Available online at www.sciencedirect.com
ScienceDirect Physics Procedia 73 (2015) 308 – 312
4th h International Confereence Photon nics and Info formation Optics, O PhIO O 2015, 28-330 January 2015 2
Sp pecificityy of corrrelation pattern p rrecognitiion meth hods appplication n in securitty hologrrams ideentity co ontrol app paratus E.Yu. E Zlokkazova *, R.S. R Stariko ova, S.B. O Odinokovb, I.K. Tsygaanovb, V.E . Talalaevb, V.V. Kooluchkinb a
National N Researcch Nuclear University MEPhI (Mo oscow Engineerinng Physics Institute), Kashirskoye shosse 31, Moscoow, 115409, Russsia b Bauman Moscoow State Technica al University, ul. Baumanskaya 2--ya, 5/1, Moscow, 105005, Russia
Absttract Automatic insspection of secuurity hologram (SH) identity iis highly deman nded issue due high distributioon of SH world dwide to prrotect documennts such as passsports, driving licenses, l banknnotes etc. Whilee most of the known approachhes use inspectiion of SH design d features nnone of these approaches a inspect the featuress of its surface relief r that is a direct contributioon to original master m matriix used for thesse holograms production. In our o previous woorks we represeented the devicee that was deveeloped to provid de SH identtification by proocessing of cohherent responsess of its surface elements. Mostt of the algorith hms used in thiss device are bassed on appliication of correelation pattern recognition meethods. The maain issue of thee present articlee is a descriptioon of these meethods appliication specificiities. © Published by Elsevier B.V.B.V. This is an open access article under the CC BY-NC-ND license © 2015 20 015The TheAuthors. Authorrs. Published byy Elsevier (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer--review under rresponsibility of the National Research R Nucleear University MEPhI M (Moscow w Engineering PPhysics Institutte). Peer-review under responsibility of the National Research Nuclear University MEPhI (Moscow Engineering Physics Institute) Keyw words: security hoolograms; automaatic identification; optical-electronnic scanner; distorrtion invariant pattern recognition..
1. In ntroduction Application of special holographic marrks, also know wn as security y holograms (SH) ( or opticaal variable deevices (OV VD), is one off the widespreead methods to t provide higgh level of seecurity of valuable papers and documen nts in Russsian Federatioon and worlddwide. High distribution oof SH is sup pported by itss application to secure su uch a docu uments as ID D cards, passpports, driving licenses, creedit cards, and d banknotes etc. This stattement raises high interrests in develoopment of autoomatic devicees of identity ccontrol of SH..
* Corresponding C autthor. Tel.: +7-9266 215 12 18; fax: +7-499-324-74-003. E-mail E address:
[email protected]
1875-3892 © 2015 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the National Research Nuclear University MEPhI (Moscow Engineering Physics Institute) doi:10.1016/j.phpro.2015.09.143
E. Yu. Zlokazov et al. / Physics Procedia 73 (2015) 308 – 312
309
High production difficulty of master-matrix, that is used for stamping the SH complex diffractive surface structures, causes the holograms produced by this matrix to be highly unique elements. Thus the operative verification whether the chosen SH were produced by its original master-matrix or not can be very important procedure for counterstand the falsification of SH. One of the methods to provide such a procedure can be based on observation of SH diffractive structure local specificities. In our previous publications we introduced the Optical-Electronic Scanner (OES) that was developed for automatic inspection of local SH specificities by numeric analysis of coherent diffraction responses of small surface elements [Zlokazov et. al. (2013), Zlokazov et.al. (2014)]. Capturing the diffracted light intensity by matrix photodetector and comparing the captured image responses with the same sample of the same local element of template SH in different points allows to obtain originality of the hologram under investigation. Specific problem of this approach is the requirement of precise positioning of control laser beam head to provide image matching with low error rate. Two ways were realized to decrease the affect of positioning error: 1) realisation of control head positioning in connection to SH global design and 2) application of the knowledge about possible distortions and variability in template image data that will be used in digital processing of surface diffraction responses captured by coherent control system of OES. In first case the aiming channel were developed to recognise SH design images and to provide the shift of optical head in front of the desired element of SH. In second case, using coherent control channel, we gathered data from different training examples of template SH with different shift error of optical head positioning. In both cases application of image processing methods based on invariant pattern recognition algorithms allowed to realize the SH identity procedure in OES with high precision. 2. Image recognition using distortion invariant correlation methods Correlation pattern recognition (CPR) is widely used technique in digital image processing. In the basis of these methods lies the calculation and analysis of correlation dependence between input data and template object functions. Advantage of CPR methods application is high precision in positioning of target object on image field and possibility of stable target recognition in presence of distortions due to application of distortion-invariant composite correlation filters (CCF) [Kumar (1992)]. CCF is a digitally synthesized two-dimensional data massive which elements can be complex or real values. The CCF calculation algorithms imply the application of training images set which represent the template object under different distortions determined a priori. One of the most investigated and efficiently applied CCF are filters with optimization of output correlation field parameters which are mathematically efficient and showed promising recognition capabilities of objects represented by grayscale images [Zlokazov et.al. (2012)].
