CN102306415A - Portable valuable file identification device - Google Patents

Portable valuable file identification device Download PDF

Info

Publication number
CN102306415A
CN102306415A CN201110218463A CN201110218463A CN102306415A CN 102306415 A CN102306415 A CN 102306415A CN 201110218463 A CN201110218463 A CN 201110218463A CN 201110218463 A CN201110218463 A CN 201110218463A CN 102306415 A CN102306415 A CN 102306415A
Authority
CN
China
Prior art keywords
image
value document
point
infrared
portable
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
Application number
CN201110218463A
Other languages
Chinese (zh)
Other versions
CN102306415B (en
Inventor
刘梦涛
向拓闻
黎明
徐朝阳
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
GRG Banking Equipment Co Ltd
Guangdian Yuntong Financial Electronic Co Ltd
Original Assignee
Guangdian Yuntong Financial Electronic Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Guangdian Yuntong Financial Electronic Co Ltd filed Critical Guangdian Yuntong Financial Electronic Co Ltd
Priority to CN 201110218463 priority Critical patent/CN102306415B/en
Publication of CN102306415A publication Critical patent/CN102306415A/en
Application granted granted Critical
Publication of CN102306415B publication Critical patent/CN102306415B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Landscapes

  • Image Processing (AREA)

Abstract

The invention discloses a portable valuable file identification device comprising portable equipment, wherein a camera head is arranged on the portable equipment; the camera head is additionally provided with an infrared filter; and the portable equipment is also provided with an acquisition module for acquiring original infrared images of valuable files, a preprocessing module for carrying out image smoothness, image restoration and image segmentation on the original infrared images, a projection correction module for carrying out projection correction on the valuable file images and an identification module for identifying the truth or the falsity of the valuable files. The portable valuable file identification device is simple in operation high in identification accuracy and can bring great convenience to users.

