CN104050749A - Method for making judgment by automatically selecting different feature points of banknotes - Google Patents
Method for making judgment by automatically selecting different feature points of banknotes Download PDFInfo
- Publication number
- CN104050749A CN104050749A CN201310128538.6A CN201310128538A CN104050749A CN 104050749 A CN104050749 A CN 104050749A CN 201310128538 A CN201310128538 A CN 201310128538A CN 104050749 A CN104050749 A CN 104050749A
- Authority
- CN
- China
- Prior art keywords
- bank note
- true
- qualification
- average waveform
- information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Inspection Of Paper Currency And Valuable Securities (AREA)
Abstract
The invention discloses a method for making judgment by automatically selecting different feature points of banknotes. The method comprises the following steps: a, authentication feature information and authenticity identification information of a real banknote are acquired; b, first image information of a banknote to be authenticated is acquired; c, the face value and direction of the banknote to be authenticated are judged; d, second image information of the banknote to be authenticated is acquired; and e, the second image information is compared with authenticity identification information corresponding to the banknote to complete authenticity identification. According to the invention, the face value and entering direction of the banknote can be judged according to the first image information, authenticity identification equipment can access authenticity identification information of a real banknote corresponding to the face value and the entering direction, and by reading the authenticity identification information, a processor can determine the number, specific position points and width of feature points for banknote authenticity judgment in the transverse direction and/or the longitudinal direction so that different feature points can be automatically selected to read and authenticate a second image to finally complete authenticity judgment.
Description
Technical field
The present invention relates to a kind of forge or true or paper money authentication method, particularly relate to a kind of method of automatic selection bank note different characteristic point judgement.
Background technology
Along with scientific and technological development, present many public places, in order to meet people's convenience and agility demand, are all provided with vending machine, money exchanger or deposit machine etc., thereby have significantly saved the configuration of human resources.But counterfeit money spreads unchecked, the requirement of the resolving ability to forge or true or paper money is more and more higher.In order to promote the resolving ability of forge or true or paper money, relevant dealer has been developed the various judgment means for banknote image feature, for example: utilize Infrared irradiation bank note, obtain luminous flux information by photoelectric sensor, in the prior art because be subject to the fixing restriction in position in the quantity of sensor or scanning process, so the unique point of more options fixed position and the mode standard of genuine notes are compared, to judge the true and false of bank note.Although above prior art can reach the object that judges forge or true or paper money really, but the unique point of fixed position not necessarily can contain every kind of bank note face amount than more sensitive unique point, thereby reduce the judgement of forge or true or paper money, therefore reduced the discrimination of bank note.Need badly now one can be by reading the identification information of the crucial identifying position of difference of different bank note (comprising bill etc.) the banknotes identifying method of different bank note being carried out to accurate identification.
Summary of the invention
Technical matters to be solved by this invention is that a kind of forge or true or paper money authentication method with higher identification capability is provided.
For solving above technical matters, technical scheme of the present invention is: a kind of method of automatic selection bank note different characteristic point judgement, comprises the steps:
A, obtain true bank note at different denominations from identification mark information under different bank note input direction situations and authenticity information and store in advance;
B, obtain intend qualification bank note the first image information;
C, the identification mark information obtaining by step a and described the first image information of obtaining by step b are compared, to judge face amount and the direction of intending qualification bank note;
D, obtain and intend qualification bank note the second image information;
E, select the forge or true or paper money authentication information of corresponding face amount and bank note input direction according to the result of determination of step c, the second image information and this forge or true or paper money authentication information that steps d is obtained are compared, thereby complete the authenticity of bank note;
Concrete, described step c is:
C1, get the first image information in the specific region of the plan qualification bank note obtaining in step b as sample one;
C2, sample one and the identification mark information of the true bank note of all face amounts are compared, record each by the first related coefficient and organize comparison result;
C3, after completing, all comparisons obtain maximum the first related coefficient A;
C4, judge whether maximum the first related coefficient A is greater than designated value, be greater than and can judge that face amount, the bank note input direction of plan qualification bank note are identical with the object of comparing with it, if maximum the first related coefficient A is less than designated value, explanation is intended qualification bank note not within the scope of qualification.
Wherein, the valuable papers of described " bank note " for being copied, includes but not limited to banknote, stamp, duty stamp, stock, bond, treasury stock, commercial paper, acceptance bill, bank's time deposit certificate etc.
As improvement of the present invention, described steps d specifically comprises the steps:
D1, determine that according to the face amount of the definite plan of step c qualification bank note and bank note input direction information photoelectric sensor is to particular location and the width intending identifying that bank note the second image information is distinguished;
D2, radiation source irradiate bank note from intending qualification bank note one side, by being placed in described the second image information under the determined particular location of photoelectric sensor read step d1 and the width of intending qualification bank note opposite side as sample two.
Further improve as the present invention, described step e specifically comprises the steps:
E1, by sample two and face amount and bank note input direction, all the authenticity information of identical true bank note is compared with intending qualification bank note, records each organize comparison result by the second related coefficient;
E2, the mean value of obtaining all the second related coefficients after all comparisons complete are the second related coefficient B;
E3, judge whether described the second related coefficient B is greater than designated value, be greater than and can judge and intend identifying that bank note is as true bank note, if described the second related coefficient B is less than designated value, explanation is intended identifying that bank note is false bank note.
As further improvement of the present invention, described step b specifically comprises the steps:
B1, determine reading position and reading width of the first image information;
B2, radiation source irradiate and intend qualification bank note from intending qualification bank note one side, intend by being placed in that the photoelectric sensor read step b1 of qualification bank note opposite side is determined to be read position and read described the first image information under width.
Further improve as the present invention, described step a specifically comprises the steps:
A1, set all true bank note particular location at identification mark information place and read width under four bank note input direction situations respectively;
A2, set all true bank note particular location at authenticity information place and read width under four bank note input direction situations respectively;
A3, radiation source irradiate described true bank note from described true bank note one side respectively, by being placed in the photoelectric sensor read step a1 of true bank note opposite side and particular location that a2 sets and reading the luminous flux information under width, this luminous flux information exchange is crossed and formed respectively the identification mark information of true bank note and authenticity information storage after analog to digital converter conversion in storer.
