrying out black point projection in the X direction and the Y direction on the first crown word number binary image and the second crown word number binary image, then respectively calculating projection energy in the X direction and the Y direction, and carrying out absolute difference on the projection energy in the X direction of the first crown word number and the projection energy in the Y direction corresponding to the second crown word number, and marking as a characteristic T2; the absolute difference between the projection energy in the Y direction of the first crown word number and the projection energy in the X direction corresponding to the second crown word number is recorded as a feature T3, which specifically includes:
step 4.1: performing X-direction and Y-direction black point projection on the blocks H3CharPart (i, j) (j ═ 1.. 2n) and V3CharPart (i, j) (j ═ 1.. 2n) of the binarized image of each character, wherein the X-direction and Y-direction projection of the binarized image of each character of the first crown form projection marks of H3 charshardawx (i) (i ═ 1.. 2n) and H3 charshardawy (i) (i ═ 1.. 2n), respectively, and the X-direction and Y-direction projection of the binarized image of each character of the second crown are marked of V3 charshardawx (i) (i ═ 1.. 2n) and V3 charshardawy (i) (i ═ 1.. 2n), respectively;
step 4.2: obtaining a projection energy H3 charsharewxpower (i) (i ═ 1.. 2n) and H3 charsharewxpower (i) (i ═ 1.. 2n) of each character binarized image X-direction projection H3 charsharewx (i) (i ═ 1.. 2n) and Y-direction projection H3 charsharewy (i) (i ═ 1.. 2n) of the first prefix number, according to the shadpower ═ k ∑ f (k) where k denotes the signal point position and f (k) denotes the signal point amplitude, and H3 charsharewypower (i) (i ═ 1.. 2 n); the projection energy of each character binarization image X-direction projection V3 charshardowx (i) (i 1.. 2n) and Y-direction projection V3 charshardowy (i) (i 1.. 2n) of the second prefix number is respectively marked as V3 charshardowxpower (i) (i 1.. 2n), and V3 charshardowypower (i) (i 1.. 2 n);
step 4.3: performing an absolute difference between the projection energy H3charshadow xpower (i) (i ═ 1.. 2n) in the X direction of each character binary image of the first prefix number and the projection energy V3charshadow ypower (i) (i ═ 1.. 2n) in the Y direction corresponding to each character binary image of the second prefix number, and recording as a feature T2; the projection energy H3CharShadowYpower (i) (i 1.. 2n) in the Y direction of the first-prefix-number binary image and the projection energy V3CharShadowXpower (i) (i 1.. 2n) in the X direction of the second-prefix-number binary image are subjected to absolute difference, and the absolute difference is recorded as a feature T3.
6. The double-crown-number-based paper money discriminating method according to claim 5, wherein the step 5: carrying out coupling degree statistics on the characteristic T1, the characteristic T2 and the characteristic T3, and judging whether the first crown word number is consistent with the second crown word number according to the coupling degree value so as to obtain an authenticity identification result of whether the corresponding paper money is changed, wherein the method specifically comprises the following steps:
step 5.1: the coupling statistics were performed according to the following formula:
FinalSameLevel(i)＝coef1*T1+coef1*T2+coef1*T3
wherein coef1+ coef2+ coef3 is 1;
step 5.2, judging the coupling degree value according to the following conditions,
when FinalJudge (i) is 1, the first crown number is different from the second crown number, and when FinalJudge (i) is 0, the two crown numbers are the same;
and 5.3, finally giving the counterfeit identification result of whether the paper money is changed.
7. A paper currency false distinguishing system based on double-crown word numbers comprises:
the image acquisition unit is used for acquiring image information of the paper money and respectively extracting a first crown word number area and a second crown word number area;
the image processing unit is used for carrying out size normalization on the first crown word number region and the second crown word number region and carrying out binarization processing to form a first crown word number binarization image and a second crown word number binarization image;
an image blocking and black point counting unit, which is used for carrying out the same at least 2n equal partition on the first crown word number binary image and the second crown word number binary image, wherein n is a natural number, calculating the proportion of black points in each partition area, carrying out absolute difference value counting on the proportion of the black points in each corresponding area of the first crown word number and the second crown word number in sequence, summing all the difference values, and recording the sum value as a characteristic T1;
an image black point projection and energy difference value calculation unit, which is used for carrying out black point projection in the X direction and the Y direction on the first crown word number binary image and the second crown word number binary image, then calculating projection energy in the X direction and the Y direction respectively, and carrying out absolute difference on the projection energy in the X direction of the first crown word number and the projection energy in the Y direction corresponding to the second crown word number, and marking as a characteristic T2; carrying out absolute difference on the projection energy in the Y direction of the first crown word number and the projection energy in the X direction corresponding to the second crown word number, and recording as a characteristic T3;
and the characteristic coupling degree judging unit is used for carrying out coupling degree statistics on the characteristic T1, the characteristic T2 and the characteristic T3 and judging whether the first crown word number is consistent with the second crown word number according to the coupling degree so as to obtain a counterfeit identification result of whether the corresponding paper money is changed.
