EP1195045B1 - Automatic authentification of documents protected with security features - Google Patents

Abstract

Entire document is scanned with multi-spectral beam so that security features (2-6) on the document react with different spectral responses which are detected by sensors. Complete document is scanned with multi-spectral light to detect position of security features relative to coordinate system of the document. Position of markings is detected due to brightness, contrast, color and pattern differences relative to the rest of the document background. Then the security feature is checked only in the image areas corresponding to the position detected. Security features are machine readable text (2) with a line which does not reflect in selected scan of near infrared light, a two-dimensional bar code (3), diffractive element (4) which reflects diffraction pattern in coherent light of specific wavelength and picture (5) containing steganographically hidden information.

Description

Kategorie A:

für das menschliche Auge unter normaler Beobachtung unsichtbare optische Merkmale, welche unter bestimmten nichtkohärenten Beleuchtungs- und Aufnahmebedingungen visuell oder maschinell erkennbar sind.

Numerous non-electronic personal documents such as identity card, passport, visa documents or similar, value documents such as shares, banknotes, bills and similar. are provided with security features to protect against counterfeiting during production. These features can be divided into four broad categories:

Category A:

invisible optical features to the human eye under normal observation, which are visually or mechanically detectable under certain non-coherent illumination and acquisition conditions.

These include printed structures, introduced threads, etc. which are visible under ultraviolet light in the visible by fluorescence. Personal documents and banknotes are often provided with such features. The control is done manually by visual observation under a black light lamp, i. an illumination source which emits only in the invisible ultraviolet.

Included in this category are also incorporated or imprinted dyes with a special reflection behavior, often in the wavelength range of the near infrared (abbreviated NIR). The spectral distribution of the remitted infrared light contains a typical signature which can be checked by spectrometric methods. This is called a machine detection.

Kategorie B:

für das menschliche Auge unter normaler Beobachtung sichtbare optische Merkmale, welche sich unter einer bestimmten, nichtkohärenten Beleuchtung verändern.

Also included in this category are the information hidden in images, the so-called steganography, which are introduced by certain modulation methods of the grid in the rasterized image printing and can be extracted by machine using image processing methods.

Category B:

visual features visible to the human eye under normal observation, which change under a certain non-coherent illumination.

Kategorie C:

für das menschliche Auge unter normaler Beobachtung sichtbare optische Merkmale, welche unter bestimmten, kohärenten Beleuchtungsanordnungen aufgrund von Lichtbeugung bestimmte Beugungsmuster erzeugen.

These include, for example, on many personal documents applied strokes, which are visible in the visible, but with illumination of a certain wavelength, usually located in the NIR, disappear.

Category C:

visual features visible to the human eye under normal observation which produce certain diffraction patterns under certain coherent illumination arrangements due to light diffraction.

Kategorie D:

für das menschliche Auge unter normaler Beobachtung sichtbare optische Merkmale, welche einen informationstragenden, maschinell lesbaren Code darstellen.

These include diffraction gratings applied to clear or metallized films, which generate specific diffraction patterns in a particular laser illumination, which can be checked by means of photodetectors or cameras.

Category D:

visual features visible to the human eye under normal observation, constituting an information-bearing, machine-readable code.

These include the machine-readable imprints, 1-dimensional and 2-dimensional barcodes, 2-dimensional matrix codes, and the like. which are decoded with the help of scanners or camera systems.

According to the state of the art, only the features of category D are currently evaluated automatically with the aid of readers. These devices contain one or more cameras as well as one or more spectrally different illuminations and are able to automatically decode machine-readable documents such as the so-called lCAO line for personal documents, applied 1D or 2D barcodes. The positions of these codes relative to a document-oriented coordinate system are usually defined by international standards. The documents must be placed in a specific position on the viewing window of these devices in order to work correctly.

Therefore, these automatic devices perform only the "decode" functions; they can not be used with documents in which the position of the codes is not defined or the position of the document is free. In order to be able to work correctly nevertheless, an automatic function "positioning" is needed, i. automatic finding of the code or codes on the document and automatic positioning of an evaluation window at these locations.