Wherein, described identification mark information or the first image information are average waveform A, and described average waveform A obtains as follows:
A310, the P of photoelectric sensor from setting along the Y-direction of true bank note or plan qualification bank note
0position starts to read the luminous flux of true bank note or plan qualification bank note, and reading width is W
0;
A311, processor identify that to true bank note or plan bank note is from P
0position starts to W
0the view data of width range averages calculating, obtains a line view data, and described view data forms average waveform A along true bank note or plan qualification bank note directions X.
And described authenticity information or the second image information are average waveform B and/or C, or be the set of average waveform B and/or C, the set of described average waveform B or average waveform B obtains by following steps:
A320, the preliminary election P of photoelectric sensor from setting along the Y-direction of true bank note or plan qualification bank note
ymposition starts to read the luminous flux of true bank note or plan qualification bank note, and reading preselected width is W
ym,
A321, processor identify that to true bank note or plan bank note is from preliminary election P
ymposition starts to preliminary election W
ymthe view data of width range averages calculating, obtains a line view data, and described view data forms average waveform B along bank note horizontal direction
m, wherein
mfor identify described average waveform B quantity, be more than or equal to 1 natural number;
The set of described average waveform C or average waveform C obtains by following steps:
A330, the preliminary election P of photoelectric sensor from setting along the directions X of true bank note or plan qualification bank note
xnposition starts to read the luminous flux of true bank note or plan qualification bank note, and reading preselected width is W
xn,
A331, processor identify that to true bank note or plan bank note is from preliminary election P
xnposition starts to preliminary election W
xnthe view data of width range averages calculating, obtains a line view data, and described view data forms average waveform Cn along true bank note or plan qualification bank note Y-direction, wherein
nfor identify described average waveform C quantity, be more than or equal to 1 natural number;
The average waveform A of described true bank note and described average waveform B or its set, described average waveform C or its set, or corresponding with the common set forming of average waveform C with average waveform B.
As optimal way of the present invention, described processor is digital processing unit DSP, microprocessor MPU or System on chip SoC; Described photoelectric sensor is several linearly aligned CIS contact-type image sensors, CMOS or ccd image sensor; Described radiation source is one or more groups infrared light, visible ray and/or the ultraviolet light that can carry out penetration or reflective irradiation.
By implement the present invention desirable following beneficial effect:
By implementing method and technology scheme of the present invention, first genuine/counterfeit discriminating equipment can judge face amount and the approach axis of bank note by the first image information, genuine/counterfeit discriminating equipment can be transferred the authenticity information of the true bank note under this corresponding face amount and approach axis situation, by reading this authenticity information, processor can be identified for the unique point of forge or true or paper money judgement at quantity, particular location point and width laterally and/or longitudinally, thereby automatically selects preliminary election different characteristic point to carry out reading of the second image.Afterwards, processor is by asking for the second related coefficient and the concrete differentiation of the second related coefficient B being judged to the true and false of intending qualification bank note.
The present invention is by implementing above method, be actually corresponding with average waveform A to the face amount of bank note, direction, by the true and false information of bank note with average waveform B or its set, average waveform C or its set, or corresponding with average waveform B and the common set forming of average waveform C, by asking for the first related coefficient and the differentiation of maximum the first related coefficient A being judged to face amount and the direction of intending qualification bank note, by asking for the second related coefficient and the concrete differentiation of the second related coefficient B being judged to the true and false of intending qualification bank note.Because the average waveform A of true bank note and described average waveform B or its set, described average waveform C or its set, average waveform B are with one of set of the common formation of average waveform C or combine corresponding, thereby forge or true or paper money discriminating gear is judged and is intended can getting the authenticity information of judging this forge or true or paper money with average waveform B corresponding to average waveform A or one of its set, described average waveform C or its set, average waveform B and the common set forming of average waveform C or combination after the qualification face amount of bank note and direction.Like this, forge or true or paper money discriminating gear can be according to the face amount of bank note and direction, identify that to intending position and width that bank note is set as best preliminary election identification than more sensitive unique point carry out authenticity, instead of defend tenaciously in one and specifically judge position, thereby judgment mode is more flexible, be conducive to make full use of the difference of different denominations bank note characteristic point position under the different situations of approach axis, directly obtain than more sensitive unique point image, thereby can greatly improve the rate of distinguishing of forge or true or paper money.
Brief description of the drawings
Below in conjunction with Figure of description, the present invention will be further described in detail, wherein:
Fig. 1 is the forge or true or paper money discriminating gear structural representation of implementing the inventive method;
Fig. 2 is the scanning head module structural representation of implementing the forge or true or paper money discriminating gear of the inventive method;
Fig. 3 is the forge or true or paper money discriminating gear circuit connection diagram of implementing the inventive method;
Fig. 4 implements imageing sensor in the inventive method process to read the output waveform figure of the positive a certain position of bank note;
Fig. 5 implements imageing sensor in the inventive method process to read the output waveform figure of a certain position of bank note reverse side;
Fig. 6 implements the schematic diagram that in the inventive method process, selection different characteristic point carries out true and false judgement automatically;
Fig. 7 implements the schematic flow sheet that the inventive method is carried out forge or true or paper money judgement;
Fig. 8 is the schematic flow sheet of implementing to extract in the inventive method process sample one;
Fig. 9 is the schematic flow sheet of implementing to judge in the inventive method process bank note face amount and approach axis;
Figure 10 is the schematic flow sheet that extracts sample two and carry out forge or true or paper money judgement in enforcement the inventive method process;
Figure 11 is the division schematic diagram to 10 yuan note unique points;
Figure 12 judges the schematic diagram a of bank note face amount and bank note input direction by sample one;
Figure 13 judges the schematic diagram b of bank note face amount and bank note input direction by sample one;
Figure 14 judges the schematic diagram c of bank note face amount and bank note input direction by sample one;
Figure 15 judges the schematic diagram d of bank note face amount and bank note input direction by sample one;
Figure 16 is the schematic diagram a that judges forge or true or paper money by sample two;
Figure 17 is the schematic diagram b that judges forge or true or paper money by sample two;
Figure 18 is the schematic diagram c that judges forge or true or paper money by sample two.