Fig. 1. block-scheme of distortion invariant correlation image recognition method implementation based on CCF.
310
E. Yu. Zlokazov et al. / Physics Procedia 73 (2015) 308 – 312
Figure 1 represents the inpput image corrrelation recoggnition algoritthm using CC CF. Training im mages captureed by data acquisition ssystem during the preliminaary adjustmennt process of OES image reecognition unnit are subjectted to prep processing in oorder to supprress the undessired noises annd emphasize the characterristic features of template object o in diifferent possibble variations of representattion. Further tthe preprocesssed training im mages are sent nt to CCF syntthesis unit that stores thhe synthesizedd CCF in OES S memory. A Also the process of thresholld determinatiion (not show wn on s This proocess includes the figurre 1) is requuired for eachh synthesized CCF on thee preliminary adjustment step. statistical investiggation of CCF F response on input imagess of true and false f target ob bjects under ddifferent variaations i cessing unit m must be used for fo processing of input imagges during the OES deterrmined a priorri. The same image-preproc perfo formance in opperational reggime. The acq quired and preeprocessed inp put images arre sent to corrrelation proceessing unit “xcorr”. Thee output distriibution of corrrelation functtion amplitud de is subjected d to thresholdding operation n that h threshold vaalue calculated d on preliminaary adjustmennt step. If the value v searcches peaks annd compares itts values with of co orrelation peaak exceeds thrreshold value the decision is “true” and d input object is classified aas target objeect. If correelation peak vvalue is less thhan threshold value v the inpuut object is claassified as falsse. 3. Aiming C Channel Most of the real SH’s thhat used in ou ur researches during OES development have deviatioon of optical head posittion relative tto secured doccument body content such as printed tex xt, images, an nd other securrity elements. Thus the positioning p off control head in front of thee hologram el ement to be in nvestigated must m be made in independently y from posittion of docum ment “non hoolographic” elements. For tthis case an aiming a channel was moundded to provid de the recognition of SH H design imagees. Figure 2 reppresents the opptical scheme of aiming chaannel. Several LED-sourcees were used to eliminate th he SH surfaace in order tto reconstructt the encrypted images andd project thesee images on array a of photoo-detectors (P PDA). Prov viding a docum ment scan byy aiming chan nnel head allow ws to achievee the SH design images froom several an ngular posittions light soources. Mergiing of all thee achieved im mages alloweed the detaileed visualizatioon the SH design d elem ments. Neverthheless the reeference elem ments of SH ddesign that was w chosen to o provide thee mapping co ontrol appeeared roughly imaged in moost cases (see figure 2(b)). he decision of o chosen SH H design refeerence Application of correlatioon pattern reccognition metthods gave th elem ments rough im mages recogniition problem. We used com mputer modelss of reference elements subj bjected to disto ortion similar to those, w which were obbserved during the capturinng and mergin ng of different template exaamples. Developed algorithm allowedd to obtain thee positioning precision p of opptical head to be about 1mm m.
(a)
(b))
Fig 2: a) Aimiing channel opticcal scheme; b) exaample of the imagge processed and d merged in aimin ng channel, three different referencce elem ments were recogn nized using distorrtion invariant pattern recognition methods.
311
E. Yu. Zlokazov et al. / Physics Procedia 73 (2015) 308 – 312
4. Coheren nt control chaannel The local sttructure of SH H is representeed by diffractiion grating paatterns with siizes from 2 too 200um, perio ods in the range from 00.4 to 2.0um and a different orientations. Illumination of such a patterns by coheerent light pro ovides pearance of chharacteristic diiffraction peak ks in scatteredd light intensiity distribution n that can be captured by matrix m app pho oto-detector. E Elements with different set of gratinggs will produ uce diffraction n images witth different spatial s disttribution of difffraction peakks. To capture such an imagges an optical scheme represented on figuure 3 can be used u in coh herent control head construcction. m contained by four llaser diodes: blue, green, red, r and infrar ared. Flexible fiberThe schemee uses laser module optiical light guiddes were usedd to deliver an a illuminationn from all the lasers to op ptical head. T The opened en nds of optiical fibers were mounted on visualising disc such thaat to provide th he normal illu umination of SSH surface. Such S a con nstruction allowed to visuallise a 1st diffrraction orderss on a top surrface of visuallizer and captture of these bright b poin nts by photo ddetector arrayy (PDA) whilee DC order waas reflected back to visualiizer and attenuuated by fiberr ends mou unting. The ccollimating obbjective refraccts the emittedd laser light such s that the +/- 1st diffracction orders appear a focu used on lowerr surface of visualizer. As the sizess of SH surfacce grating pattterns are less than precisio on of optical head h positioniing error, two ways of solving s this prroblem were developed. d Th he first one waas the size off the laser spott on SH surfac ace was set up p to be abo out 0.5mm and the second one was the application oof CCF impleementation co oncept in trainning of the control chan nnel image reecognition algoorithms [Kum mar (1992)]. Foor every chosen cell of SH surface with tthe size 0.5x0 0.5mm the set of training images wass contained by y the diffractiion images off cell itself alsso the imagess of neighbourr cells a chann nel positioningg and increasse the werre added. Thiis allowed to compensate the lack of pprecision of aiming stab bility of OES performancee in presence of local streetches of SH surface. In order to increaase the stabillity of reco ognition algorrithms in preesence of posssible imperfeections of tem mplate examp ples of choseen SH, severaal SH exam mples must be used to gathher the training data. Thus tthe set of train ning spectral images i for eacch SH surfacee zone inclluded the speectral images of zone itseelf, spectral iimages of neeighbour zonees taken from m several tem mplate exam mples of chossen SH type. Applicationn of CCF impplementation methods in iimage processing algorithm ms allowed to obtain the mean disccrimination errror to be less than 5%, in so ome cases it w was less than 1%. 1
(a) (
(b)
Fig. 3: a) Cohereent control channeel optical schemee; b) image of SH H small surface eleement coherent diffraction responsse captured by OES control chaannel camera.
312
E. Yu. Zlokazov et al. / Physics Procedia 73 (2015) 308 – 312
5. Conclusion Represented article illustrated the examples of efficient implementation of distortion-invariant composite correlation filters in optical-electronic scanner of security holograms surface. Two OES basic units performance is based on recognition of images captured by its own image acquisition cameras. CCF application in Aiming Channel processing algorithms allowed to provide the mapping of Control Head using rough images of SH design elements with 1mm precision. Implementation of CCF methods for Control Head shift-invariant processing of surface local elements coherent diffraction responses allowed to provide the identification of holograms with about 0.5-5% of error probability. Acknowledgements This work was supported by grant 14-19-01751 from the Russian Science Foundation (RSF). References Evtikhiev, N.N., Shaulskiy D.V., Zlokazov E.Yu., Starikov R.S., 2012. Variants of minimum correlation energy filters: comparative study, Proc. SPIE, 8398, 83980G. Kolyuchkin, V.V., Zherdev, A.Yu., Zlokazov, E.Yu., Lushnikov, D.S., Odinokov, S.B., 2013. Correlation method for quality control of master matrix used for embossing security holograms, Proc. SPIE, 8776, 87760A. Tsiganov, I.K, Kolyuchkin, V.V., Piryutin, N.V., Odinokov, S.B., Talalaev, V.E., Zlokazov, E.Yu., 2014. Optoelectronic coherent scanner for operational control of authenticity of security holograms on passport documents, presentation on conference HOLOEXPO-2014. Vijaya Kumar, B.V.K., 1992.Tutorial survey of composite filter designs for optical correlators, Appl. Opt., 31, 4773-4801.