Description

Portable value document recognition device
Technical field
The present invention relates to the valuable document identification technology field, relate in particular to a kind of portable value document recognition device.
Background technology
At present, though various bank card is widely used, the circulation of cash still is main mode, and increasing.Some offenders utilize valuable documents such as high-tech technology forged banknote, and society has been caused serious harm.In order to prevent the circulation of imitated value document such as counterfeit money, the user to the precision of value document identification device with portable require increasingly high.
Yet value document identification equipment comparatively accurately is not only expensive, and not portable like large-scale machines such as cash inspecting machine, ATMs at present, big limitations the field of employment of these equipment; And some easy portable equipment like equipment such as e-Pointers, are difficult to accurately differentiate the true and false of value document again.
Summary of the invention
The embodiment of the invention proposes a kind of portable value document recognition device, and is simple to operate, bring great convenience to the user, and the identification precision is high.
The portable value document recognition device that the embodiment of the invention provides comprises portable set, and said portable set is provided with camera, and said camera adds infrared fileter;
On said portable set, also dispose:
Acquisition module is used to control said camera value document to be identified is taken, and obtains original infrared image; Said original infrared image includes value document zone and background area, and there is aberration in two zones;
Pre-processing module is used for that said original infrared image is carried out image smoothing and handles with recovering, and searches the marginal point of value document the original infrared image after handling, and is partitioned into the value document image;
Projection correction's module, application image projective transformation technology is carried out projection correction to said value document image, and the size of value document image and the template image that prestores after the correction is consistent;
Identification module judges whether the infrared signature data of the value document image after proofreading and correct meet the infrared characteristic standard that presets; If meet, judge that then value document is a legitimate files, otherwise judge that value document is illegal file.
Further, said portable set also disposes communication module;
Said communication module receives the original infrared image that remote equipment sends through wireless communication networks, and sends said original infrared image to said acquisition module;
Said communication module also is sent to remote equipment through wireless communication networks with the value document recognition result.
Wherein, said pre-processing module comprises:
The image noise reduction unit adopts Gauss's smoothing technique that said original infrared image is carried out picture smooth treatment;
Image restoration unit, the image-recovery technique of employing PDE is carried out image to the original infrared image after the smoothing processing and is recovered to handle;
The marginal point of value document is searched in the framing unit the original infrared image after recover handling, define the value document zone;
The image segmentation unit is made as 0 with the pixel value of the background area outside the value document zone in the said original infrared image, is partitioned into the value document image.
The portable value document recognition device that the embodiment of the invention provides; The camera collection value document infrared image of infrared fileter has been installed in employing additional; Again to the value document infrared image carry out denoising, recover, cut apart, processing such as correction; Reduce the influence of factors such as noise, shake, deformation as far as possible, improve the identification precision of value document, reliability is high.This portable value document recognition device is simple to operate, brings great convenience to the user, helps to take precautions against the circulation of illegal value document.
Description of drawings
Fig. 1 is the structural representation of the portable value document recognition device that provides of the embodiment of the invention one;
Fig. 2 is the structural representation of the pre-processing module that provides of the embodiment of the invention two;
Fig. 3 is the structural representation of the framing unit that provides of the embodiment of the invention three;
Fig. 4 is the structural representation of the image correction module that provides of the embodiment of the invention four;
Fig. 5 is the synoptic diagram of the collection value document infrared image that provides of the embodiment of the invention five;
Fig. 6 is the synoptic diagram of the value document image coordinate system that provides of the embodiment of the invention six.
Embodiment
To combine the accompanying drawing in the embodiment of the invention below, the technical scheme in the embodiment of the invention is carried out clear, intactly description, obviously, described embodiment only is the present invention's part embodiment, rather than whole embodiment.Based on the embodiment among the present invention, those of ordinary skills are not making the every other embodiment that is obtained under the creative work prerequisite, all belong to the scope of the present invention's protection.
Referring to Fig. 1, be the structural representation of the portable value document recognition device that provides of the embodiment of the invention one.
The portable value document recognition device of present embodiment comprises portable set, and this portable set is provided with camera, and said camera adds infrared fileter.During practical implementation, said portable set comprises simple and easy mobile devices such as mobile phone, digital camera.
As shown in Figure 1, on portable set, also dispose:
Acquisition module 1 is used to control said camera value document to be identified is taken, and obtains original infrared image; Said original infrared image includes value document zone and background area, and there is aberration in two zones;
Pre-processing module 2 is used for that said original infrared image is carried out image smoothing and handles with recovering, and searches the marginal point of value document the original infrared image after handling, and is partitioned into the value document image;
Projection correction's module 3, application image projective transformation technology is carried out projection correction to said value document image, and the size of value document image and the template image that prestores after the correction is consistent;
Identification module 4 judges whether the infrared signature data of the value document image after proofreading and correct meet the infrared characteristic standard that presets; If meet, judge that then value document is a legitimate files, otherwise judge that value document is illegal file.
In a preferred implementation, extract the infrared signature data in the infrared signature zone of the value document image of identification module 4 after proofread and correct; Said infrared signature data comprise at least one in the variance of gray-scale value of average and infrared image of gray-scale value of gradient eigenwert, infrared image of the gray-scale value of infrared image.Identification module 4 compares said infrared signature data and the template image characteristic that prestores; If comparative result meets the infrared characteristic standard that presets, judge that then value document is a legitimate files, otherwise judge that value document is illegal file.