Embodiment
As Fig. 1, shown in 3, a kind of forge or true or paper money discriminating gear of the method for applying automatic selection bank note different characteristic point judgement, comprise bank note placement platform 10, be arranged on the photoelectric sensor 21 of described bank note placement platform 10 1 sides, also comprise the infrared transmitter 101 that is arranged on described bank note placement platform 10 opposite sides, in order to detect bank note 102 and the photoelectric encoder 204 of described photoelectric sensor 21 because of the change in location information of caused by relative motion, the processor 202 being connected with described displacement detector (not shown) and the digital to analog converter 201 being connected with described processor 202 and in order to store the storer 206 of banknotes identifying correlation data, described photoelectric sensor 21 connects with described digital to analog converter 201, the described photoelectric sensor 21 contact-type image sensor group that the contact-type image sensor 103 of several single arrangements forms of serving as reasons.The orientation of described contact-type image sensor 103 is with to be placed on bank note 102 on described bank note placement platform vertical with the direction of relative movement of described photoelectric sensor 21.Banknotes identifying correlation data in described storer 206 comprises the standard identification information of different bank note and the identification information of corresponding denomination and bank note approach axis, and different bank note refers to the Currency Type difference of bank note and/or the denomination difference of bank note.Bank note transport establishment 203 drives bank note 102 to move to realize relative motion between bank note 102 and described photoelectric sensor 21.Described bank note conveying mechanism is driven by motor driver 205.
Wherein, described processor is digital processing unit DSP, microprocessor MPU or System on chip SoC; Described photoelectric sensor is several linearly aligned CIS contact-type image sensors, CMOS or ccd image sensor.As the radiation source of bank note, it is one or more groups infrared light, visible ray and/or ultraviolet light that can carry out penetration or reflective irradiation.In the present embodiment, radiation source is selected infrared light, and this infrared light obtains by infrared transmitter 101.
As shown in Figure 3, bank note transport establishment 203 is controlled by motor driver 205, and feeds back bank note displacement information by photoelectric encoder 204, realizes transmission and the scanning of bank note; The infrared light that utilizes contact-type image sensor 103 to obtain infrared transmitter 101 penetrated the luminous flux information (or being expressed as graphical information) of bank note 102, convert digital quantity to through analog to digital converter 201 and be sent to processor 202, automatically select different characteristic point and the mode standard being stored on storer 206 to compare, judge bank note face amount and the true and false.
Described contact-type image sensor 103 is linear contact-type image sensor, and it is one group and has hundreds of continuous arrangements photoelectric sensor together, can detect trickle image change; Motor driver 205 drives bank note transport establishment 203 to transmit bank note 102, and the scanning of feeding back displacement information and complete bank note 102 complete images by photoelectric encoder 201, simultaneously, contact-type image sensor 103 receives the infrared light that infrared transmitter 101 launches and penetrates the image that the specialty inks on bank note 102 and bank note transmits, and converts digital quantity to be transferred to processor 202 and judge processing through digital to analog converter 201.
Described photoelectric sensor 21 adopts linear contact-type image sensor 103 to ensure to implement device of the present invention to have reduced again the manufacturing cost of described photoelectric sensor 21 in can reading all identification informations of bank note, thereby is conducive to of the present invention applying.Infrared transmitter 101 can ensure that as light source the image information that photoelectric sensor 21 reads is more accurate, thereby has ensured the accuracy of identification.
Refer to the 2nd figure, this figure is the scanning head module structural representation of implementing the forge or true or paper money discriminating gear of the inventive method.This scanning head module is applicable to detect the specialty inks of bank note, and this module at least includes an infrared transmitter 101 and a contact-type image sensor 103; Wherein, infrared transmitter 101 and contact-type image sensor 103 are separately positioned on the both sides of bank note 102, infrared transmitter 101 is for launching infrared light, and after transmission imaging, the image of the specialty inks on bank note 102 is touched formula imageing sensor 103 and receives.
As shown in Figure 4, Figure 5, wherein, Fig. 4 implements imageing sensor in the inventive method process to read the output waveform figure of the positive a certain position of bank note, and Fig. 5 implements imageing sensor in the inventive method process to read the output waveform figure of bank note reverse side relevant position.Wherein, transverse axis represents arranging of pixel, and the longitudinal axis represents the luminous flux number percent of different pixels point.Due to the difference of ink and/or the pattern etc. of different pixels point, the luminous flux number percent of different pixels point is different.According to the difference of luminous flux part ratio, can remittance abroad one oscillogram along the arragement direction of different pixels point.Waveform shown in Fig. 5 be corresponding with Fig. 4, same card Securities position is identical but luminous flux number percent oscillogram under reverse side input situation.Particularly, bank note for example bank note when its back side is upwards time, the oscillogram that certain ad-hoc location oscillogram and bank note face up is just reverse, so the direction that left enters and the right-hand direction entering have two mutual reverse ad-hoc location oscillograms, one has 4 different waveforms, and corresponding 4 kinds of bank note input directions are note input direction.