The documents saved under categories A (except UV characteristics), B and C can only be machine-written, i. be checked by means of technical, usually optical procedures. There are individual solution proposals, such as the use of micro spectrometers for spectral analysis, image processing methods for the recognition of the information contained in the printed images, which each fulfill a specific task.

1. a) aufgrund von Produktionstoleranzen kann z.B. die Position eines aufgeprägten Beugungsgmerkmals schwanken
2. b) aufgrund von fehlenden Normen gibt es keine einheitliche Vereinbarungen
3. c) aufgrund von unsicherer Position des Dokumentes im Bildfeld des automatischen Prüfgerätes sind die Positionen relativ zum Koordinatensystem des Abtasters nicht bekannt

Recent documents are often backed up at the same time as many of the A, B, C and D categories. Very often, the geometric position of these features on the document is not well defined:

1. a) due to production tolerances, for example, the position of an impressed diffraction feature may vary
2. b) there are no harmonized agreements due to missing standards
3. c) due to the uncertain position of the document in the image field of the automatic tester, the positions relative to the coordinate system of the scanner are not known

These documents can either not be checked automatically at the moment or semi-manually. The user must place and position the document in a specific orientation. This process is uncertain and takes time.

In addition, as the variety of concurrent features greatly increases, there is a great need for systems that automatically detect which features are present in the document, automatically locate and verify them. Such systems are not available today.

The verification of the document itself is called "authentication". The question of whether the document shown, e.g. An identity card also belongs to the carrier is called "Verification".

As a result of the elimination of intra-European customs controls and the growth in passenger numbers, the number of highly secure personal documents that can be authenticated and verified in the airports in a very short time is growing rapidly. Today's visual inspection, possibly with the help of simple UV lighting systems or ICAO handheld readers can not meet this requirement. Similar conditions also apply to the examination of value documents such as banknotes, secured shipping documents and the like. Ä. Documents. There is therefore a great need for secure, fast and robust automatic testable and readable documents.

The object of the invention is a method and an arrangement for the fast and automatic authentication of documents secured with numerous optical features, in which the position of the features relative to the coordinate system of the document is not or only approximately known.

This object is achieved by the technical teaching of claims 1 or 2. The document is illuminated in its entirety or in some areas in a first step with lighting means whose spectral and geometric distribution, polarization, coherence and time course of the optical properties of the illumination means are selected so that an imageable by imaging sensors image arises in which the Positions of the security features due to brightness, contrast, color, pattern differences relative to the rest of the document background are automatically recognizable with methods of image processing, that in a second step, the examination of the corresponding features is performed only in the image areas, which found in the first step Corresponding positions and for this purpose the corresponding image areas are illuminated in such a way that the corresponding features generate images and / or spectral signatures, from which with methods of image processing and / or spectrometry that for d ie checking the characteristics required information can be derived.

It does not matter whether a single path or a multipath scan takes place; in both Verfahrensbetriebsweisen can be worked according to the invention.

The classification of the security features into the described categories A to D here represents only a division useful for simpler description of the inventive concept. Other features which do not fit into one of these explicit categories are also part of the inventive concept.

The inventive method allows the determination of the position of the security features, both relative to the coordinate system of the scanned document, and of course to the coordinate system of the scanner itself. By detecting the sheet edge against the dark Background of the cover, the scanner "knows" how the document is aligned to its own coordinate system.

Thus one can find the position of the features relative to the document coordinate system, which is important if there are no standardized rules for this or even if due to production tolerances the exact position is not known.

This position information is first found on the first scan relative to the coordinate system of the scanner, regardless of the exact orientation of the document on the scanning window of the scanner. With this information, the corresponding illuminations and / or line sensors of the scanner can be moved to these points of the document with high precision and the actual examination of the authenticity features can be carried out with high precision.

If one wishes to also check whether the position of the security features coincides with the desired positions hidden by a standard or by other features relative to the document coordinate system, a transformation between the scanner coordinate system and the coordinate system of the document itself is established by eg The top left corner of the sheet is defined as the X0, Y0 origin of the document. By detecting at least two corners in the sample window, the translation rotation matrix is established between the scanner coordinate system and the document coordinate system.