In another embodiment of the present invention, portable set can also receive original infrared image through wireless communication networks except the original infrared image that obtains value document through camera.Concrete, the portable set of present embodiment also disposes communication module; This communication module receives the original infrared image that remote equipment sends through wireless communication networks, and sends said original infrared image to acquisition module 1; And this communication module also is sent to remote equipment through wireless communication networks with the value document recognition result.
Referring to Fig. 2, be the structural representation of the pre-processing module that provides of the embodiment of the invention two.
Concrete, pre-processing module 2 comprises:
Image noise reduction unit 21 adopts Gauss's smoothing technique that said original infrared image is carried out picture smooth treatment;
Image restoration unit 22, the image-recovery technique of employing PDE is carried out image to the original infrared image after the smoothing processing and is recovered to handle;
The marginal point of value document is searched in framing unit 23 the original infrared image after recover handling, define the value document zone;
Image segmentation unit 24 is made as 0 with the pixel value of the background area outside the value document zone in the said original infrared image, is partitioned into the value document image.
Referring to Fig. 3, be the structural representation of the framing unit that provides of the embodiment of the invention three.
Concrete, framing unit 23 comprises:
Coordinate system construction unit 231 is an initial point with the summit, the lower left corner of said original infrared image, sets up the x axle along the Width of said original infrared image, sets up the y axle along the length direction of said original infrared image; The width of said original infrared image is W, and length is L;
Marginal point is searched unit 232, on the x=W/2 straight line, checks and accepts value document edge of image point, obtains the coboundary point
Figure BDA0000080290960000051
With the lower limb point
Figure BDA0000080290960000052
On the y=L/2 straight line, check and accept value document edge of image point, obtain the left hand edge point With the right hand edge point
Figure BDA0000080290960000054
Respectively with
Figure BDA0000080290960000055
With Being reference point, is step-size in search with Δ w on the x direction of principal axis, on the y direction of principal axis, is step-size in search with Δ L, searches the next marginal point of value document image; Marginal point of every acquisition is that reference point repeats search procedure with this marginal point just, in the hunting zone, can not find till the marginal point; All the coboundary points, lower limb point, left hand edge point and the right hand edge point that obtain in the whole search procedure are formed coboundary point sequence P respectively U, lower limb point sequence P D, left hand edge point sequence P LWith right hand edge point sequence P R
Straight line construction unit 233 is used for respectively to P U, P D, P LAnd P RCarry out least square linear fit, obtain coboundary straight-line equation L U, lower limb straight-line equation L D, left hand edge straight-line equation L LWith right hand edge straight-line equation L R
Summit computing unit 234 is used for according to L U, L D, L LAnd L RCalculate the intersection point between the adjacent straight line, obtain the coordinate on four summits of value document image, four quadrilateral areas that the summit surrounded are the value document zone.
Referring to Fig. 4, be the structural representation of the image correction module that provides of the embodiment of the invention four.
Concrete, projection correction's module 3 comprises:
Modeling unit 31 is used for the mapping relations according to four apex coordinates of four apex coordinates of said value document image and template image, obtains mapping function;
Correcting unit 32 is used for according to said mapping function said value document image being carried out projection correction, and value document image and the size of template image after the correction are consistent.
In a preferred implementation, the value document image function before proofreading and correct is g (x ', y '), and the template image function is that (x, y), both mapping functions are Bilinear Equations: x '=c to f 1X+c 2Y+c 3Xy+c 4, y '=c 5X+c 6Y+c 7Xy+c 8
In four said Bilinear Equations of apex coordinate substitution of modeling unit 31 with four apex coordinates of template image and said value document image, draw parameter c 1, c 2, c 3, c 4, c 5, c 6, c 7And c 8Value, obtain the mapping function of said value document image and template image.Correcting unit 32 is according to said mapping function, and the pixel value of each point in the said value document image is mapped on the said template image value document image after obtaining to proofread and correct.
In addition, as shown in Figure 1, also dispose on the portable set:
Storage unit 5 is used to store original infrared picture data, template image data and infrared characteristic normal data;
Output unit 6 is used to export the value document recognition result;
Interactive operation unit 7 is used to import the kind, denomination of current value document to be measured, towards data;
Control module 8 is used to control co-ordination between above-mentioned each module.
Below in conjunction with Fig. 5 and Fig. 6, the flow process of the portable value document recognition device identification value document that the embodiment of the invention is provided is described in detail.
One, IMAQ
The acquisition module 1 of portable value document recognition device; Can be through installing additional of portable set itself the original infrared image of camera collection value document of infrared fileter, can also receive the original infrared image that remote equipment sends through wireless communication networks.
Referring to Fig. 5, be the synoptic diagram of the collection value document infrared image that provides of the embodiment of the invention five.The original infrared image of value document take to be obtained by camera, and during practical implementation, camera can obtain the two dimensional image of value document from all angles, i.e. 0＜α≤90 °; Wherein, shooting angle α is the angle of camera axis and value document acquisition plane.
If shooting angle is near level, the anamorphose of then obtaining is bigger, and recognition effect is relatively poor; Otherwise if shooting angle is approaching vertical, the anamorphose of then obtaining is less, and recognition effect is better.As shown in Figure 2, when camera was in A, C position, the infrared image that is obtained was an oblique view (a) and (c); When camera was in the B position, the infrared image that is obtained was front elevation (b).
Preferably, shooting angle α＞60 °.And in order to improve the discriminating precision, background is single as far as possible during images acquired, and with value document certain aberration is arranged.
Two, image pre-service
When utilizing the original infrared image of camera collection value document; Because the restriction of complex environment and the performance of picture pick-up device own; Can cause decrease in image quality; Therefore before value document was carried out truth identification, the pre-processing module 2 of portable value document recognition device was carried out the image rectification earlier, returns to the original image of degenerating without disturbing.Specific as follows:
(1), image denoising