As shown in Figure 6, by the W of the ad-hoc location of the bank note facing up
0after wide number of scanning lines does on average, obtain the waveform of Fig. 4, wherein suppose that every bank note sweep trace has 800 pixels, the number percent of the luminous flux of each pixel is different obviously.Particularly, automatically select the judgement of different characteristic point, in two stages, stage one: judge bank note face amount and note input direction, note input direction be generally divided into 4 may, be respectively that the left that faces up enters, right-hand the entering that face up, the back side upward left enters and the back side right-hand enter etc. upward.Stage two: differentiate forge or true or paper money with specific feature locations and width.Process needed sample for the judgement in stage one and be referred to as stage one sample A, process needed sample for the judgement in stage two and be referred to as stages two sample B and/or C.The quantity of stage one sample only has one, its position P
0and width W
0fix P
0conventionally be less than 1/4th of bank note; Quantity, position and the width of stages two sample are unfixed, have independently definition for each note input direction of every kind of face amount.After extraction stage one sample, judge bank note face amount and note input direction, again according to quantity, position and the width information of face amount and horizontal and vertical sample of note input orientation determination stages two, finally utilize the mode standard of stages two sample and storage to compare, judge forge or true or paper money.
Shown in Fig. 3, in the time that bank note 102 is transferred to contact-type image sensor 103 by bank note transport establishment 203, photoelectric encoder 204 is counting up to P
0, given an order by processor 202, by W when the position
0wide number of scanning lines is average standard feature figure rear and that be stored in storer 206 and compares.5 kinds of values of money as preset in equipment, the standard feature figure being stored in storer 206 has 4=20 comparison figures of 5 X at least.The representative of closest figure is belonged to the differentiation which kind of value of money and direction send processor 202 to and do next round forge or true or paper money after equipment has been compared all characteristic patterns.
As shown in Figure 7, the judgement of the true and false of bank note can comprise the following step on the whole: step 601: extraction stage one sample; Step 602: judge bank note face amount and note input direction; Step 603: horizontal and vertical sample of extraction stage two; Step 604: judge forge or true or paper money.It should be noted that, the extraction to stages two sample in step 603, can only extract horizontal sample or longitudinal sample in stage two, also can not only extract the horizontal sample in stage two but also extract longitudinal sample.In Fig. 7 embodiment, what adopt is the situation of not only having extracted the horizontal sample in stage two but also having extracted longitudinal sample.
The position of stage one sample is specific, is conventionally set in 1/4th regions that bank note scans the earliest.As bank note has 1200 sweep traces, P
0and P
0+ W
0all be set between 1 to 1200/4=300, such as P
0be made as 100, W
0be made as 10.Due to four of scanning paper currency at the most/once after can get stage one sample, so, before all scannings that complete bank note, can carry out and complete the judgement of bank note face amount and note input direction.
As shown in Figure 8, the leaching process of stage one sample comprises the following step: step 701: the signal that judges whether to receive reading images; If so, carry out step 702; If not, carry out step 701; Step 702: read a line view data; Step 703: judge whether to detect bank note boundary line; If so, carry out step 704; If not, carry out step 705; Step 704: judge apart from boundary line P
0+ W
0whether the image of row is accepted complete; If so, carry out step 706; If not, carry out step 701; Step 705: detect bank note boundary line; Step 706: take out from P
0to P
0+ W
0banknote image, as stage one sample.
Shown in Fig. 3, processor 202 receives the pulse signal of photoelectric encoder 204 and counts, and triggers the operation of reading images in the time counting down to specific quantity, when bank note 102 moves specific range, starts to read a line view data.Non-bank note region in image is due to by infrared light direct irradiation, its gradation of image value approaches saturation region, and the bank note region logical light quantity certain because bank note has blocked, its gradation of image value is less than the gradation of image value in non-bank note region, so there is obvious feature bank note boundary line, can detect the boundary line of bank note according to this feature.Detect behind the Article 1 widthwise edge boundary line of bank note, taking out is P apart from this position, boundary line
0, width is W
0view data, and remove the data of non-bank note part wherein, as stage one sample.
As shown in Figure 9, the flow process that judges bank note (for example bill) face amount and approach axis comprises the following step: step 801: the average waveform of calculation stages one sample is average waveform A; Step 802: compare with the corresponding reference model of certain note input direction of certain face amount, ask i.e. the first related coefficient of its correlation parameter; Step 803: whether four note input directions that judge this face amount are compared complete; If so, carry out step 804; If not, carry out step 802; Step 804: judge whether that all face amount comparisons are complete; If so, carry out step 805; If not, carry out step 802; Step 805: ask maximum correlation coefficient to ask maximum the first related coefficient A; Step 806: whether maximum correlation coefficient is greater than designated value; If so, carry out step 807; If not, carry out step 808; Step 807: determine face amount and note input direction; Step 808: non-designated bank note.
To the W of stage one sample
0row view data averages calculating, obtain a line view data, be average waveform A(waveform as shown in Fig. 4 or Fig. 5), then compare with the waveform of the reference model of four note input directions of all appointment face amounts, calculate its first related coefficient, finally get maximum the first related coefficient A wherein, if maximum the first related coefficient A is greater than designated value, can obtain face amount and the note input direction of bank note.
As shown in figure 10, implement to extract sample two in the inventive method process and carry out the flow process that forge or true or paper money judges to comprise the following step: taking bill as example, step 901: the position and the width that obtain all samples of stages two according to face amount and note input direction; Step 902: extraction stages two sample to calculate its average waveform be average waveform B or average waveform C; Step 903: compare with corresponding mode standard, ask its related coefficient to ask for the second related coefficient; Step 904: judge whether horizontal and vertical all samples are compared complete; If so, carry out step 904; If not, carry out step 902; Step 905: calculate the i.e. mean value of the second related coefficient of all related coefficients; Step 906: whether average correlation coefficient i.e. the second related coefficient B is greater than designated value; If so, carry out step 907; If not, carry out step 908; Step 907: bank note is true; Step 908: bank note is false.
Shown in Fig. 3, according to bill face amount and note input direction thereof, read the information such as quantity, position and the width of corresponding all samples of stages two from storer 206, then extract all samples, and calculate their average waveform B and/or average waveform C, the waveform of corresponding mode standard in these average waveform and storer 206 is compared, calculate its second related coefficient, finally calculate the i.e. second related coefficient B of mean value of all the second related coefficients, if the second related coefficient B is greater than certain designated value, bank note is true, otherwise bank note is false.