This double effect is further characterized in the specific part of the description and in the claims.

Among other things, the invention consists of finding, with suitable illumination methods, the position of the authenticity features relative to the document and / or relative to the scanner, these being Types of illumination may be different from those used in testing the authenticity features.

Part of the concept of the invention is also the storage and archiving of all captured pictorial information of the secured document in a database simultaneously with the review of the document.

For clarification, the concept of the invention is explained on the basis of a typical document, namely a personal document secured by a combination of numerous features.

Fig. 1

zeigt den schematischen Aufbau eines Personaldokumentes mit den Sicherheits-Merkmalen ICAO Zeile mit NIR-Unterstrich, Barcode, Portraitdruck mit im Bild versteckten Informationen und ein auf einer metallisierten Folie aufgebrachtes Beugungsmerkmal

Fig. 2

zeigt beispielhaft die einzelnen Positionen der zu prüfenden Merkmale, welche unabhängig von der Lage des Dokumentes auf dem Gerät und von der genauen Position bezüglich des Dokuments selbst bestimmt werden

Fig. 3

zeigt schematisch einen typischen Lösungsvorschlag, bei welchem das gesamte Dokument mit einem hochauflösenden Zeilenkamerascanner unter verschiedenen spektral, geometrisch und zeitlich unterschiedliche Beleuchtungen abgetastet wird

The following figures are used for this:

Fig. 1

shows the schematic structure of a personal document with the security features ICAO line with NIR underscore, barcode, portrait print with hidden information in the image and a deposited on a metallized film diffraction feature

Fig. 2

shows by way of example the individual positions of the features to be tested, which are determined independently of the position of the document on the device and the exact position with respect to the document itself

Fig. 3

schematically shows a typical solution proposal, in which the entire document is scanned with a high-resolution line camera scanner under different spectrally, geometrically and temporally different illuminations

The idea of the invention is explained by way of example but not restrictively on the basis of a personal document secured by several features at the same time.

1. a) durch einen maschinell lesbaren Text 2 mit einer Linie, welche bei einer spektral selektierten Beleuchtung im Nahen Infrarot nicht mehr reflektiert
2. b) durch einen 2-dimensionalen Barcode 3, welcher kryptographisch gesicherte Informationen enthält
3. c) durch ein diffraktives Element 4, welches im Zentrum einer sternförmigen metallisierten Folie aufgebracht ist und welches bei kohärenter Beleuchtung bei einer spezifischen Wellenlänge bestimmte Beugungsmuster zurückstrahlt
4. d) durch steganographisch im Portraitbild 5 versteckte Informationen.

According to FIG. 1, this document 1 is to be secured by the following features which are to be positioned and checked by machine:

1. a) by a machine-readable text 2 with a line which no longer reflects at a spectrally selected illumination in the near infrared
2. b) by a 2-dimensional barcode 3, which contains cryptographically secured information
3. c) by a diffractive element 4, which is applied in the center of a star-shaped metallized film and which radiates certain diffraction patterns with coherent illumination at a specific wavelength
4. d) by steganographically hidden in the portrait image 5 information.

The text field 6 contains no security features and no machine-readable information. However, it still has to be recognized by machine as a field that is not to be evaluated.

The layout of the document can be described by the position of these individual features with respect to an x, y coordinate system aligned at the edge of the document. These positions should not or only insecure be known when scanning by the arrangement according to the invention, so that according to the invention in a first step, these positions of these features must be determined with respect to the coordinate system.

FIG. 2 shows by way of example an arrangement for carrying out the method step "positioning".

The document 1 is placed on the device 12 with the page to be evaluated. Through the viewing window 13, the page is optically scanned line by line with the motor-driven sensor head 15, which is linearly moved along the storage 14-.

By way of example, some methods for finding the individual feature positions and their identification by a specific illumination according to the inventive concept for the first method step will be described.

1. a) durch die Beleuchtung im Nahen Infrarot verschwindet die durchgehende Linie 11 im Textfeld 2 während die Linie 11 im sichtbaren Licht deutlich erkennbar bleibt. Die Texte, sowohl im Feld 2 als auch im Feld 6, sind sowohl im NIR-Bereich als auch im sichtbaren Licht deutlich erkennbar, da schwarzer Druck einen breiten Bereich der Beleuchtungswellenlängen absorbiert.Durch Auswertung der bildhaften Informationen, welche sich in den beiden Bildern A und B deutlich unterscheiden, kann das Textfeld 2 von dem Textfeld 6 unterschieden werden.
   Fig. 3 zeigt die Bilder A und B mit den entsprechenden Bildinhalten. Durch die Position der zwischen A und B verschwindenen Linie 11 kann die Position des maschinell auszuwertenden Textfeldes bezüglich des xy- Koordinatensystem ermittelt werden. Gleichzeitig wird durch das Verschwinden der Linie 11 im Bild A bereits der zweite Verfahrensschritt, nämlich die Prüfung des Sicherheitsmerkmals mit abgearbeitet.
2. b) das Feld 3 mit dem 2-dimensionalen Barcode ist in beiden Bildern A und B sichtbar. Es kann mit dem Fachmann bekannten Methoden der Mustererkennung anhand der hochfrequenten Ortsfrequenzen, welche das Codemuster auszeichnen, vom übrigen Bildinhalt erkannt werden
3. c) die automatische Positionierung des farbigen Portraitbildes 5 kann z.B. dadurch erkannt werden, daß die Farbpigmente des Portraits 5 ihr Reflexionsverhalten zwischen NIR- Beleuchtung und Beleuchtung im sichtbaren Bereich stark ändern, während das Reflexionsverhalten des Hintergrundpapiers ähnlich bleibt.
   In Fig. 3 ist diese Eigenschaft durch einen schwächeren Kontrast zwischen Portait und Hintergrund im Bild A verglichen mit dem Kontrast in Bild B verdeutlicht. Dieser lokale Unterschied erlaubt es dem Fachmann der Bildverarbeitung, die Position des Portraits bezüglich des Dokumentenkoordinatensystems zu bestimmen.
4. d) durch die Beleuchtung des Dokumentes mit der gerichteten Beleuchtung 17 kann die Position der metallisierten Folie, welche im Zentrum das diffraktive Element trägt, erkannt werden. Während unter der diffusen Beleuchtung 16 der Kontrast zwischen metallisierter Folie und Hintergrundpapier relativ schwach erscheint und dies auch relativ unabhängig von der beleuchtenden Wellenlänge, wird durch die schräg gerichtete Beleuchtung der Kontrastunterschied zwischen der spiegelnd reflektierenden Metallfolie und dem Papier des Hintergrundes sehr groß.
   Die schräg einfallende gerichtete Beleuchtung reflektiert an der Eintrittpupille des abbildenden Objektivs 18 vorbei. Dadurch erscheint die Folie dunkler als der Hintergrund. Dies ist in Fig. 3 im Bild C dargestellt.
   Es ist dem Fachmann der Optik bekannt, daß dieser Kontrastunterschied auch durch eine sog. Hellfeldbeleuchtung erzielt werden kann, bei welcher die Beleuchtung so angeordnet wird, daß sie durch die spiegelnde Folie direkt in die Pupille des abbildenden Objektivs reflektiert und damit heller als der diffus reflektierende Hintergrund erscheint.
   Wird die schräg einfallende gerichtete Beleuchtung gleichzeitig zumindest teilweise kohärent gewählt so wird gleichzeitig erreicht, daß das eigentliche Sicherheitsmerkmal im Zentrum des metallisierten Sterns, nämlich das diffraktive Element mit einer der Beugungsordnungen in die Pupille der abbildenden Optik zurückstrahlt und damit hell erscheint.
   Es ist für den Fachmann der Bildverarbeitung ein leichtes, sowohl die Position des metallisierten Sterns als auch die des diffraktiven Elementes in seinem Zentrum aufgrund dieser hohen Kontrastunterschiede zu erkennen.