Image noise reduction unit 21 adopts Gauss's smoothing technique that original infrared image is carried out picture smooth treatment, suppresses because the picture noise that the performance limitations of collecting device own is caused.Adopt Gauss's smoothing technique to handle image, not only can noise reduction, and can reduce fuzzy in the smoothing processing, improve the accuracy that value document is differentiated.
(2), image recovers
Value document possibly obtain under various environment, and indoor etc. as under sunny open air, the daylight lamp owing to gather the influence of shaking in environment and the gatherer process, makes the infrared image that collects possibly have certain degradation phenomena.Image restoration unit 22 is utilized the image-recovery technique of PDE, the original infrared image after the smoothing processing is carried out image recover to handle, can be with the optimal estimation value of the image restoration that is collected to original image.
(3), framing
Because when utilizing camera that value document is taken, can background environment together be taken in the original infrared image, promptly original infrared image includes value document zone and background area, therefore need from original infrared image, cut out the value document image.The location dividing processing has improved the accuracy of infrared image projection correction effectively.During practical implementation, under a simple relatively background environment, gather the value document image, help realizing accurate location the value document image.
Concrete, the treatment scheme that the 23 pairs of value document images in framing unit position is following:
Referring to Fig. 6, be the synoptic diagram of the value document image coordinate system that provides of the embodiment of the invention six.
Quadrilateral ABCO shown in Figure 6 is original infrared image, and its length is L, and width is W; Quadrilateral MNPQ is the value document image in the original infrared image.Coordinate system construction unit 231 is an initial point with summit, the lower left corner O of original infrared image, sets up the x axle along the Width (being straight line OC direction) of original infrared image, sets up the y axle along the length direction (being straight line OA direction) of original infrared image.
Marginal point is searched unit 232 and from original infrared image, is searched value document edge of image point, comprising:
On the x=W/2 straight line; From up to down search for the coboundary point of value document image; Obtain coboundary point on the x=W/2 straight line; The lower limb point of bottom-up search value document image obtains the lower limb point P 0 D . Order P 0 U = ( x 0 U , y 0 U ) , P 0 D = ( x 0 D , y 0 D ) .
On the y=L/2 straight line; Search for the left hand edge point of value document image from left to right; Obtain left hand edge point on the y=L/2 straight line; The right hand edge point of right-to-left search value document image obtains the right hand edge point P 0 R . Order P 0 L = ( x 0 L , y 0 L ) , P 0 R = ( x 0 R , y 0 R ) .
Above marginal point
Figure BDA0000080290960000089
is a reference point, on straight line, searches the coboundary point of value document image; Wherein, The hunting zone of y is the step-size in search on the x direction of principal axis for
Figure BDA00000802909600000811
Δ w, and Δ L is the step-size in search on the y direction of principal axis.As shown in Figure 6, on
Figure BDA00000802909600000812
straight line, search coboundary point
Figure BDA00000802909600000813
x = x 0 U - Δ w Search the coboundary point on the straight line P - 1 U . Order P 1 U = ( x 1 U , y 1 U ) , P - 1 U = ( x - 1 U , y - 1 U ) .
Above again marginal point is a reference point, on
Figure BDA00000802909600000819
straight line, searches the coboundary point of value document image; Above marginal point
Figure BDA00000802909600000820
is a reference point, on
Figure BDA00000802909600000821
straight line, searches the coboundary point of value document image; Wherein, The hunting zone of y is that
Figure BDA00000802909600000822
is as shown in Figure 6; On
Figure BDA00000802909600000823
straight line, search coboundary point
Figure BDA00000802909600000824
, search is less than the coboundary point on
Figure BDA00000802909600000825
straight line.Afterwards, above again marginal point repeats search procedure for reference point.The rest may be inferred, and marginal point of every acquisition is that reference point repeats search procedure with this marginal point just, in the hunting zone, can not find till the marginal point.
All coboundary points that obtain in the whole search procedure are formed coboundary point sequence P U:
P U = ( P - M U U , P - M U + 1 U . . . , P - 1 U , P 0 U , P 1 U , . . . , P N U - 1 U , P N U U ) .
In like manner; According to the above-mentioned method of searching marginal point;
Figure BDA00000802909600000828
is reference point with marginal point; On the x direction of principal axis, being step-size in search with Δ w, is step-size in search with Δ L on the y direction of principal axis, searches the next marginal point of value document image; Marginal point of every acquisition is that reference point repeats search procedure with this marginal point just, in the hunting zone, can not find till the marginal point.
All lower limb points that obtain in the whole search procedure are formed lower limb point sequence P D:
P D = ( P - M D D , P - M D + 1 D , . . . , P - 1 D , P 0 D , P 1 D , . . . , P N D - 1 D , P N D D ) .
The marginal point sequence P that straight line construction unit 233 usefulness are above-mentioned UAnd P DCarry out least square linear fit, obtain coboundary straight-line equation L UWith lower limb straight-line equation L D
Marginal point is searched unit 232 according to above-mentioned method; With left hand edge point
Figure BDA0000080290960000091
and right hand edge point
Figure BDA0000080290960000092
is reference point; On the x direction of principal axis, be step-size in search with Δ w; On the y direction of principal axis, be step-size in search, search value document edge of image point with Δ L.All the left hand edge points, the right hand edge point that obtain in the whole search procedure are formed left hand edge point sequence P respectively LWith right hand edge point sequence P RStraight line construction unit 233 is again to P LAnd P RCarry out least square linear fit, obtain left hand edge straight-line equation L LWith right hand edge straight-line equation L R
Summit computing unit 234 is according to L U, L D, L LAnd L RCalculate the intersection point between the adjacent straight line, can obtain the coordinate on four summits of value document image, four quadrilateral areas that the summit surrounded are the value document zone.As shown in Figure 6, behind the coordinate on the M of acquisition value document image, N, P, four summits of Q, just can from photo OABC, orient value document image M NPQ.
(4), image segmentation
Image segmentation unit 24 remains unchanged the pixel value in the value document image-region, and the pixel value of the background area outside the value document zone is made as 0, is partitioned into the value document zone.
Three, select template characteristic
The kind of the value document that interactive operation unit 7 input is to be identified, denomination, towards; Projection correction unit 3 and identification module 4 are called in the characteristic attribute (as: image, size, characteristic area etc.) of corresponding value document template image from storage unit 5, for the real and fake discrimination of value document is prepared.
Four, projection correction
When utilizing camera that value document is taken; Reason owing to shooting angle; Might obtain the two-dimentional oblique view of value document; Therefore template image in the size of the original infrared image of the value document that is collected and the storage unit 5 is difference to some extent, needs application image projective transformation technology that the value document infrared image is proofreaied and correct, and makes consistent size in value document infrared image and the value document template characteristic attribute.