In conjunction with Fig. 6 to Figure 10, a complete description specific embodiment of the present invention is as follows: a kind of method of automatic selection bank note different characteristic point judgement, comprises the steps:
A, obtain true bank note at different denominations from identification mark information under different bank note input direction situations and authenticity information and store in advance;
B, obtain intend qualification bank note the first image information;
C, the identification mark information obtaining by step a and described the first image information of obtaining by step b are compared, to judge face amount and the direction of intending qualification bank note;
D, obtain and intend qualification bank note the second image information;
E, select the forge or true or paper money authentication information of corresponding face amount and bank note input direction according to the result of determination of step c, the second image information and this forge or true or paper money authentication information that steps d is obtained are compared, thereby complete the authenticity of bank note;
Concrete, described step c is:
C1, get the first image information in the specific region of the plan qualification bank note obtaining in step b as sample one;
C2, sample one and the identification mark information of the true bank note of all face amounts are compared, record each by the first related coefficient and organize comparison result;
C3, after completing, all comparisons obtain maximum the first related coefficient A;
C4, judge whether maximum the first related coefficient A is greater than designated value, be greater than and can judge that face amount, the bank note input direction of plan qualification bank note are identical with the object of comparing with it, if maximum the first related coefficient A is less than designated value, explanation is intended qualification bank note not within the scope of qualification.
Wherein, the valuable papers of described " bank note " for being copied, include but not limited to banknote, stamp, duty stamp, stock, bond, treasury stock, commercial paper, acceptance bill, bank's time deposit certificate etc., but in the present embodiment, select banknote to be specifically addressed.The present embodiment illustrates and should not be construed as limiting the scope of the present invention the selectivity of bank note concrete manifestation form.
Described steps d specifically comprises the steps:
D1, determine that according to the face amount of the definite plan of step c qualification bank note and bank note input direction information photoelectric sensor is to particular location and the width intending identifying that bank note the second image information is distinguished;
D2, radiation source irradiate bank note from intending qualification bank note one side, by being placed in described the second image information under the determined particular location of photoelectric sensor read step d1 and the width of intending qualification bank note opposite side as sample two.
Accordingly, described step e specifically comprises the steps:
E1, be that note input direction is all compared with the authenticity information of intending the identical true bank note of qualification bank note by sample two and face amount and bank note input direction, record each by the second related coefficient and organize comparison result;
E2, the mean value of obtaining all the second related coefficients after all comparisons complete are the second related coefficient B;
E3, judge whether described the second related coefficient B is greater than designated value, be greater than and can judge and intend identifying that bank note is as true bank note, if described the second related coefficient B is less than designated value, explanation is intended identifying that bank note is false bank note.
Accordingly, described step b specifically comprises the steps:
B1, determine reading position and reading width of the first image information;
B2, radiation source irradiate and intend qualification bank note from intending qualification bank note one side, intend by being placed in that the photoelectric sensor read step b1 of qualification bank note opposite side is determined to be read position and read described the first image information under width.
Accordingly, described step a specifically comprises the steps:
A1, set all true bank note particular location at identification mark information place and read width under four bank note input direction situations respectively;
A2, set all true bank note particular location at authenticity information place and read width under four bank note input direction situations respectively;
A3, radiation source irradiate described true bank note from described true bank note one side respectively, by being placed in the photoelectric sensor read step a1 of true bank note opposite side and particular location that a2 sets and reading the luminous flux information under width, this luminous flux information exchange is crossed and formed respectively the identification mark information of true bank note and authenticity information storage after analog to digital converter conversion in storer.
Wherein, described identification mark information or the first image information are average waveform A, and described average waveform A obtains as follows:
A310, the P of photoelectric sensor from setting along the Y-direction of true bank note or plan qualification bank note
0position starts to read the luminous flux of true bank note or plan qualification bank note, and reading width is W
0;
A311, processor identify that to true bank note or plan bank note is from P
0position starts to W
0the view data of width range averages calculating, obtains a line view data, and described view data forms average waveform A along true bank note or plan qualification bank note directions X.
And described authenticity information or the second image information are average waveform B and/or C, or be the set of average waveform B and/or C, the set of described average waveform B or average waveform B obtains by following steps:
A320, the preliminary election P of photoelectric sensor from setting along the Y-direction of true bank note or plan qualification bank note
ymposition starts to read the luminous flux of true bank note or plan qualification bank note, and reading preselected width is W
ym,
A321, processor identify that to true bank note or plan bank note is from preliminary election P
ymposition starts to preliminary election W
ymthe view data of width range averages calculating, obtains a line view data, and described view data forms average waveform B along bank note horizontal direction
m, wherein
mfor identify described average waveform B quantity, be more than or equal to 1 natural number;
The set of described average waveform C or average waveform C obtains by following steps:
A330, the preliminary election P of photoelectric sensor from setting along the directions X of true bank note or plan qualification bank note
xnposition starts to read the luminous flux of true bank note or plan qualification bank note, and reading preselected width is W
xn,
A331, processor identify that to true bank note or plan bank note is from preliminary election P
xnposition starts to preliminary election W
xnthe view data of width range averages calculating, obtains a line view data, and described view data forms average waveform Cn along true bank note or plan qualification bank note Y-direction, wherein
nfor identify described average waveform C quantity, be more than or equal to 1 natural number;
The average waveform A of described true bank note and described average waveform B or its set, described average waveform C or its set, or corresponding with the common set forming of average waveform C with average waveform B.
As the preferred embodiments of the present invention, described processor is digital processing unit DSP, microprocessor MPU or System on chip SoC; Described photoelectric sensor is several linearly aligned CIS contact-type image sensors, CMOS or ccd image sensor; Described radiation source is one or more groups infrared light, visible ray and/or the ultraviolet light that can carry out penetration or reflective irradiation.
The bank note that is 10 yuan below by a face amount illustrates the face amount and the bank note input direction that how judge bank note by method of the present invention, and and then judges the true and false of bank note.