To find the positions of the text field 2, the 2-dimensional bar code 3 and the port 5, the document 1 is illuminated during the scanning by the diffuse illumination 16 alternately with a wavelength in the near infrared and with a wavelength in the visible light. This can be done by temporally switching of appropriately spectrally selected light-emitting diodes (LEDs). If the switchover takes place synchronously with the line frequency, two arithmetically interlaced images A and B are produced in the image memory of the arithmetic unit 21, where A is intended to illuminate the area illuminated in the NIR range and B to represent the image illuminated in the visible.

1. a) the illumination in the near infrared disappears the continuous line 11 in the text field 2 while the line 11 remains clearly visible in visible light. The texts, both in field 2 and in field 6, are clearly visible both in the NIR range and in visible light, since black printing absorbs a wide range of illumination wavelengths. By evaluating the pictorial information shown in the two images A and B clearly distinguish the text field 2 from the text field 6.
   Fig. 3 shows the images A and B with the corresponding image contents. Due to the position of the line 11 disappearing between A and B, the position of the text field to be evaluated by machine with respect to the xy coordinate system can be determined. At the same time by the disappearance of the line 11 in the image A already the second process step, namely the examination of the security feature processed with.
2. b) the field 3 with the 2-dimensional barcode is visible in both images A and B. It can be recognized from the rest of the image content using methods of pattern recognition known to those skilled in the art on the basis of the high-frequency spatial frequencies which characterize the code pattern
3. c) the automatic positioning of the colored portrait image 5 can be recognized, for example, by the fact that the color pigments of the portrait 5 change their reflection behavior between NIR illumination and illumination in the visible range, while the reflection behavior of the background paper remains similar.
   In Fig. 3, this property is illustrated by a weaker contrast between portability and background in image A as compared to the contrast in image B. This local difference allows the image processing expert to determine the position of the portrait relative to the document coordinate system.
4. d) by the illumination of the document with the directional illumination 17, the position of the metallized film, which carries the diffractive element in the center, can be detected. While the contrast between the metallized film and the background paper appears relatively weak under the diffuse illumination 16, and this also relatively independent of the illuminating wavelength, the oblique illumination makes the difference in contrast between the specularly reflective metal foil and the paper of the background very large.
   The obliquely incident directional illumination reflects past the entrance pupil of the imaging lens 18. This makes the slide appear darker than the background. This is shown in FIG. 3 in FIG.
   It is known to those skilled in the optics that this contrast difference can also be achieved by a so-called bright field illumination, in which the illumination is arranged so that it reflects through the specular film directly into the pupil of the imaging lens and thus brighter than the diffuse reflective Background appears.
   If the obliquely incident directional illumination is simultaneously selected to be at least partially coherent, it is simultaneously achieved that the actual security feature in the center of the metallized star, namely the diffractive element with one of the diffraction orders, reflects back into the pupil of the imaging optics and thus appears bright.
   It is easy for the person skilled in the image processing art to recognize both the position of the metallised star and that of the diffractive element in its center due to these high contrast differences.

Since the positioning step does not require the high image resolution, which requires the actual machine evaluation of the text field, the 2D code, the information hidden in the portrait, and the diffraction patterns, the recording of the individual images associated with the respective illumination types can be recorded with reduced resolution become. This can technically be done by the imaging line sensor working at a coarser pitch during this positioning step.

Another idea of the invention is to switch the individual illuminations in synchronism with the line clock of the imaging line sensor and thus to obtain a series of interlaced lower resolution images in a scan, from which due to the systematic contrast and pattern differences, the positions of the individual features using methods of pattern recognition and Image processing are determined.

It is important that both in the single-path and in the multipath method each of the entire document is illuminated with a multispectral radiation, so that the authenticity features distributed correspondingly on the document react with different spectral responses and are also detected with spectrally different receiving sensors ,

In the simplest case, clocked light emitting diodes are used as radiation sources because only a relatively low resolution is required.

There is a line sensor which includes a row of image sensing sensors, which line sensor is now moved along the document. Such a silicon sensor would be a z. B. have a spectral sensitivity of visible light to the NIR spectrum. In the method of this line sensor, the LEDs are then alternately driven per line, which emit visible light and alternately the LEDs, which emit light in the NIR spectral range. This creates in the image memory, which receives the signal, a partially interlinked overall image, which z. B. consists of 1000 lines. Of these, 500 lines are subjected to signals corresponding to the spectral response in visible light, while 500 lines are subjected to signals corresponding to the spectral response of the authenticity features in the NIR range.

Using common software methods, these two image areas are separated and processed separately in the image array.

This process is carried out mainly in the single path method, because this is particularly fast and provides a relatively low resolution.

In the image memory would be now z. B. the locations and types of authenticity features according to the figures 3a to 3c virtually stored. In the infrared range z. B. then in the one image to be examined, the portrait no longer visible according to paragraph 5, while the other authenticity features, eg. B. the defractive element 4, in the NIR range is easy to read.

In the visible (white light) but, for example, the Portrait 5 is very easy to read.

From this template, z. Example, a 1000-line image array containing 500 lines of one image and 500 lines of the other image is now determined by difference determination between the two read images, where the authenticity features are.

In relation to the portrait image 5, therefore, the portrait image would be easily readable, while in the NIR region the portrait is only badly readable or completely disappears, so that the determination of this difference image clearly establishes that one is in visible light at this location readable portrait image is located.

Thus, the location of this portrait image is clearly determined on the document.

Of course, not only the location of the portrait can be deduced from the white light portion, but also z. B. color information or other information that can be derived from the white light spectrum.

In an analogous manner, this also applies to the location of the other elements in which z. For example, with respect to the machine-readable text 2, line 11 in the NIR range disappears while it is present in the visible region that so that without further notice the location of this machine-readable text 2 can be determined.

Of course, the invention is not limited to the determination of a differential image of two different spectral wavelengths, but it is - as previously stated in the invention - even the direction of the incident light and the resulting spectral response is a measure of the location and type of used authenticity features.

In the current claim 3 it is also stated that the geometric distribution of the illumination is used to find the authenticity features. The term "geometric distribution" is z. B. understood, the direction of the incident light, but also the geometric distribution, z. B. in the form of a lattice structure or a particularly distributed, geometric structure such. As concentric rings or other geometric arrangements of the illumination sources.

All mentioned in the current claim 3 features may contribute individually or in different combinations to the solution of the invention task.

Likewise, the degree of polarization of the illumination source can be used to obtain a particular readout of authentication features.

Also, the invention is not limited to the use of coherent light, but non-coherent or medium-coherent light may be used.

The term "temporal activity" is understood to mean that the authenticity feature is illuminated with a specific pulse and the impulse response and its decay are examined in a specific time window, so that the temporal activity of the impulse response can also be used to read the authenticity feature.

All measures mentioned above are therefore initially used to determine the authenticity features that are distributed on the document.

There are now several possibilities for carrying out the method according to the invention:

According to a first embodiment it is provided that - in accordance with the above description - first scans the entire document with a lower resolution and determines the location of the authenticity features. This has been explained in detail above.

Now, if the location of the authenticity features has been determined by appropriate signal examinations, scanning is carried out in a second scan with higher resolution, and the resulting signal image is only examined in more detail at the locations and subjected to special investigation methods, which for the authenticity features found specifically in this location apply (multi-path scanning).

Of course, the second scan can be limited to specific areas of the document to save time. However, it is important that the second scan can be performed at a relatively high speed, because only at certain points where authenticity features were detected, would have to be scanned with high resolution.

Ie. that is, you can change the scan speed, over the length of the document, so that you scan at higher resolution only at the locations where there are authenticity features, while driving over the "empty" area of the document at a higher speed. This significantly increases the processing speed.

In another embodiment (additional claim 2), the entire document with the highest, for the recognition of the authenticity features scanned resolution once, then immediately determine the location of the authenticity features from the scanned image and the resulting signal response.

If the location of the authenticity features is determined from the signal response, then at this location, the correspondingly obtained signal can be examined more accurately for authenticity features by software.

Thus, a digital image is generated in a multi-line array of the entire document and the authenticity features contained thereon. In the next step, the authenticity check of each individual feature now takes place only exactly at the place where the authenticity feature was previously determined. This saves especially computing time and provides a much higher production capacity for the device according to the invention. Incidentally, it is now possible in a second scan passage to bring certain illumination means only for this recognized location and to illuminate this recognized location and perform the corresponding authenticity check.

Thus, all detection methods and features for the detection and verification of this authenticity feature can be used individually or in combination with each other, as described in the general part (and in particular in claim 3). This applies in particular to the type and form of radiation, to the spectral distribution of the radiation source and the like. And always focused on the recognized place of the authenticity feature and only on this place. Thus, there is the significant advantage that now due to the location of the authenticity feature found the laser, which z. B. is required for the detection of an authenticity feature, can now be positioned with high accuracy with respect to this authenticity feature to expose this and measure the radiation response accordingly. This results in a much better signal-to-noise ratio, which is no longer dependent on inaccuracies in the positioning of the document.

This type of evaluation of authenticity features in the case of authenticity features, which are constructed on the basis of diffraction structures, is of great importance. Here, it depends on a very accurate positioning of the radiation source with respect to the authenticity feature, and here are the advantages of the invention.

Other authenticity features can also be examined, e.g. For example, the fact that there is a certain fluorescent thread at a certain point in a document. This thread is then illuminated with a corresponding radiation source. Again, the positioning of the radiation source with respect to the thread is extremely important in order to obtain a favorable signal-to-noise ratio. This results in the further advantage that due to the detected location and the location of authenticity features other areas in the document can be approached to possibly find singular hidden (ie, arbitrarily mounted) authenticity features. For example, in bank robberies involving blackmail, certain documents (banknotes) are specially marked, and such identifiers may well be located elsewhere than they are on the document by design or by default. Such hidden and singular existing authenticity features can be quickly and easily recognized and evaluated with the teaching of the technical invention.

A further embodiment according to the invention is that all the spectrally different illuminations are switched on simultaneously and the corresponding images are simultaneously imaged on different image sensors in the beam path of the imaging optics by one or more wavelength-dependent beam splitters.

The arrangement described represents only an example of the method and the design of the arrangement according to the invention. The use of matrix sensors, point-by-point scanning optical sensors or on a Chip-integrated, spectral, coherence and / or direction-dependent sensor elements are part of the invention and the expert in optics known, so that they need not be further explained here.

According to the invention, the diffractive features are checked by using a coherent light source, e.g. a laser diode is directed to the position, which were determined in the first step. This can be done, for example, by mechanically moving a laser diode so that its light beam is directed onto the diffractive element directly or via optical intermediate elements such as mirrors, prisms, beam splitters, etc. The resulting diffraction image is e.g. converted via a ground glass into a real image of punctiform structures and detected by an image sensor. The examination of the feature can be done by checking the position and arrangement and the radiometric properties such as brightness ratios of these structures in the image detected by the image sensor 20 with the aid of a computing unit 21.

According to the invention, the position of a diffractive feature can also be determined by locally illuminating the document in a search grid with a coherent illumination source in the presumed positions of the feature and obtaining an image of this local region with the aid of an image sensor. If the diffractive element is hit, significant, a priori, definitions of the diffractive element appear in known structures so that the positioning has occurred when these expected structures occur

The definitions of the diffractive element with respect to its position, the type of diffraction structures generated, the information or decoding rule contained therein are stored according to the invention in other machine-readable features of the same document, eg cryptographically encoded in 2-dimensional barcode or in the steganographically hidden information in the portrait image , This procedure has the big one The advantage that such standards do not need to have fixed standards for the specific document. It allows you to create individual diffractive elements for each document and to hide the information required for verification in the same document but in one or more other security features.

This method of the individual security feature, whose test specification is encoded in one or more other features of the same document, according to the invention not only applies to the diffractive security features, but also fundamental for all others. For example, the shape or type of UV-visible features can be hidden in the portrait image. With the decoding of this information, the test specification for the automatic over-checking of the UV feature (s) is ready at the same time, without having to ask for information in a database, a standard or another source of information stored outside the actual document itself.

drawing Legend

1

document

2

Machine readable text

3

barcode

4

diffractive element

5

Portrait image

6

text box

7

8th

9

10

11

line

12

device

13

window

14

storage

15

sensor head

16

diffused lighting

17

directed lighting

18

imaging lens

19

20

image sensor

21

computer unit

Claims (11)

1. Method for the quick, automatic verification of documents secured with optical features, in which the positions of the features relative to the coordinate system of the document and/or the automatic testing instrument are not known or are known only inaccurately, having the following steps:

   1) in the first method step, the document is examined for the presence of security features with a low resolution of the detection unit;

   2) in the second method step, the location of the individual security features on the document is determined with a likewise low resolution of the detection unit;

   3) in the third method step, at the location of the security feature found, the latter is examined for authenticity with high resolution,

   in the first method step, the document being illuminated with a wavelength in the near infrared and with a wavelength in the visible light and depicted on an image sensor,
   and in the second method step, the positions of security features being detected on the basis of lightness, contrast or colour differences between the images.
2. Method for the quick, automatic verification of documents secured with optical features, in which the positions of the features relative to the coordinate system of the document and/or the automatic testing instrument are not known or are known only inaccurately, having the following steps:

   1) in the first method step, the document is scanned with the highest resolution necessary for the detection of the authenticity features and the location of the authenticity features is determined immediately from the scanned image,

   2) in a second method step, at the location of the security features determined, the signal response obtained there is examined with high accuracy by means of software analysis of the signal response only at this location,

   in the first method step, the document being illuminated with a wavelength in the near infrared and with a wavelength in the visible light and depicted on an image sensor, and the positions of security features being detected on the basis of lightness, contrast or colour differences between the images.
3. Method according to Claim 1 or 2, the position of a diffractive feature (4) being determined in that, in a search pattern, a coherent illumination source illuminates the document locally in the supposed positions of the feature and an image of this local region is obtained with the aid of an image sensor, and in that this position is varied until the expected image structures, known a priori from the definition of the diffractive element (4), occur.
4. Method according to one of Claims 1 to 3, the illuminations being switched alternately in synchronism with the line change of a recording image sensor (20) and thus an interleaved arrangement of images being produced, which, line by line, in each case correspond to one illumination required for the detection of the position of the security features.
5. Method according to one of Claims 1 to 3, a plurality of the required illumination sources (16, 17) being active simultaneously and the image of the document surface corresponding to each illumination source being projected onto a dedicated image sensor via beam splitters.
6. Method according to one of Claims 1 to 5, the type of security feature present being determined from the characteristics of the determined position of the security features.
7. Method according to one of Claims 1 to 6, the information and definitions needed for the positioning and testing of a security feature being stored in one or more other machine-readable features of the same document.
8. Method according to Claim 7, the definitions of a specific security feature for each document (1) being defined individually and the part of the definitions which is required for the positioning and testing being encoded in one or more other machine-readable features of the same document (1).
9. Method according to one of Claims 1 to 8, the image and other information accumulating during the automatic evaluation of the document being stored in a document database, from which further characteristics of the document about the correctness of the security features and emerging from the machine-readable information can be determined at any desired times.
10. Arrangement for implementing the method according to one of Claims 1 to 9, in which the document to be tested is scanned with an optical arrangement of various illumination sources and one or more image sensors covering the document, and the images associated with the individual illumination sources and produced by the image sensor or sensors are converted into electric digital signals and stored and evaluated in a computing unit, for the purpose of checking diffractive elements (4), a coherent light source formed as a laser diode being deflected onto the position determined in the first method step and the laser diode being moved mechanically in such a way that its light beam is deflected onto the diffractive element (4) directly or via intermediate optical elements such as mirrors, prisms and beam splitters, and the diffraction image produced being converted into a real image via a matt disc and registered by an image sensor.
11. Arrangement according to Claim 10, in which all the spectrally different illuminations are switched on simultaneously and the corresponding images are projected simultaneously onto various image sensors by means of one or more wavelength-dependent beam splitters in the beam path of the imaging optics.

Images