Value document image before proofreading and correct is distorted image g (x ', y '), the template image of standard value document be f (x, y), mapping relations between the two be s (x, y) and t (x, y).Modeling unit 31 utilize Bilinear Equations set up distorted image g (x ', y ') and template image f (x, y) in the mapping relations of each coordinate:
x′＝s(x，y)＝c 1x+c 2y+c 3xy+c 4
y′＝t(x，y)＝c 5x+c 6y+c 7xy+c 8
In 4 pairs of above-mentioned Bilinear Equations of reference point substitution, can calculate parameter c 1, c 2, c 3, c 4, c 5, c 6, c 7And c 8Value, thereby confirm mapping relations s (x, y) and t (x, y).By the inverse process of above-mentioned model, can by distorted image g (x ', y ') return to template image f (x, y).
During practical implementation, according to the Currency Type of value document to be measured, face amount, transfer the characteristic attribute of value document template image from storage unit 5, obtain the size of original image towards, modeling unit 31; Supposing that width is K, highly is H, and four summits of original image are: (0; 0), (0, H), (K, 0), (K; H), four summits of distorted image can obtain in above-mentioned image pretreatment stage, are assumed to be: (x 1, y 1), (x 2, y 2), (x 3, y 3), (x 4, y 4).Above 4 pairs of data accordingly in the Bilinear Equations, can be calculated parameter c by 8 groups of data 1, c 2, c 3, c 4, c 5, c 6, c 7And c 8Value, as follows:
c 1 = ( x 3 - x 1 ) / K c 2 = ( y 3 - y 1 ) / H c 3 = ( x 4 - x 3 - x 2 + x 1 ) / K / H c 4 = x 1 c 5 = ( y 3 - y 1 ) / K c 6 = ( y 2 - y 1 ) / H c 7 = ( y 4 - y 3 - y 2 + y 1 ) / K / H c 8 = y 1
Point on the template image f (x, y), the point on the corresponding deformation pattern g (s (x, y), t (x, y)), utilize bilinear interpolation obtain point (s (and x, y), t (x, y)) pixel value; If point (s (x, y), t (x, y)) is not in image g, and then (x, y) pixel puts 0 with the point on the template image f.Be consistent through the image value document of projection correction and the size of template image.
The present invention adopts Bilinear Equations to calculate mapping relations; Only need to find four pairs of corresponding point to set up Bilinear Equations between value document image and the template image before correction as reference point; Solve 8 parameters of Bilinear Equations; Mapping relations in value document image before can confirming to proofread and correct and the template image between each point coordinate are calculated simple; Come that according to above-mentioned mapping relations the value document image is carried out projection correction and handle, farthest restored image is avoided image fault.Need to prove that reference point can be the summit of image, also can be other unique points.
Compared with prior art; Portable value document recognition device provided by the invention has projection correction's module; Using the projective transformation technology is consistent the original infrared image of the value document that collects and the size of the value document template image that prestores; On the one hand, make the present invention, can collect the original infrared image that meets requirement of the present invention as long as utilize the camera that has installed infrared fileter additional; Make the present invention can be applicable to have in the equipment such as mobile phone, camera of infrared photography function, be easy to carry and cost low; On the other hand, when gathering the original infrared image of value document, can take,, also can not influence the accuracy of identification of value document even collect the oblique view of value document from all angles, simple to operate; Again on the one hand, even the collecting device of prior art is stable again, the infrared image that collects is rectified again, also certain error can occur, and the present invention proofreaies and correct the image that collects through projection correction's module, has effectively improved the identification precision of value document.
Five, truth identification
Truth identification is the infrared signature zone with the value document image after proofreading and correct, and makes comparisons with the infrared signature zone of value document template image in the storage unit, thereby identifies the true and false of value document.Wherein, The infrared signature zone is meant: in the value document manufacture process; In order to increase the anti-forgery security of value document; And use infrared absorption printing ink or similarly infrared light activated specialty inks is printed with value document, on value document, formed the relatively more responsive zone of infrared radiation (with respect to other zones).
Infrared signature data in the infrared signature zone of value document image comprise the variance of gray-scale value of average and infrared image of gray-scale value of gradient eigenwert, the infrared image of the gray-scale value of infrared image.The flow process that 4 pairs of value documents of identification module carry out truth identification is following:
(1), calculates the infrared signature region gradient
Calculate interior point (x, Grad G y) of infrared signature zone Ω of value document image to be measured Ω(x, y), statistics meets G Ω(x, y)＞TH gThe number N of the point of condition gG 0(x is the Grad of the point in the template image corresponding region y), meets G 0(x, y)＞TH gThe number of the point of condition is N 0If N g/ N 0∈ [0.95,1.05] judges that then value document to be measured meets the gradient standard; Otherwise judge that value document to be measured is illegal file.
Wherein, TH gBe the threshold value that is used for gradient analysis that presets, 1.0＜TH g＜25.0.
(2), calculate infrared signature zone average
Calculate the interior average M of infrared signature zone Ω of value document image to be measured ΩM 0Be average in the template image corresponding region; If M Ω/ M 0∈ [0.90,1.10] judges that then value document to be measured meets the average standard; Otherwise judge that value document to be measured is illegal file.
(3), calculate infrared signature zone variance
Calculate the infrared signature zone Ω internal variance V of value document image to be measured ΩV 0Be template image corresponding region internal variance; If V Ω/ V 0∈ [0.80,1.25] judges that then value document to be measured meets variance criterion; Otherwise judge that value document to be measured is illegal file.
If the infrared signature data in the Ω of the infrared signature of value document image to be measured zone all meet gradient, average and variance criterion, judge that then said value document to be measured is a legitimate files, otherwise judge that said value document to be measured is illegal file.At last, output module 6 output value document recognition results.
The portable value document recognition device that the embodiment of the invention provides; The camera collection value document infrared image of infrared fileter has been installed in employing additional; Again to the value document infrared image carry out denoising, recover, cut apart, processing such as correction; Reduce the influence of factors such as noise, shake, deformation as far as possible, improve the identification precision of value document, reliability is high.This portable value document recognition device is simple to operate, brings great convenience to the user, helps to take precautions against the circulation of illegal value document.
The above is a preferred implementation of the present invention; Should be pointed out that for those skilled in the art, under the prerequisite that does not break away from the principle of the invention; Can also make some improvement and retouching, these improvement and retouching also are regarded as protection scope of the present invention.

Claims (8)

1. a portable value document recognition device is characterized in that, comprise portable set, said portable set is provided with camera, and said camera adds infrared fileter;
On said portable set, also dispose:
Acquisition module is used to control said camera value document to be identified is taken, and obtains original infrared image; Said original infrared image includes value document zone and background area, and there is aberration in two zones;
Pre-processing module is used for that said original infrared image is carried out image smoothing and handles with recovering, and searches the marginal point of value document the original infrared image after handling, and is partitioned into the value document image;
Projection correction's module, application image projective transformation technology is carried out projection correction to said value document image, and the size of value document image and the template image that prestores after the correction is consistent;
Identification module judges whether the infrared signature data of the value document image after proofreading and correct meet the infrared characteristic standard that presets; If meet, judge that then value document is a legitimate files, otherwise judge that value document is illegal file.
2. portable value document recognition device as claimed in claim 1 is characterized in that said portable set also disposes communication module;
Said communication module receives the original infrared image that remote equipment sends through wireless communication networks, and sends said original infrared image to said acquisition module;
Said communication module also is sent to remote equipment through wireless communication networks with the value document recognition result.
3. according to claim 1 or claim 2 portable value document recognition device is characterized in that said pre-processing module comprises:
The image noise reduction unit adopts Gauss's smoothing technique that said original infrared image is carried out picture smooth treatment;
Image restoration unit, the image-recovery technique of employing PDE is carried out image to the original infrared image after the smoothing processing and is recovered to handle;
The marginal point of value document is searched in the framing unit the original infrared image after recover handling, define the value document zone;
The image segmentation unit is made as 0 with the pixel value of the background area outside the value document zone in the said original infrared image, is partitioned into the value document image.
4. portable value document recognition device as claimed in claim 3 is characterized in that, said framing unit comprises:
The coordinate system construction unit is an initial point with the summit, the lower left corner of said original infrared image, sets up the x axle along the Width of said original infrared image, sets up the y axle along the length direction of said original infrared image; The width of said original infrared image is W, and length is L;
Marginal point is searched the unit, on the x=W/2 straight line, checks and accepts value document edge of image point, obtains the coboundary point
Figure FDA0000080290950000021
With the lower limb point
Figure FDA0000080290950000022
On the y=L/2 straight line, check and accept value document edge of image point, obtain the left hand edge point
Figure FDA0000080290950000023
With the right hand edge point
Figure FDA0000080290950000024
Respectively with
Figure FDA0000080290950000025
With
Figure FDA0000080290950000026
Being reference point, is step-size in search with Δ w on the x direction of principal axis, on the y direction of principal axis, is step-size in search with Δ L, searches the next marginal point of value document image; Marginal point of every acquisition is that reference point repeats search procedure with this marginal point just, in the hunting zone, can not find till the marginal point; All the coboundary points, lower limb point, left hand edge point and the right hand edge point that obtain in the whole search procedure are formed coboundary point sequence P respectively U, lower limb point sequence P D, left hand edge point sequence P LWith right hand edge point sequence P R
The straight line construction unit is used for respectively to P U, P D, P LAnd P RCarry out least square linear fit, obtain coboundary straight-line equation L U, lower limb straight-line equation L D, left hand edge straight-line equation L LWith right hand edge straight-line equation L R
The summit computing unit is used for according to L U, L D, L LAnd L RCalculate the intersection point between the adjacent straight line, obtain the coordinate on four summits of value document image, four quadrilateral areas that the summit surrounded are the value document zone.
5. portable value document recognition device as claimed in claim 4 is characterized in that, said projection correction module comprises:
Modeling unit is used for the mapping relations according to four apex coordinates of four apex coordinates of said value document image and template image, obtains mapping function;
Correcting unit is used for according to said mapping function said value document image being carried out projection correction, and value document image and the size of template image after the correction are consistent.
6. portable value document recognition device as claimed in claim 5 is characterized in that, the value document image function before proofreading and correct is g (x ', y '), and the template image function is that (x, y), both mapping functions are Bilinear Equations: x '=c to f 1X+c 2Y+c 3Xy+c 4, y '=c 5X+c 6Y+c 7Xy+c 8
In four said Bilinear Equations of apex coordinate substitution of said modeling unit with four apex coordinates of template image and said value document image, draw parameter c 1, c 2, c 3, c 4, c 5, c 6, c 7And c 8Value, obtain the mapping function of said value document image and template image;
Said correcting unit is according to said mapping function, and the pixel value of each point in the said value document image is mapped on the said template image value document image after obtaining to proofread and correct.
7. portable value document recognition device as claimed in claim 6 is characterized in that, extracts the infrared signature data in the infrared signature zone of the value document image of said identification module after proofread and correct; Said infrared signature data comprise at least one in the variance of gray-scale value of average and infrared image of gray-scale value of gradient eigenwert, infrared image of the gray-scale value of infrared image;
Said identification module compares said infrared signature data and the template image characteristic that prestores; If comparative result meets the infrared characteristic standard that presets, judge that then value document is a legitimate files, otherwise judge that value document is illegal file.
8. portable value document recognition device as claimed in claim 7 is characterized in that, also disposes on the said portable set:
Storage unit is used to store original infrared picture data, template image data and infrared characteristic normal data;
Output unit is used to export the value document recognition result;
The interactive operation unit is used to import the kind, denomination of current value document to be measured, towards data.
CN 201110218463 2011-08-01 2011-08-01 Portable valuable file identification device Active CN102306415B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN 201110218463 CN102306415B (en) 2011-08-01 2011-08-01 Portable valuable file identification device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN 201110218463 CN102306415B (en) 2011-08-01 2011-08-01 Portable valuable file identification device

Publications (2)

Publication Number Publication Date
CN102306415A true CN102306415A (en) 2012-01-04
CN102306415B CN102306415B (en) 2013-06-26

Family

ID=45380272

Family Applications (1)

Application Number Title Priority Date Filing Date
CN 201110218463 Active CN102306415B (en) 2011-08-01 2011-08-01 Portable valuable file identification device

Country Status (1)

Country Link
CN (1) CN102306415B (en)

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102592351A (en) * 2012-02-28 2012-07-18 广州广电运通金融电子股份有限公司 Authentication device and method of papery medium
WO2013040933A1 (en) * 2011-09-19 2013-03-28 广州广电运通金融电子股份有限公司 Identification method for valuable file and identification device thereof
CN104464078A (en) * 2014-12-08 2015-03-25 深圳怡化电脑股份有限公司 Method and system for identifying damaged bank note through optical variability printing ink
CN105632015A (en) * 2015-12-24 2016-06-01 中钞实业有限公司 Bank bill fingerprint feature anti-counterfeiting identification method based on cloud platform
CN106845508A (en) * 2015-12-07 2017-06-13 腾讯科技（深圳）有限公司 The methods, devices and systems of release in a kind of detection image
CN106952393A (en) * 2017-03-28 2017-07-14 深圳怡化电脑股份有限公司 Paper Currency Identification and device, electronic equipment and storage medium
CN108292457A (en) * 2015-11-26 2018-07-17 凸版印刷株式会社 Identification device, recognition methods, recognizer and the computer-readable medium comprising recognizer
CN109658584A (en) * 2018-12-14 2019-04-19 泰康保险集团股份有限公司 A kind of bill bank slip recognition method and device

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0722597B1 (en) * 1993-10-06 1999-06-02 Minnesota Mining And Manufacturing Company Security reader for automatic detection of tampering and alteration
CN2512040Y (en) * 2001-06-28 2002-09-18 上海印钞厂 Multifunctional quality inspection device banknotes
JP2005196627A (en) * 2004-01-09 2005-07-21 Olympus Corp Camera with false ticket detecting function and false ticket detecting method using camera
CN2804968Y (en) * 2005-07-05 2006-08-09 张保华 Device for bank-notes identification
CN101366060A (en) * 2005-12-16 2009-02-11 Ncr公司 Media validation
CN101718528A (en) * 2009-12-10 2010-06-02 北京科技大学 Digital image based rapid solving method of circle parameters

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0722597B1 (en) * 1993-10-06 1999-06-02 Minnesota Mining And Manufacturing Company Security reader for automatic detection of tampering and alteration
CN2512040Y (en) * 2001-06-28 2002-09-18 上海印钞厂 Multifunctional quality inspection device banknotes
JP2005196627A (en) * 2004-01-09 2005-07-21 Olympus Corp Camera with false ticket detecting function and false ticket detecting method using camera
CN2804968Y (en) * 2005-07-05 2006-08-09 张保华 Device for bank-notes identification
CN101366060A (en) * 2005-12-16 2009-02-11 Ncr公司 Media validation
CN101718528A (en) * 2009-12-10 2010-06-02 北京科技大学 Digital image based rapid solving method of circle parameters

Cited By (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2013040933A1 (en) * 2011-09-19 2013-03-28 广州广电运通金融电子股份有限公司 Identification method for valuable file and identification device thereof
US9014459B2 (en) 2011-09-19 2015-04-21 Grg Banking Equipment Co., Ltd. Identification method for valuable file and identification device thereof
CN102592351A (en) * 2012-02-28 2012-07-18 广州广电运通金融电子股份有限公司 Authentication device and method of papery medium
CN102592351B (en) * 2012-02-28 2013-11-27 广州广电运通金融电子股份有限公司 Authentication method of papery medium
CN104464078A (en) * 2014-12-08 2015-03-25 深圳怡化电脑股份有限公司 Method and system for identifying damaged bank note through optical variability printing ink
CN108292457A (en) * 2015-11-26 2018-07-17 凸版印刷株式会社 Identification device, recognition methods, recognizer and the computer-readable medium comprising recognizer
CN106845508A (en) * 2015-12-07 2017-06-13 腾讯科技（深圳）有限公司 The methods, devices and systems of release in a kind of detection image
CN106845508B (en) * 2015-12-07 2019-05-17 腾讯科技（深圳）有限公司 The methods, devices and systems of release in a kind of detection image
US10410053B2 (en) 2015-12-07 2019-09-10 Tencent Technology (Shenzhen) Company Limited Method, apparatus, system, and storage medium for detecting information card in image
CN105632015A (en) * 2015-12-24 2016-06-01 中钞实业有限公司 Bank bill fingerprint feature anti-counterfeiting identification method based on cloud platform
CN105632015B (en) * 2015-12-24 2019-03-01 中钞实业有限公司 A kind of bank money fingerprint characteristic anti false authentication method based on cloud platform
CN106952393A (en) * 2017-03-28 2017-07-14 深圳怡化电脑股份有限公司 Paper Currency Identification and device, electronic equipment and storage medium
CN109658584A (en) * 2018-12-14 2019-04-19 泰康保险集团股份有限公司 A kind of bill bank slip recognition method and device

Also Published As

Publication number Publication date
CN102306415B (en) 2013-06-26

Similar Documents

Publication Publication Date Title
CN102306415B (en) Portable valuable file identification device
US9014459B2 (en) Identification method for valuable file and identification device thereof
CN101169827B (en) Method and device for tracking characteristic point of image
CN109684987B (en) Identity verification system and method based on certificate
KR101515256B1 (en) Document verification using dynamic document identification framework
CN106529414A (en) Method for realizing result authentication through image comparison
CN104464079B (en) Multiple Currencies face amount recognition methods based on template characteristic point and topological structure thereof
CN111680588A (en) Human face gate living body detection method based on visible light and infrared light
Mushtaq et al. Digital image forgeries and passive image authentication techniques: a survey
CN108229427A (en) A kind of identity-based certificate and the identity security verification method and system of recognition of face
CN103679147A (en) Method and device for identifying model of mobile phone
CN104112115A (en) Three-dimensional face detection and identification technology
CN103679636A (en) Rapid image splicing method based on point and line features
CN100498820C (en) Automatic recognising method and automatic recognising recording system for number of paper money
CN103793642B (en) Mobile internet palm print identity authentication method
CN104021419B (en) A kind of method and apparatus for counting paper number
CN103577818A (en) Method and device for recognizing image characters
CN108734235A (en) A kind of personal identification method and system for electronic prescription
CN103996033B (en) Human identification method based on tracking of human face five-sense-organ coordinates
WO2016123903A1 (en) Folded bill identification method and device
CN103810488B (en) Image characteristic extracting method, image search method, server, terminal and system
CN107122775A (en) A kind of Android mobile phone identity card character identifying method of feature based matching
CN110378351A (en) Seal discrimination method and device
CN105184225A (en) Multinational paper money image identification method and apparatus
CN106991448A (en) A kind of portrait compares processing method

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