As shown in figure 11, on this selected bank note, set position and the width information of sample as follows:
P
0 : 250
W
0 : 30
m : 2
P
y1 : 350 P
y2 : 680
W
y1 : 45 W
y2 : 40
n : 1
P
x1 : 360
W
x1 : 40
The first related coefficient and the second related coefficient that the present invention is alleged are defined as follows: X is reference sample data, and Y is test sample book data:
The absolute value of related coefficient is larger, shows that X and the Y degree of correlation are higher.In the time that X is relevant to Y linear positive, related coefficient value is 1.
The designated value of using while being related coefficient comparison for the comparison threshold value of each reference sample is different, is the empirical value that Experiment Training obtains.
To test bank note fixed position P
0be 250 and fixed width W
0the reference sample data that are 4 note input directions of the bank note of 30 test sample book data and all face amounts are compared, find out the sample of related coefficient maximum, this example maximum correlation coefficient i.e. the first related coefficient A is 0.930, and be greater than comparison threshold value 0.87, meet the requirements, so the face amount of this bank note is 10 yuan, note input direction is oppositely positive.If greatest coefficient is not greater than comparison threshold value, this bank note is counterfeit money or illustrates that this test bank note is not within the scope of qualification.
Be the bank note of 10 yuan as Figure 12-14 are respectively face amount, under different input direction situations, the average waveform A obtaining by the inventive method, wherein dotted line is the average waveform A of identification mark information, solid line is the average waveform A of the first image information.
In the time that the first image information data is compared with the positive reverse identification mark information data of true bank note, the first related coefficient obtaining by the inventive method is that 0.931> threshold value 0.87(average waveform A comparison situation is as shown in 12); In the time that positive to the first image information data and true bank note reverse identification mark information data is compared, the first related coefficient obtaining by the inventive method is for-0.460(average waveform A comparison situation is as shown in 13); In the time that the first image information data is compared with the identification mark information data of true bank note reverse side forward, the first related coefficient obtaining by the inventive method is for-0.194(average waveform A comparison situation is as shown in 14); In the time that the first image information data is compared with the identification mark information data of true bank note front forward, the first related coefficient obtaining by the inventive method is that 0.173(average waveform A comparison situation is as shown in 15).
In the comparison of above-mentioned four kinds of situations, the first maximum related coefficient is 0.931 to be the first related coefficient A, and in view of this first related coefficient, A is greater than designated value 0.87, therefore can judge that test bank note oppositely inputs as front.
Certainly,, in the time specifically judging, different denominations is different from its first waveform of different input direction bank note A.By the judgement to the first related coefficient A, not only can obtain the input direction information of bank note, can also obtain the face amount information of bank note.
By method of the present invention, can obtain the authenticity information of true bank note obtaining after the test face amount of bank note and input direction information, thereby can enter next step forge or true or paper money judgement.
As shown in figure 12, get two, horizontal sample, longitudinally 1, sample.By sample two and face amount and bank note input direction, all the authenticity information of identical true bank note is compared with intending qualification bank note, can obtain the second related coefficient 0.906(position and width information and be: P
y1: 350, W
y1: 45, average waveform B comparison situation is as shown in 16), the second related coefficient 0.965(position and width information be: P
y2: 680, W
y2: 40, average waveform B comparison situation is as shown in 17), the second related coefficient 0.937(position and width information be: P
x1: 360, W
x1: 40, average waveform C comparison situation is as shown in 18).
The mean value of asking for above three the second related coefficients is that the second related coefficient B is (0.906+0.965+0.937)/3=0.936, and this second related coefficient B is greater than designated value and compares threshold value 0.88, and wherein comparing threshold value is the empirical value that Experiment Training obtains.Therefore, can judge that test bank note is true bank note.
Must be pointed out, some indefinitenesses that above-described embodiment is just made the present invention illustrate.But person of skill in the art will appreciate that, do not departing under aim of the present invention and scope, can the present invention be made amendment, replaces and be changed, these amendments, replacement and change still belong to protection scope of the present invention.
Claims (11)
1. a method of automatically selecting the judgement of bank note different characteristic point, is characterized in that comprising the steps:
A, obtain true bank note at different denominations from identification mark information under different bank note input direction situations and authenticity information and store in advance;
B, obtain intend qualification bank note the first image information;
C, the identification mark information obtaining by step a and described the first image information of obtaining by step b are compared, to judge face amount and the direction of intending qualification bank note;
D, obtain and intend qualification bank note the second image information;
E, select the forge or true or paper money authentication information of corresponding face amount and bank note input direction according to the result of determination of step c, the second image information and this forge or true or paper money authentication information that steps d is obtained are compared, thereby complete the authenticity of bank note;
Concrete, described step c is:
C1, get the first image information in the specific region of the plan qualification bank note obtaining in step b as sample one;
C2, sample one and the identification mark information of the true bank note of all face amounts are compared, record each by the first related coefficient and organize comparison result;
C3, after completing, all comparisons obtain maximum the first related coefficient A;
C4, judge whether maximum the first related coefficient A is greater than designated value, be greater than and can judge that face amount, the bank note input direction of plan qualification bank note are identical with the object of comparing with it, if maximum the first related coefficient A is less than designated value, explanation is intended qualification bank note not within the scope of qualification.
2. the method for a kind of automatic selection bank note different characteristic point judgement according to claim 1, is characterized in that: described steps d specifically comprises the steps:
D1, determine that according to the face amount of the definite plan of step c qualification bank note and bank note input direction information photoelectric sensor is to particular location and the width intending identifying that bank note the second image information is distinguished;
D2, radiation source irradiate bank note from intending qualification bank note one side, by being placed in described the second image information under the determined particular location of photoelectric sensor read step d1 and the width of intending qualification bank note opposite side as sample two.