8. The double crown-size based bill authentication system according to claim 7, wherein the image capturing unit further comprises a character region extracting unit for performing character region extraction on the first crown-size region and the second crown-size region.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710026017.8A CN106780953B (en) | 2017-01-13 | 2017-01-13 | Paper money counterfeit distinguishing method and system based on double-crown-word number |
| PCT/CN2018/071569 WO2018130119A1 (en) | 2017-01-13 | 2018-01-05 | Double prefix number-based paper money authenticity identification method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710026017.8A CN106780953B (en) | 2017-01-13 | 2017-01-13 | Paper money counterfeit distinguishing method and system based on double-crown-word number |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106780953A CN106780953A (en) | 2017-05-31 |
| CN106780953B true CN106780953B (en) | 2020-06-02 |
Family
ID=58945406
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710026017.8A Active CN106780953B (en) | 2017-01-13 | 2017-01-13 | Paper money counterfeit distinguishing method and system based on double-crown-word number |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN106780953B (en) |
| WO (1) | WO2018130119A1 (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106780953B (en) * | 2017-01-13 | 2020-06-02 | 广州广电运通金融电子股份有限公司 | Paper money counterfeit distinguishing method and system based on double-crown-word number |
| CN111046866B (en) * | 2019-12-13 | 2023-04-18 | 哈尔滨工程大学 | Method for detecting RMB crown word number region by combining CTPN and SVM |
| CN111445620B (en) * | 2020-03-09 | 2022-05-13 | 广州中智融通金融科技有限公司 | Storehouse cash storage method and device |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101809631B (en) * | 2007-09-27 | 2012-07-25 | 光荣株式会社 | Paper sheet processor |
| JP5213649B2 (en) * | 2008-11-12 | 2013-06-19 | 日立オムロンターミナルソリューションズ株式会社 | Paper sheet processing apparatus and paper sheet processing program |
| CN102890840B (en) * | 2012-08-22 | 2016-03-23 | 山东新北洋信息技术股份有限公司 | Banknote discriminating method and device |
| CN103136845B (en) * | 2013-01-23 | 2015-09-16 | 浙江大学 | A kind of Renminbi false distinguishing method based on crown word number characteristics of image |
| CN104916029A (en) * | 2014-03-13 | 2015-09-16 | 广州南沙资讯科技园有限公司博士后科研工作站 | Paper money verification system and paper money verification method based on system |
| CN204856686U (en) * | 2015-08-16 | 2015-12-09 | 浙江维融电子科技股份有限公司 | A pseudo - system of novel mirror for 2015 editions RMB |
| CN105335741A (en) * | 2015-10-28 | 2016-02-17 | 深圳怡化电脑股份有限公司 | Smudged serial number classification method and system |
| CN205670338U (en) * | 2016-06-17 | 2016-11-02 | 光荣电子工业（苏州）有限公司 | There is the cleaning-sorting machine of the splicing pseudo-function of coin inspection |
| CN106780953B (en) * | 2017-01-13 | 2020-06-02 | 广州广电运通金融电子股份有限公司 | Paper money counterfeit distinguishing method and system based on double-crown-word number |
-
2017
- 2017-01-13 CN CN201710026017.8A patent/CN106780953B/en active Active
-
2018
- 2018-01-05 WO PCT/CN2018/071569 patent/WO2018130119A1/en active Application Filing
Also Published As
| Publication number | Publication date |
|---|---|
| WO2018130119A1 (en) | 2018-07-19 |
| CN106780953A (en) | 2017-05-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP5344668B2 (en) | Method for automatically confirming securities media item and method for generating template for automatically confirming securities media item | |
| CN103761799B (en) | A kind of bill anti-counterfeit method based on texture image feature and device | |
| CN102542660B (en) | Bill anti-counterfeiting identification method based on bill watermark distribution characteristics | |
| CN104616392B (en) | A kind of paper money discrimination method based on local binary patterns | |
| CN104680161A (en) | Digit recognition method for identification cards | |
| Mirza et al. | Paper currency verification system based on characteristic extraction using image processing | |
| CN107103683B (en) | Paper money identification method and device, electronic equipment and storage medium | |
| CN106780953B (en) | Paper money counterfeit distinguishing method and system based on double-crown-word number | |
| CN107331026B (en) | Paper money identification method and device | |
| CN107067533A (en) | The method and device that a kind of bank note differentiates | |
| CN104050745A (en) | High-speed coin sorting technology based on image identification | |
| EP3417429A1 (en) | Method and apparatus for detection of document tampering | |
| CN103886309A (en) | Method for identifying dollar denominations through facial recognition | |
| CN106845469B (en) | Paper money identification method and device | |
| CN104537364A (en) | Dollar bill denomination and edition identifying method based on texture analysis | |
| CN107742359B (en) | banknote identification method and device, terminal equipment and readable storage medium | |
| CN106447904B (en) | Method and device for identifying paper money | |
| CN107527423B (en) | Paper money anti-counterfeiting identification method and paper money identification device | |
| CN103646458B (en) | The method of the principal component analysis identification note true and false | |
| CN108269349A (en) | A kind of bill handling method, apparatus, equipment and storage medium | |
| CN103035061B (en) | Anti-counterfeit characteristic generation method of valuable file and identification method and device thereof | |
| CN103955988A (en) | Paper money transaction device | |
| KR102094234B1 (en) | Serial number recognition Apparatus and method for paper money | |
| CN108154596B (en) | Double-crown-number paper currency discrimination method based on image matching | |
| CN109410420B (en) | Image detection method and device and bill discriminator |
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 |