3. the method for a kind of automatic selection bank note different characteristic point judgement according to claim 2, is characterized in that: described step e specifically comprises the steps:
E1, by sample two and face amount and bank note input direction, all the authenticity information of identical true bank note is compared with intending qualification bank note, records each organize comparison result by the second related coefficient;
E2, the mean value of obtaining the second related coefficient after all comparisons complete are the second related coefficient B;
E3, judge whether described the second related coefficient B is greater than designated value, be greater than and can judge and intend identifying that bank note is as true bank note, if described the second related coefficient B is less than designated value, explanation is intended identifying that bank note is false bank note.
4. according to the method for a kind of automatic selection bank note different characteristic point judgement described in any one in claims 1 to 3, it is characterized in that: described step b specifically comprises the steps:
B1, determine reading position and reading width of the first image information;
B2, radiation source irradiate and intend qualification bank note from intending qualification bank note one side, intend by being placed in that the photoelectric sensor read step b1 of qualification bank note opposite side is determined to be read position and read described the first image information under width.
5. the method for a kind of automatic selection bank note different characteristic point judgement according to claim 4, is characterized in that: described step a specifically comprises the steps:
A1, set all true bank note particular location at identification mark information place and read width under four bank note input direction situations respectively;
A2, set all true bank note particular location at authenticity information place and read width under four bank note input direction situations respectively;
A3, radiation source irradiate described true bank note from described true bank note one side respectively, by being placed in the photoelectric sensor read step a1 of true bank note opposite side and particular location that a2 sets and reading the luminous flux information under width, this luminous flux information exchange is crossed and formed respectively the identification mark information of true bank note and authenticity information storage after analog to digital converter conversion in storer.
6. the method for a kind of automatic selection bank note different characteristic point judgement according to claim 5, is characterized in that: described identification mark information or the first image information are average waveform A, and described average waveform A obtains as follows:
A310, the P of photoelectric sensor from setting along the Y-direction of true bank note or plan qualification bank note
0position starts to read the luminous flux of true bank note or plan qualification bank note, and reading width is W
0;
A311, processor identify that to true bank note or plan bank note is from P
0position starts to W
0the view data of width range averages calculating, obtains a line view data, and described view data forms average waveform A along true bank note or plan qualification bank note directions X.
7. the method for a kind of automatic selection bank note different characteristic point judgement according to claim 6, it is characterized in that: described authenticity information or the second image information are average waveform B and/or C, or be the set of average waveform B and/or C, the set of described average waveform B or average waveform B obtains by following steps:
A320, the preliminary election P of photoelectric sensor from setting along the Y-direction of true bank note or plan qualification bank note
ymposition starts to read the luminous flux of true bank note or plan qualification bank note, and reading preselected width is W
ym,
A321, processor identify that to true bank note or plan bank note is from preliminary election P
ymposition starts to preliminary election W
ymthe view data of width range averages calculating, obtains a line view data, and described view data forms average waveform B along bank note horizontal direction
m, wherein
mfor identify described average waveform B quantity, be more than or equal to 1 natural number;
The set of described average waveform C or average waveform C obtains by following steps:
A330, the preliminary election P of photoelectric sensor from setting along the directions X of true bank note or plan qualification bank note
xnposition starts to read the luminous flux of true bank note or plan qualification bank note, and reading preselected width is W
xn,
A331, processor identify that to true bank note or plan bank note is from preliminary election P
xnposition starts to preliminary election W
xnthe view data of width range averages calculating, obtains a line view data, and described view data forms average waveform Cn along true bank note or plan qualification bank note Y-direction, wherein
nfor identify described average waveform C quantity, be more than or equal to 1 natural number;
The average waveform A of described true bank note and described average waveform B or its set, described average waveform C or its set, or corresponding with the common set forming of average waveform C with average waveform B.
8. the method for a kind of automatic selection bank note different characteristic point judgement according to claim 7, is characterized in that: described processor is digital processing unit DSP, microprocessor MPU or System on chip SoC.
9. the method for a kind of automatic selection bank note different characteristic point judgement described in claim 2 or 3, is characterized in that: described photoelectric sensor is several linearly aligned CIS contact-type image sensors, CMOS or ccd image sensor.
10. the method for a kind of automatic selection bank note different characteristic point judgement described in claim 2 or 3, is characterized in that: described radiation source is one or more groups infrared light, visible ray and/or the ultraviolet light that can carry out penetration or reflective irradiation.
The method of a kind of automatic selection bank note different characteristic point judgement in 11. claims 1 to 3 described in any one claim, is characterized in that: described bank note is banknote, stamp, duty stamp, stock, bond, treasury stock, commercial paper, acceptance bill or bank's time deposit certificate.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310128538.6A CN104050749B (en) | 2013-04-15 | 2013-04-15 | A kind of method automatically selecting valuable papers's different characteristic point judgement |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310128538.6A CN104050749B (en) | 2013-04-15 | 2013-04-15 | A kind of method automatically selecting valuable papers's different characteristic point judgement |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104050749A true CN104050749A (en) | 2014-09-17 |
| CN104050749B CN104050749B (en) | 2016-09-14 |
Family
ID=51503514
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310128538.6A Expired - Fee Related CN104050749B (en) | 2013-04-15 | 2013-04-15 | A kind of method automatically selecting valuable papers's different characteristic point judgement |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104050749B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106296972A (en) * | 2016-08-10 | 2017-01-04 | 恒银金融科技股份有限公司 | A kind of banknote identification module |
| CN106447897A (en) * | 2016-08-15 | 2017-02-22 | 深圳怡化电脑股份有限公司 | Method and apparatus for detecting magnetic characteristics of paper note |
| CN106504403A (en) * | 2016-11-17 | 2017-03-15 | 深圳怡化电脑股份有限公司 | A kind of method and device of paper money discrimination |
| CN106600810A (en) * | 2016-12-05 | 2017-04-26 | 深圳怡化电脑股份有限公司 | Currency identification method and currency identification device |
| CN108475452A (en) * | 2016-04-19 | 2018-08-31 | 日立欧姆龙金融***有限公司 | Paper verifies processing system, bill handling device, paper discriminating device and paper and verifies processing method |
| CN110599672A (en) * | 2019-08-30 | 2019-12-20 | 深圳怡化电脑股份有限公司 | Cash-sticking detection method and device for financial equipment, terminal equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6529269B1 (en) * | 1999-09-28 | 2003-03-04 | Nippon Conlux Co., Ltd. | Paper sheet identification method and device |
| CN101105873A (en) * | 2006-07-13 | 2008-01-16 | 阿鲁策株式会社 | Bank note treatment device |
| CN101405772A (en) * | 2006-03-16 | 2009-04-08 | 阿鲁策株式会社 | Bank note authenticating method and bank note authenticating device |
| JP2010117803A (en) * | 2008-11-12 | 2010-05-27 | Hitachi Omron Terminal Solutions Corp | Paper sheet processor and paper sheet processing program |
| CN102222384A (en) * | 2011-05-27 | 2011-10-19 | 尤新革 | Analysis method of multispectral image of note |
-
2013
- 2013-04-15 CN CN201310128538.6A patent/CN104050749B/en not_active Expired - Fee Related
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6529269B1 (en) * | 1999-09-28 | 2003-03-04 | Nippon Conlux Co., Ltd. | Paper sheet identification method and device |
| CN101405772A (en) * | 2006-03-16 | 2009-04-08 | 阿鲁策株式会社 | Bank note authenticating method and bank note authenticating device |
| CN101105873A (en) * | 2006-07-13 | 2008-01-16 | 阿鲁策株式会社 | Bank note treatment device |
| JP2010117803A (en) * | 2008-11-12 | 2010-05-27 | Hitachi Omron Terminal Solutions Corp | Paper sheet processor and paper sheet processing program |
| CN102222384A (en) * | 2011-05-27 | 2011-10-19 | 尤新革 | Analysis method of multispectral image of note |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108475452A (en) * | 2016-04-19 | 2018-08-31 | 日立欧姆龙金融***有限公司 | Paper verifies processing system, bill handling device, paper discriminating device and paper and verifies processing method |
| CN106296972A (en) * | 2016-08-10 | 2017-01-04 | 恒银金融科技股份有限公司 | A kind of banknote identification module |
| CN106447897A (en) * | 2016-08-15 | 2017-02-22 | 深圳怡化电脑股份有限公司 | Method and apparatus for detecting magnetic characteristics of paper note |
| CN106504403A (en) * | 2016-11-17 | 2017-03-15 | 深圳怡化电脑股份有限公司 | A kind of method and device of paper money discrimination |
| CN106600810A (en) * | 2016-12-05 | 2017-04-26 | 深圳怡化电脑股份有限公司 | Currency identification method and currency identification device |
| CN106600810B (en) * | 2016-12-05 | 2019-04-26 | 深圳怡化电脑股份有限公司 | A kind of recognition methods of bank note and device |
| CN110599672A (en) * | 2019-08-30 | 2019-12-20 | 深圳怡化电脑股份有限公司 | Cash-sticking detection method and device for financial equipment, terminal equipment and storage medium |
| CN110599672B (en) * | 2019-08-30 | 2021-10-08 | 深圳怡化电脑股份有限公司 | Cash-sticking detection method and device for financial equipment, terminal equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104050749B (en) | 2016-09-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105957238B (en) | A kind of paper currency management method and its system | |
| CN104050749A (en) | Method for making judgment by automatically selecting different feature points of banknotes | |
| CN100437625C (en) | Currency verification | |
| CN102542660B (en) | Bill anti-counterfeiting identification method based on bill watermark distribution characteristics | |
| CN101303779B (en) | Automatic teller machine and tracking method of paper currency | |
| CN100595799C (en) | Two-dimensional currency automatic recognition method and system | |
| CN105243731B (en) | A kind of multinational banknote image identifying device of spectrum self adaptation, system and method | |
| CN107610322B (en) | Banknote version identification method and device, electronic equipment and storage medium | |
| CN103021069B (en) | High-speed note image acquisition processing system and acquisition processing method thereof | |
| CN105184957A (en) | Paper currency discrimination method and system | |
| CN106934915B (en) | Bank note treatment device | |
| CN104992496A (en) | Paper money face identification method and apparatus | |
| CN107331031B (en) | A kind of recognition methods of note denomination, device, equipment and storage medium | |
| CN102324132A (en) | Bank-note detecting method by identifying ultraviolet-light image and system | |
| CN108320373A (en) | A kind of method and device of the detection of guiding against false of paper currency mark | |
| CN105788064B (en) | Bill identification method and device | |
| CN106683261A (en) | Method and device for authentic identification of banknote | |
| CN204990480U (en) | A smooth pattern differentiation system that becomes of luster for 2015 editions RMB | |
| US11842593B2 (en) | Systems and methods for detection of counterfeit documents | |
| Amirsab et al. | An automated recognition of fake or destroyed Indian currency notes | |
| CN202257774U (en) | Device and system for screening magnetic bank card | |
| CN103903329A (en) | Financial instrument authenticity identification device with front CIS | |
| CN109543554B (en) | Bill detection method, device, terminal and computer readable storage medium | |
| CN106296975A (en) | The recognition methods of a kind of dollar bill face amount and device | |
| WO2015172304A1 (en) | Method for making judgment by automatically selecting different feature points of bank notes |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20190412 Address after: 518055 Room 1803, Jinqi Zhigu Business Building, No. 4 Chongwen Garden, Taoyuan Street, Nanshan District, Shenzhen City, Guangdong Province Patentee after: Shenzhen city Baolige Micro Technology Co. Ltd. Address before: 518034 No. 406, 4 Wanke Golden Home Building, Lianhua Road, Futian District, Shenzhen City, Guangdong Province Patentee before: Many intelligent Science and Technology Ltd.s are won in Shenzhen |
|
| TR01 | Transfer of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20160914 Termination date: 20200415 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |