EP2394250B1 - Method and device for verifying document using a wavelet transformation - Google Patents
Method and device for verifying document using a wavelet transformation Download PDFInfo
- Publication number
- EP2394250B1 EP2394250B1 EP10705107.0A EP10705107A EP2394250B1 EP 2394250 B1 EP2394250 B1 EP 2394250B1 EP 10705107 A EP10705107 A EP 10705107A EP 2394250 B1 EP2394250 B1 EP 2394250B1
- Authority
- EP
- European Patent Office
- Prior art keywords
- document
- features
- wavelet
- pixels
- documents
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims description 63
- 230000009466 transformation Effects 0.000 title claims description 35
- 238000012795 verification Methods 0.000 claims description 54
- 230000003595 spectral effect Effects 0.000 claims description 40
- 238000001514 detection method Methods 0.000 claims description 16
- 238000011156 evaluation Methods 0.000 claims description 15
- 238000009795 derivation Methods 0.000 claims description 10
- 238000013528 artificial neural network Methods 0.000 claims description 5
- 238000012706 support-vector machine Methods 0.000 claims description 5
- 239000013598 vector Substances 0.000 claims description 4
- 230000006870 function Effects 0.000 claims description 3
- 230000003746 surface roughness Effects 0.000 claims description 3
- 238000000844 transformation Methods 0.000 claims 1
- 230000000875 corresponding effect Effects 0.000 description 12
- 238000000354 decomposition reaction Methods 0.000 description 5
- 230000005284 excitation Effects 0.000 description 4
- 238000013507 mapping Methods 0.000 description 4
- 230000005855 radiation Effects 0.000 description 4
- 230000003068 static effect Effects 0.000 description 4
- 238000010972 statistical evaluation Methods 0.000 description 4
- 239000003086 colorant Substances 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000005684 electric field Effects 0.000 description 2
- 238000005401 electroluminescence Methods 0.000 description 2
- 230000005670 electromagnetic radiation Effects 0.000 description 2
- 230000005484 gravity Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000000926 separation method Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000011109 contamination Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000004020 luminiscence type Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000001454 recorded image Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- 238000011179 visual inspection Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07D—HANDLING OF COINS OR VALUABLE PAPERS, e.g. TESTING, SORTING BY DENOMINATIONS, COUNTING, DISPENSING, CHANGING OR DEPOSITING
- G07D7/00—Testing specially adapted to determine the identity or genuineness of valuable papers or for segregating those which are unacceptable, e.g. banknotes that are alien to a currency
- G07D7/20—Testing patterns thereon
- G07D7/202—Testing patterns thereon using pattern matching
Definitions
- the invention relates to a method and a device for verifying documents, in which a plurality of images of the document are detected, characteristics are derived from the images, the features are compared with expert knowledge, and verification, for example, authentication of a document by means of a Result of the comparison.
- One aspect of the method relates to the derivation of the features.
- documents which may include, for example, security documents, such as in particular passports, identity cards, identification cards, visas, driver's licenses, etc., or documents of value, in particular banknotes, checks, credit cards, postage stamps, etc., with regard to their To check authenticity and authenticity.
- security documents such as in particular passports, identity cards, identification cards, visas, driver's licenses, etc.
- documents of value in particular banknotes, checks, credit cards, postage stamps, etc.
- Verification also means identifying a document as a document type. This means that recognition of a banknote as a document of a specific value of an issuing authority is also understood as verification.
- features of the document are analyzed for verification and compared with expert knowledge, which includes information about the individual features and their design and / or about feature combinations of a document in order to perform a verification on the basis of, for example, a combination of features.
- a device and a method for reading in and authenticating documents are known. These documents include, for example, passports, driver's licenses, identity documents and value documents. They are read in to receive and verify information stored on it and to examine and record it for security features. The security features are used to check whether the document is genuine, fake or falsified. For reading a first image of the document is taken. The figure determines the size of the document. Based on the size of the will Associate a document with a class of documents that are the same size. Reference documents for all document types are segmented and each segment has a correlation number. A correlation number is determined for each segment of the first document, which includes predefined static information immutable to a document type.
- correlation numbers are compared with the correlation numbers of the corresponding segments of the reference documents that are the same size as the examined document.
- the identity of the document is performed by comparing the correlation numbers. Taking advantage of the knowledge of the identity of the document, a set of security features is captured and the identified document is examined using the security features to determine if the document is genuine, a forgery, or a corrupted document.
- a method and apparatus for validation and verification are known.
- An apparatus and method is described that automatically identifies and validates a document to validate the identity of a user of the document and to verify that the user is authorized to participate in an activity represented by the document, and to compare the information on the document with information from databases to determine if there are known concerns about a document's authenticity or a user's identity and / or trust.
- a document class is narrowed down by first determining the size of the document. An attempt is then made to identify the document as one of a group, ie, its membership in a document type, by examining colors or other characteristics at specific locations on the document until the document is identified as belonging to a document type.
- the order in which the locations of the document are searched depends on a history of the document types that have been identified.
- An identified document is verified by looking up reference information about the document type. Using this information, certain content is read in and only special security and other features are used to validate the document.
- the contents of the document are matched against a database to verify its contents. Other databases are checked for stolen documents, known valid documents, and travel or access restrictions.
- biometric Data on the document and biometric data obtained directly from the user of the document are compared with each other to verify the user.
- the known methods are essentially based on a direct pattern comparison of a captured image of the document.
- preferably only limited imaging areas are compared with each other in order to avoid only static areas, i. such areas, which are identical for all documents of one type, for example, for all passports of an issuing nation, are compared. Since different types of documents have such static information at different locations, different segments or areas must be evaluated for comparison. Only when the document is identified, for example, as a passport of a particular nation, does a subsequent review of security features take place, with the aid of which a genuineness and authenticity of the documents is checked.
- the known methods are thus based on a multi-stage examination procedure, in which it is necessary to first identify the document and then to compare known security features for the identified document. Falsifications, for example a change of printed personal data, on an original security document can be detected with the known methods only with difficulty or not at all.
- an authentic security document comprises security features that are integrated into the security document.
- the security features include characteristic visual features that are intrinsic to the method used to fabricate the security document.
- the authentication method comprises the steps of (i) capturing an image of at least one region of interest of the surface of a document to be authenticated, the region having at least one security feature, (ii) digitally processing the image by decomposing the image into at least one subspace that is high resolution Details of the mapping include and extracting classifying features from the subspace and (iii) deriving an authenticity score for the document based on the derived classification features.
- a method for determining the denomination of a bill is known.
- a wavelet transformation is used.
- the image of a banknote is preprocessed so that edges of the image are captured.
- the preprocessed image is wavelet transformed a predetermined number of times and the magnitude of the wavelet coefficient absolute values compared to a reference value to more accurately determine the value of the bill without being affected by partial or total banknote contamination ,
- the invention is based on the technical problem of providing a method and a device for verifying documents, in particular security and / or value documents, with which a quick, simple and reliable verification of documents can be carried out.
- the invention is based on the idea to first subject a plurality of images of the document to a wavelet transformation and to derive features from individual wavelet pixels or wavelet pixel regions.
- a wavelet transform By using a wavelet transform, spatial-frequency information of the image is obtained that can still be assigned to a location on the document. This makes it possible, both for a human observer visually perceptible information and more complex features that can not be derived by a pure pattern comparison of the optical raw data, in a simple manner evaluate.
- a method for verifying documents comprising the steps of: capturing at least one image of the document; Deriving characteristics from the figure; Comparing the features with expert knowledge; Verifying the document based on a result of the comparison, wherein it is provided that the image is subjected to a wavelet transformation and the features are derived from individual wavelet pixels or wavelet pixel regions.
- a corresponding device comprises a detection unit for detecting an image of the document as well as an evaluation unit, which derives characteristics on the basis of the acquired image. Furthermore, the device is designed to compare the derived features with expert knowledge and to perform a verification, for example authentication, of the document based on the result of the comparison.
- the evaluation unit comprises a transformation unit which subjects the image to a wavelet transformation and that the features are derived from individual wavelet pixels or wavelet pixel regions.
- a tuple which comprises a coefficient value of the wavelet transformation and the information necessary to assign the coefficient value.
- the necessary information includes, for example, a position and an indication of a spatial direction with respect to which the wavelet transformation has made a spectral decomposition, a frequency range and optionally an indication of which of several recorded images of the same document the coefficient value is to be assigned.
- wavelet coefficient value, location information and membership of a frequency range can be taken from a wavelet pixel.
- a spatial frequency is in each case a measure of a recurrence of a feature, with reference to one, unless otherwise clear from the direct relationship of the use of the term spatial distance, understood. If, for example, dark pixels appear on a otherwise white or light background along a spatial direction in, for example, equidistant distances, a frequency is defined by the distance.
- a frequency range is a range that includes a frequency interval. In the above example, in the frequency range including the frequency at which the dark dots occur, non-zero coefficients occur in a wavelet transform.
- the frequency While in general parlance the frequency indicates a recurrence with respect to time, the frequency here is understood in each case as the recurrence in a spatial distance with respect to a spatial direction. Therefore, it is necessary to specify the corresponding spatial direction in each case.
- a spectral range is understood to be a summary (amount) of the wavelet pixels of a wavelet transformation, which belong to a spatial direction, a local area and a frequency range. With a wavelet transformation, it is possible to generate several spectral ranges for different frequency ranges. Spectral ranges can thus differ with regard to their spatial direction and / or their frequency range.
- the wavelet transformation decomposes at least one of the multiple images into a plurality of spectral regions, each spectral region comprising wavelet pixels for a spatial region corresponding to the spatial region of the image with respect to a spatial direction of the local region of the at least one image and with respect to a frequency region includes.
- each spectral region comprises wavelet pixels that can be assigned to a location of the image.
- the local area of the spectral region corresponds to the local area of the image.
- each location area (not necessarily with each individual pixel) corresponds a location area of the corresponding spectral range comprising the coefficient values obtained by the wavelet transformation.
- An apparatus for implementing the method therefore expediently provides in that the detection device is designed to detect the document under different recording conditions.
- a wavelet transformation can be carried out in a device, for example by means of a program-controlled computing unit.
- circuit arrangement which performs a wavelet transformation on the basis of the mapping raw data.
- each wavelet transform generates for each frequency range three spectral ranges which are assigned according to a spatial direction of a base vector of a coordinate system and a diagonal of the coordinate system of the spatial domain of the image.
- a Cartesian coordinate system which is aligned in a rectangular document so that the coordinate axes are aligned parallel to the long sides of the rectangular document or the image of the document. Since the information is usually or often printed in parallel or along a diagonal, relative to the edges of the rectangular document, such a wavelet transformation is particularly suitable for evaluation.
- Features such as a used character spacing can be determined, for example, particularly simple and accurate. In principle, any other coordinate system can be selected.
- the method is designed so that a wavelet pixel can be represented as a tuple, which comprises a coefficient value and an indication which allow an association with the image to a spectral region of the corresponding image and to a position in the spatial region of the spectral region.
- the wavelet pixels are preferably stored as tuples or vectors in a memory.
- memory constructs are used, which are designed so that they can optimally store the information given, ie as space-saving as possible.
- one or more wavelet pixel areas are used in an evaluation that include wavelet pixels that are not associated with a contiguous location area of a spectral range. This means that when evaluating the wavelet pixels, wavelet pixels that are not arranged in a contiguous location area of a spectral range can be evaluated together. It is likewise possible for several wavelet pixels of different spectral ranges to be combined to form a wavelet pixel range. These wavelet pixel areas do not necessarily correspond to the same image of the document to be verified.
- the multiple images it is possible to detect the multiple images so that different wavelength ranges of the light scattered / reflected and / or emitted on the document are detected.
- light may be emitted from the document due to an excited luminescence.
- different color separations of the document can be captured as images.
- the document is irradiated with monochromatic and / or polychromatic electromagnetic radiation.
- the detection device thus comprises one or more monochromatic light sources and optionally additionally a polychromatic light source.
- the detection device may comprise one or more wavelength-selective detection units that detect a polychromatically illuminated document.
- an excitation with UV radiation takes place before or during a capture of an image.
- other embodiments may provide for excitation in the IR wavelength range.
- the multiple images are acquired in such a way that the document is illuminated and / or excited differently.
- the document is excited / illuminated by IR radiation, visible radiation and / or UV radiation.
- Excitation mechanisms for example in the form of an electric field, etc., may be used to excite electroluminescence, for example.
- the different mappings may be captured with different quality and / or resolution and represent different information contents of the document
- the wavelet transforms for the different mappings of the same document are performed with different wavelet basis functions and / or scalings . It is understood that also images of the same quality and resolution with different wavelet basis functions and / or scaling can be transformed.
- a plurality of wavelet pixels are evaluated in combination, which are assigned to the different images of the same security document and / or different location areas and / or different spectral areas respectively of the same security document are.
- the features can be easily derived from the wavelet pixels or a group of wavelet pixels that are combined in a wavelet pixel area.
- at least one of the features is derived by a statistical evaluation.
- a statistical evaluation here is understood as determining a frequency of individual coefficient values, the presence of coefficient values in a certain coefficient range, but also the determination of more complex statistical values, such as an average of coefficient values, a standard deviation and / or a variance or other variables.
- the derivation of a feature is performed by a comparison of determined statistical properties or property combinations with characteristic statistical default properties or default property combinations characteristic of features, where the feature is considered derived if the determined statistical property or property combination meets a predetermined match with the feature associated typical statistical Has default property or default property combination.
- it can be determined for a wavelet pixel area how many wavelet pixels have a coefficient value within a certain value range.
- a normalization to the total number of considered wavelet pixels is preferably carried out.
- individual features of security documents are correlated to a particular relative proportion of wavelet coefficient values in that range of values. This means that the feature has a default property, namely a relative frequency with which coefficient values in the considered wavelet pixel range occur in the particular value range.
- the individual wavelet pixels can still be assigned to a location area of the original image or the document, the features can be resolved locally, i. assigned to spatial areas of the image of the document.
- the derived features may be, for example, printing methods used. For this purpose, individual local areas or even the entire document can be examined with regard to the use of a specific printing method. It has been found that individual printing methods used to produce a document cause statistical properties of the wavelet coefficients that are characteristic of the particular printing method to occur in certain spectral ranges. Since certain printing processes require a great deal of technical effort, the detection of a particular printing process used in simple storage already provides sufficient certainty to determine whether it is a genuine or forged document. If, on the other hand, a printing method can be identified which is not normally used for security documents, such a feature is an indication that this document is not genuine or at least falsified.
- a relative number of the wavelet pixels of the associated location area of at least one spectral range is determined whose coefficient value is within a given value interval.
- the wavelet pixels or wavelet pixel groups are evaluated for statistical deviations from typical statistical input properties in order to determine such distortions.
- the document is classified and identified by means of a neural network or a so-called support vector machine or other classification method. Since deriving the features from the wavelet pixel areas enables derivation of general features for documents, such as the presence of information printed by certain printing methods, the presence of guilloche printing, a determination of a certain surface roughness, etc., it is possible to do so the identified features successively exclude possible document identifications. While a plurality of pixel areas between the document to be verified and a possible reference document is usually to be compared in the prior art in order to allow reliable identification, a clear identification of a document type can be brought about here by deriving a small number of features.
- a method in which it is provided that after deriving a statistical property which is or can not be assigned a feature in the expert knowledge, a response to the question whether (a) this property is a new authentic Feature of a document or (b) a new forged feature of a document or (c) a new forgery-indicating feature of a document or (d) none of the aforementioned alternatives, and depending on the detected answer in cases (a) to (c), the expert's knowledge is expanded.
- a self-learning verification method or a self-learning verification device can be created.
- a device can be designed accordingly.
- a feature combination of a document is supplied to expert knowledge if this combination of features is unknown and verification information is detected which identifies the associated document as an authentic and / or non-authentic document.
- a plurality of features is preferably derived.
- the expert knowledge comprises information about individual typical features of authenticated documents and / or about typical forged features. These are not directly assigned to individual document types.
- the authentication is based on the presence of individual and / or a combination of features that are identified and that are not explicitly assigned to a document type via the expert knowledge.
- This method and a corresponding device can be used, for example, where expert knowledge about a specific embodiment of individual document types should not be revealed. Compared to an ordinary visual inspection is the Finding non-authentic documents by using such a device or the application of such a method already significantly increased.
- features and / or combinations of features are assigned weightings, and based on the features and feature combinations derived from the weights for a document, an overall weighting is determined that indicates a measure of the presence of an authentic document.
- features that indicate a forgery can be negatively considered.
- Fig. 1 schematically a flowchart of a method for verification of documents is shown. First, several pictures will be inserted and 1. In a further method step, the multiple images are each subjected to a wavelet transformation. 2. Different spectral regions are generated here.
- FIG. 2 is an example of one Figure 3 of a document.
- Fig. 3 are the spectral ranges resulting from the wavelet transformation of the in Fig. 2 shown Figure 3 of the document is shown schematically.
- the wavelet decomposition with respect to three different spatial directions is parallel to the longitudinal edge 5 of the document (spectral region indicated by a) parallel to the short side edge 6 of the document (spectral region indicated by b) and along a diagonal 7 of the document (identification of the spectral region with c).
- coefficient values are determined which are associated with a location on the image of the document, a frequency range, and a spatial direction along which the spectral decomposition is made.
- the coefficient values are shown as gray levels at the respective location.
- a separate location area is shown for each frequency range 4-1, ..., 4-n and each spatial direction, along which the wavelet transform performs the spectral decomposition.
- the beams of the security feature 8 aligned parallel to the longitudinal axis produce high coefficient values in the spectral regions 4-1b, 4. 2b, in which the wavelet decomposition is made parallel to the short side edge. It will be apparent to those skilled in the art that the wavelet pixels can still be locally assigned to the location of the image.
- wavelet pixel areas which are also referred to as wavelet pixel areas, in order to derive a feature.
- the individual wavelet pixels of a wavelet pixel region need not be spatially adjacent or from the same spectral region or from the same image of the document. Rather, wavelet pixels can be arbitrarily combined into a wavelet pixel range as long as they come from images of one and the same document.
- a derivation of features which may be, for example, a printing method, a surface finish, a character size, etc., can most easily be done by a statistical evaluation of individual pixels or of pixel areas.
- This is exemplified by the 4 and 5 explained.
- 4 and 5 show graphs plotting a frequency of coefficient values in a wavelet pixel group. Along the abscissa 22 are the coefficient values and along the ordinate 23 their frequencies normalized to a maximum frequency of a coefficient value plotted. While in Fig. 4 An illustration of a smooth document is evaluated in Fig. 5 an illustration of a rough or roughened due to falsifications document shown. It can be clearly seen that there are characteristic statistical properties that are identical for both documents and also give characteristic properties that are different for the documents.
- the device 41 comprises a detection device 42.
- the detection device 42 comprises a graphic detection unit 43, which may be designed, for example, as a digital camera or as a scanner.
- the detection device 42 preferably comprises one or more illumination devices 44, 45. This makes it possible to expose a document 46 differently and to capture one or more images by means of the graphical detection unit 43.
- the detector 42 may include means for applying the electric field to contacts of the document to trigger, for example, electroluminescence.
- the graphical detection unit 43 is preferably designed such that it can record images with thresholds.
- the detector 42 is designed to be able to capture graphic images of the document 46.
- a plurality of graphical detection units can also be provided.
- the device 41 further comprises a transformation unit 47.
- the transformation unit 47 is designed to subject the data representing the images to a wavelet transformation. This will for each picture Spectral ranges generated, which are preferably stored in a memory 48 for further processing.
- the transformation unit 47 may also be formed by a program-controlled computing unit 55 in some embodiments.
- the device 41 further comprises an evaluation unit 49, which derives features of the document from the wavelet pixel or wavelet pixel regions contained in the generated spectral regions.
- the evaluation unit 49 may be formed by means of one or more electronic circuits. Alternatively, the evaluation unit 49 can be realized by means of the program-controlled computing unit 55 or another arithmetic unit. To derive the features, information stored in the memory unit 48, such as program modules 56, may be retrieved, which may be considered as a representation of expert knowledge for deriving the features.
- the device 41 further comprises a verification unit 50, which in the illustrated embodiment is also realized by means of the computing unit 55. This is designed to perform a classification of the document 46 on the basis of the derived characteristics on the basis of expert information stored in the memory device 48.
- the verification unit consists of an interrogator which checks whether the derived feature matches the expected feature, i.e., whether the feature could be derived.
- multiple features are evaluated in combination to classify the document and assign it to a document type.
- the classification can be done for example by means of a neural network or a support vector machine.
- the expert knowledge about the classification is then at least partially contained in the neural network or support vector machine.
- the information unit, the evaluation unit and / or the verification unit insofar as they are completely or partially realized by means of a program-controlled computer, can be realized by means of one and the same computer or microprocessor.
- the illustrated embodiment further includes an output unit 51 for outputting a verification result.
- a verification result This may, for example, be an indication as to whether the document is genuine, a forgery or falsified.
- the document type may be output as the document has been identified.
- the output unit may be any unit capable of outputting information.
- it may be a display device.
- it may be an interface that performs the output in the form of electronic signals that may be evaluated by other electronic devices and, where appropriate, translated into a form perceptible to a user.
- the device 41 further comprises an input unit 52, via which user inputs can be detected.
- the memory device 48 is part of the device 41.
- at least part of the expert knowledge is stored in an external database.
- the verification device 41 is connected there via a communication network 53 to an external database 54. This contains the necessary expert knowledge to enable a derivation of features and / or information for a classification and / or verification of the document whose images or illustrations are recorded.
- at least part of the classification is performed in a central facility.
- only one detection of the one or more images and in the transformation of the image data by means of a wavelet transformation is performed in the local verification device. Furthermore, the individual features are derived in the verification device. The features or their special characteristics are then transmitted to the central unit. Based on the transmitted feature information then a classification of the document or a verification is made. The verification result is then transmitted back to the verification device and output.
- verification is made based on individual features or feature combinations associated with document types or classes of document types.
- one or more features are derived, which may assume different characteristics, for example different values.
- a confidence value is then formed, which is a measure of whether the document examined is a genuine or a falsified or falsified document.
- the individual derived features may be provided with a weighting factor. Verification by means of this method is particularly suitable when a class of documents has a certain range of variation with respect to individual features. If, for example, a roughness of a documentary material used varies with the passports of a state, the roughness of the document can be determined as a feature.
- identification is not only possible by examining documents with a range of variation in the individual features. Rather, it is also possible to combine features that have different characteristics in this way in order to assign each document a characteristic value, which is for example referred to as a recognition value.
- the recognition values of genuine documents of a document type for example a passport of a nation, will all have a similar value. Based on the recognition values, a classification of the documents is thus possible.
- Such a method is also suitable for constructing a self-learning verification system. If, for example, all documents are examined with respect to a plurality of features, then the determined recognition values can be stored in a memory. At least during a learning phase, it is necessary to have a different way of checking the document, for example, by a reviewer or other verification device, and to capture a verification decision or other verification information for each document. In the simplest case, it only records whether it is a genuine or a forged or falsified document. Thus, it is possible to assign verification information to individual recognition values. Once a sufficiently large number of documents of the same type is captured, documents of this type can be reliably identified as being genuine, falsified or falsified.
- the concrete characteristic values of the derived features are stored for the individual documents recorded in a learning phase. This facilitates differentiated verification analyzes. If not only an authenticity check but also a classification is to be carried out, this is more reliably possible with a large number of possible documents if characteristic feature combinations or characteristic values are stored for the individual document types.
- the same features are first derived for all documents and their characteristics are stored.
- verification information identifying, for example, the document is detected. If a large number of documents of the same document type has been recorded, it is possible to determine in which characteristics or with which characteristic values the documents of a document type match or can be delimited from other document types. In this way, it can be achieved that only a limited number of features have to be derived after the learning phase in order to classify a document.
- a verification of individual features, in particular locally resolved, is performed after the document type has been identified.
- biometric data in the form of a passport photograph or alphanumeric imprints comprising a name, an address, a date of birth, a serial number, etc.
- a document 61 comprises a plurality of local areas 62-1 to 62-10, on which, for example, in each case an alphanumeric character of a machine-readable line is printed.
- wavelet pixels are statistically evaluated from corresponding local regions of the spectral regions in order to derive different features, here for example three features.
- features a, b and c in each case a statistical distribution of a frequency against the feature expression is plotted, which results in an evaluation of all areas 62-1 to 62-10.
- the characteristic values from a range in one or more of the features a to c deviate significantly from the statistical totality.
- the individual feature values for the features a to c are shown schematically as boxes, to which a number between 1 and 10 is assigned, indicating the feature expression of the corresponding area of the document, which has the same number as a trailing number in the reference character. It becomes clear that the characteristic values of the areas 62-4, 62-5 and 62-7 each have a significant deviation from the characteristic values of the remaining areas 61-1 to 61-10 for at least two features. From this it can be deduced that distortions have been made in these areas.
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Inspection Of Paper Currency And Valuable Securities (AREA)
- Image Processing (AREA)
Description
Die Erfindung betrifft ein Verfahren und eine Vorrichtung zum Verifizieren von Dokumenten, bei dem bzw. mit der mehrere Abbildungen des Dokuments erfasst werden, aus den Abbildungen Merkmale abgeleitet werden, die Merkmale mit Expertenwissen verglichen werden und ein Verifizieren, beispielsweise Authentifizieren, eines Dokuments anhand eines Ergebnisses des Vergleichs vorgenommen wird. Ein Aspekt des Verfahrens betrifft die Ableitung der Merkmale.The invention relates to a method and a device for verifying documents, in which a plurality of images of the document are detected, characteristics are derived from the images, the features are compared with expert knowledge, and verification, for example, authentication of a document by means of a Result of the comparison. One aspect of the method relates to the derivation of the features.
Aus dem Stand der Technik ist es bekannt, Dokumente, welche beispielsweise Sicherheitsdokumente, wie insbesondere Reisepässe, Personalausweise, Identifikationskarten, Visa, Führerscheine, usw., oder Wertdokumente, wie insbesondere Banknoten, Schecks, Kreditkarten, Postwertzeichen usw., umfassen können, hinsichtlich ihrer Echtheit und Unverfälschtheit zu überprüfen. Eine solche Überprüfung wird als Verifikation oder Verifizierung bezeichnet. Als Verifikation wird hier auch das Identifizieren eines Dokuments als ein Dokumenttyp verstanden. Dies bedeutet, dass eine Erkennung einer Banknote als ein Dokument eines bestimmten Werts einer Ausstellungsbehörde auch als Verifikation verstanden wird.From the prior art it is known documents, which may include, for example, security documents, such as in particular passports, identity cards, identification cards, visas, driver's licenses, etc., or documents of value, in particular banknotes, checks, credit cards, postage stamps, etc., with regard to their To check authenticity and authenticity. Such verification is called verification or verification. Verification also means identifying a document as a document type. This means that recognition of a banknote as a document of a specific value of an issuing authority is also understood as verification.
Allgemein werden zur Verifizierung Merkmale des Dokuments analysiert und mit einem Expertenwissen verglichen, welches Angaben über die einzelnen Merkmale und deren Ausgestaltung und/oder über Merkmalskombinationen eines Dokuments umfasst, um hierüber eine Verifikation anhand beispielsweise einer Merkmalskombination vorzunehmen.In general, features of the document are analyzed for verification and compared with expert knowledge, which includes information about the individual features and their design and / or about feature combinations of a document in order to perform a verification on the basis of, for example, a combination of features.
Aus der
Aus der
Die bekannten Verfahren basieren im Wesentlichen auf einem direkten Mustervergleich einer erfassten Abbildung des Dokuments. Hierbei werden jeweils vorzugsweise nur begrenzte Abbildungsbereiche miteinander verglichen, um nur statische Bereiche, d.h. solche Bereiche, die für alle Dokumente eines Typs, beispielsweise für alle Reisepässe einer Ausstellungsnation, identisch sind, miteinander verglichen werden. Da unterschiedliche Dokumenttypen solche statischen Informationen an unterschiedlichen Stellen aufweisen, müssen jeweils unterschiedliche Segmente oder Bereiche für einen Vergleich ausgewertet werden. Erst wenn das Dokument identifiziert ist, beispielsweise als Reisepass einer bestimmten Nation, findet eine nachgelagerte Überprüfung von Sicherheitsmerkmalen statt, anhand derer eine Echtheit und Unverfälschtheit der Dokumente überprüft wird.The known methods are essentially based on a direct pattern comparison of a captured image of the document. In this case, preferably only limited imaging areas are compared with each other in order to avoid only static areas, i. such areas, which are identical for all documents of one type, for example, for all passports of an issuing nation, are compared. Since different types of documents have such static information at different locations, different segments or areas must be evaluated for comparison. Only when the document is identified, for example, as a passport of a particular nation, does a subsequent review of security features take place, with the aid of which a genuineness and authenticity of the documents is checked.
Die bekannten Verfahren beruhen somit auf einer mehrstufigen Prüfungsprozedur, bei der es erforderlich ist, zunächst das Dokument zu identifizieren und anschließend für das identifizierte Dokument bekannte Sicherheitsmerkmale zu vergleichen. Verfälschungen, beispielsweise eine Änderung von aufgedruckten Personaldaten, auf einem Originalsicherheitsdokument können mit den bekannten Verfahren nur schwer oder gar nicht aufgespürt werden.The known methods are thus based on a multi-stage examination procedure, in which it is necessary to first identify the document and then to compare known security features for the identified document. Falsifications, for example a change of printed personal data, on an original security document can be detected with the known methods only with difficulty or not at all.
Aus der
Aus der
Der Erfindung liegt das technische Problem zugrunde, ein Verfahren und eine Vorrichtung zur Verifikation von Dokumenten, insbesondere Sicherheits- und/oder Wertdokumenten, zu schaffen, mit denen eine schnelle, einfache und zuverlässige Verifikation von Dokumenten durchgeführt werden kann.The invention is based on the technical problem of providing a method and a device for verifying documents, in particular security and / or value documents, with which a quick, simple and reliable verification of documents can be carried out.
Die Erfindung wird durch ein Verfahren mit den Merkmalen des Anspruchs 1 und eine Vorrichtung mit den Merkmalen des Anspruchs 15 gelöst. Vorteilhafte Ausgestaltungen ergeben sich aus den Unteransprüchen.The invention is solved by a method having the features of
Der Erfindung liegt der Gedanke zugrunde, mehrere Abbildungen des Dokuments jeweils zunächst einer Wavelet-Transformation zu unterziehen und Merkmale aus einzelnen Wavelet-Pixeln oder Wavelet-Pixelbereichen abzuleiten. Durch die Verwendung einer Wavelet-Transformation werden Ortsfrequenzinformationen der Abbildung gewonnen, die immer noch einem Ort auf dem Dokument zugeordnet werden können. Hierdurch ist es möglich, sowohl für einen menschlichen Betrachter optisch wahrnehmbare Informationen als auch komplexere Merkmale, die sich durch einen reinen Mustervergleich der optischen Rohdaten nicht ableiten lassen, auf einfache Weise auszuwerten. Insbesondere wird ein Verfahren zum Verifizieren von Dokumenten vorgeschlagen, umfassend die Schritte: Erfassen mindestens einer Abbildung des Dokuments; Ableiten von Merkmalen aus der Abbildung; Vergleichen der Merkmale mit Expertenwissen; Verifizieren des Dokuments anhand eines Ergebnisses des Vergleichs, wobei vorgesehen ist, dass die Abbildung einer Wavelet-Transformation unterzogen wird und die Merkmale aus einzelnen Wavelet-Pixeln oder Wavelet-Pixelbereichen abgeleitet werden. Eine entsprechende Vorrichtung umfasst eine Erfassungseinheit zum Erfassen einer Abbildung des Dokuments sowie eine Auswerteeinheit, die anhand der erfassten Abbildung Merkmale ableitet. Ferner ist die Vorrichtung ausgestaltet, die abgeleiteten Merkmale mit Expertenwissen zu vergleichen und anhand des Ergebnisses des Vergleichs eine Verifizierung, beispielsweise Authentifizierung, des Dokuments vorzunehmen. Hierbei kann es zum Beispiel zur Prüfung einzelner Merkmale auf einem Dokument (Verifizierung, 1:1) oder aber zur Suche dieser Merkmale in mehreren Dokumenten kommen (Authentifizierung, 1:n, wobei n die Anzahl der Dokumente ist). Erfindungsgemäß ist vorgesehen, dass die Auswerteeinheit eine Transformationseinheit umfasst, die die Abbildung einer Wavelet-Transformation unterzieht und dass die Merkmale aus einzelnen Wavelet-Pixeln oder Wavelet-Pixelbereichen abgeleitet sind. Insgesamt wird hierdurch ein Rechenaufwand deutlich reduziert, insbesondere wenn komplexere Merkmale hinsichtlich ihrer Existenz überprüft werden sollen.The invention is based on the idea to first subject a plurality of images of the document to a wavelet transformation and to derive features from individual wavelet pixels or wavelet pixel regions. By using a wavelet transform, spatial-frequency information of the image is obtained that can still be assigned to a location on the document. This makes it possible, both for a human observer visually perceptible information and more complex features that can not be derived by a pure pattern comparison of the optical raw data, in a simple manner evaluate. In particular, a method for verifying documents is proposed, comprising the steps of: capturing at least one image of the document; Deriving characteristics from the figure; Comparing the features with expert knowledge; Verifying the document based on a result of the comparison, wherein it is provided that the image is subjected to a wavelet transformation and the features are derived from individual wavelet pixels or wavelet pixel regions. A corresponding device comprises a detection unit for detecting an image of the document as well as an evaluation unit, which derives characteristics on the basis of the acquired image. Furthermore, the device is designed to compare the derived features with expert knowledge and to perform a verification, for example authentication, of the document based on the result of the comparison. This can be done, for example, to check individual features on a document (verification, 1: 1) or to search for these features in several documents (authentication, 1: n, where n is the number of documents). According to the invention, it is provided that the evaluation unit comprises a transformation unit which subjects the image to a wavelet transformation and that the features are derived from individual wavelet pixels or wavelet pixel regions. Overall, a computational effort is significantly reduced, especially when more complex features are to be checked for their existence.
Als ein Wavelet-Pixel wird ein Tupel angesehen, welches einen Koeffizientenwert der Wavelet-Transformation sowie die notwendigen Angaben umfasst, um den Koeffizientenwert zuzuordnen. Die notwendigen Angaben umfassen beispielsweise eine Position sowie eine Angabe einer Raumrichtung, bezüglich derer die Wavelet-Transformation eine spektrale Zerlegung vorgenommen hat, einen Frequenzbereich und gegebenenfalls eine Angabe, welcher von mehreren aufgenommenen Abbildungen desselben Dokuments der Koeffizientenwert zuzuordnen ist. Mit anderen Worten: Aus einem Wavlet-Pixel können Waveletkoeffizientenwert, Ortsinformation und Zugehörigkeit zu einem Frequenzbereich entnommen werden.As a wavelet pixel, a tuple is considered, which comprises a coefficient value of the wavelet transformation and the information necessary to assign the coefficient value. The necessary information includes, for example, a position and an indication of a spatial direction with respect to which the wavelet transformation has made a spectral decomposition, a frequency range and optionally an indication of which of several recorded images of the same document the coefficient value is to be assigned. In other words, wavelet coefficient value, location information and membership of a frequency range can be taken from a wavelet pixel.
Unter einer räumlichen Frequenz wird im Zusammenhang dieser Anmeldung, sofern sich nichts anderes aus dem unmittelbaren Zusammenhang der Verwendung des Begriffes ergibt, jeweils ein Maß für eine Wiederkehr eines Merkmals, bezogen auf einen räumlichen Abstand, verstanden. Treten beispielsweise entlang einer Raumrichtung in zum Beispiel äquidistanten Abständen dunkle Pixel auf einem ansonsten weißen oder hellen Hintergrund auf, so ist über den Abstand eine Frequenz festgelegt. Ein Frequenzbereich ist ein Bereich, der ein Frequenzintervall umfasst. Bei dem genannten Beispiel treten in dem Frequenzbereich, der die Frequenz umfasst, mit der die dunklen Punkte auftreten, von Null verschiedene Koeffizienten bei einer Wavelet-Transformation auf. Während im allgemeinen Sprachgebrauch die Frequenz eine Wiederkehr bezogen auf die Zeit angibt, wird hier die Frequenz jeweils auf die Wiederkehr in einem räumlichen Abstand, bezogen auf eine Raumrichtung, verstanden. Daher ist es notwendig, jeweils auch die entsprechende Raumrichtung anzugeben.In the context of this application, a spatial frequency is in each case a measure of a recurrence of a feature, with reference to one, unless otherwise clear from the direct relationship of the use of the term spatial distance, understood. If, for example, dark pixels appear on a otherwise white or light background along a spatial direction in, for example, equidistant distances, a frequency is defined by the distance. A frequency range is a range that includes a frequency interval. In the above example, in the frequency range including the frequency at which the dark dots occur, non-zero coefficients occur in a wavelet transform. While in general parlance the frequency indicates a recurrence with respect to time, the frequency here is understood in each case as the recurrence in a spatial distance with respect to a spatial direction. Therefore, it is necessary to specify the corresponding spatial direction in each case.
Unter einem Spektralbereich wird eine Zusammenfassung (Menge) der Wavelet-Pixel einer Wavelet-Transformation verstanden, die zu einer Raumrichtung, einem Ortsbereich und einem Frequenzbereich gehören. Mit einer Wavelet-Transformation ist es möglich, mehrere Spektralbereiche für unterschiedliche Frequenzbereiche zu erzeugen. Spektralbereiche können sich somit hinsichtlich ihrer Raumrichtung und/oder ihres Frequenzbereichs unterscheiden.A spectral range is understood to be a summary (amount) of the wavelet pixels of a wavelet transformation, which belong to a spatial direction, a local area and a frequency range. With a wavelet transformation, it is possible to generate several spectral ranges for different frequency ranges. Spectral ranges can thus differ with regard to their spatial direction and / or their frequency range.
Bei einer bevorzugten Ausführungsform ist vorgesehen, dass die Wavelet-Transformation mindestens eine der mehreren Abbildung in mehrere Spektralbereiche zerlegt, wobei jeder Spektralbereich Wavelet-Pixel für einen mit dem Ortsbereich der Abbildung korrespondierenden Ortsbereich bezüglich einer Raumrichtung des Ortsbereichs der mindestens einen Abbildung und bezüglich eines Frequenzbereichs umfasst. Dies bedeutet, dass jeder Spektralbereich Wavelet-Pixel umfasst, die einem Ort der Abbildung zugeordnet werden können. Der Ortsbereich des Spektralbereichs korrespondiert hierbei mit dem Ortsbereich der Abbildung. Somit gibt es in jedem Spektralbereich Wavelet-Pixel, die einem korrespondierenden Ortsbereich der Abbildung zugeordnet sind. Dies bedeutet auch, dass mit jedem Ortsbereich (nicht notwendigerweise mit jedem einzelnen Pixel) ein Ortsbereich des entsprechenden Spektralbereichs korrespondiert, der durch die Wavelet-Transformation gewonnene Koeffizientenwerte umfasst. Erfindungsgemäß ist vorgesehen, mehrere Abbildungen desselben Dokuments unter verschiedenen Aufnahmebedingungen zu erfassen, die jeweils einer Wavelet-Transformation unterzogen werden. Eine Vorrichtung zur Umsetzung des Verfahrens sieht daher zweckmäßigerweise vor, dass die Erfassungseinrichtung ausgebildet ist, das Dokument unter verschiedenen Aufnahmebedingungen zu erfassen.In a preferred embodiment, it is provided that the wavelet transformation decomposes at least one of the multiple images into a plurality of spectral regions, each spectral region comprising wavelet pixels for a spatial region corresponding to the spatial region of the image with respect to a spatial direction of the local region of the at least one image and with respect to a frequency region includes. This means that each spectral region comprises wavelet pixels that can be assigned to a location of the image. The local area of the spectral region corresponds to the local area of the image. Thus, there are wavelet pixels in each spectral range associated with a corresponding location region of the image. This also means that with each location area (not necessarily with each individual pixel) corresponds a location area of the corresponding spectral range comprising the coefficient values obtained by the wavelet transformation. According to the invention, it is provided to acquire several images of the same document under different recording conditions, which are each subjected to a wavelet transformation. An apparatus for implementing the method therefore expediently provides in that the detection device is designed to detect the document under different recording conditions.
Eine Wavelet-Transformation kann in einer Vorrichtung beispielsweise mittels einer programmgesteuerten Recheneinheit ausgeführt werden.A wavelet transformation can be carried out in a device, for example by means of a program-controlled computing unit.
Andere Ausführungsformen sehen vor, dass eine Schaltungsanordnung vorgesehen ist, die eine Wavelet-Transformation anhand der Abbildungsrohdaten ausführt.Other embodiments provide that a circuit arrangement is provided which performs a wavelet transformation on the basis of the mapping raw data.
Es ist vorgesehen, dass jede Wavelet-Transformation für jeden Frequenzbereich drei Spektralbereiche erzeugt, die entsprechend einer Raumrichtung eines Basisvektors eines Koordinatensystems und einer Diagonalen des Koordinatensystems des Ortsbereichs der Abbildung zugeordnet sind. In der Regel bietet sich die Verwendung eines kartesischen Koordinatensystems an, welches bei einem rechteckigen Dokument so ausgerichtet ist, dass die Koordinatenachsen parallel zu den Längsseiten des rechteckigen Dokuments bzw. der Abbildung des Dokuments ausgerichtet sind. Da die Informationen in der Regel oder häufig parallel oder entlang einer Diagonalen, bezogen auf die Kanten des rechteckigen Dokuments, aufgedruckt sind, eignet sich eine solche Wavelet-Transformation besonders für eine Auswertung. Merkmale wie beispielsweise ein verwendeter Zeichenabstand lassen sich beispielsweise besonders einfach und genau bestimmen. Grundsätzlich kann jedes andere Koordinatensystem gewählt werden.It is provided that each wavelet transform generates for each frequency range three spectral ranges which are assigned according to a spatial direction of a base vector of a coordinate system and a diagonal of the coordinate system of the spatial domain of the image. In general, the use of a Cartesian coordinate system, which is aligned in a rectangular document so that the coordinate axes are aligned parallel to the long sides of the rectangular document or the image of the document. Since the information is usually or often printed in parallel or along a diagonal, relative to the edges of the rectangular document, such a wavelet transformation is particularly suitable for evaluation. Features such as a used character spacing can be determined, for example, particularly simple and accurate. In principle, any other coordinate system can be selected.
Vorzugsweise ist das Verfahren so ausgebildet, dass ein Wavelet-Pixel als ein Tupel darstellbar ist, welches einen Koeffizientenwert sowie eine Angabe umfasst, die eine Zuordnung zu der Abbildung zu einem Spektralbereich der entsprechenden Abbildung und zu einer Position in dem Ortsbereich des Spektralbereichs ermöglichen. In einer Vorrichtung zur Umsetzung des Verfahrens werden die Wavelet-Pixel vorzugsweise als Tupel oder Vektoren in einem Speicher abgelegt. Hierzu werden Speicherkonstrukte verwendet, die so ausgebildet sind, dass sie die genannten Angaben optimal, d.h. möglichst platzsparend, speichern können.Preferably, the method is designed so that a wavelet pixel can be represented as a tuple, which comprises a coefficient value and an indication which allow an association with the image to a spectral region of the corresponding image and to a position in the spatial region of the spectral region. In an apparatus for implementing the method, the wavelet pixels are preferably stored as tuples or vectors in a memory. For this purpose, memory constructs are used, which are designed so that they can optimally store the information given, ie as space-saving as possible.
Bei einer Ausführungsform werden bei einer Auswertung ein oder mehrere Wavelet-Pixelbereiche verwendet, der oder die Wavelet-Pixel umfassen, die nicht einem zusammenhängenden Ortsbereich eines Spektralbereichs zugeordnet sind. Dies bedeutet, dass beim Auswerten der Wavelet-Pixel Wavelet-Pixel gemeinsam ausgewertet werden können, die nicht in einem zusammenhängenden Ortsbereich eines Spektralbereichs angeordnet sind. Ebenso ist es möglich, dass mehrere Wavelet-Pixel unterschiedlicher Spektralbereiche zu einem Wavelet-Pixelbereich zusammengefasst werden. Diese Wavelet-Pixelbereiche müssen nicht notwendigerweise mit derselben Abbildung des zu verifizierenden Dokuments korrespondieren.In one embodiment, one or more wavelet pixel areas are used in an evaluation that include wavelet pixels that are not associated with a contiguous location area of a spectral range. This means that when evaluating the wavelet pixels, wavelet pixels that are not arranged in a contiguous location area of a spectral range can be evaluated together. It is likewise possible for several wavelet pixels of different spectral ranges to be combined to form a wavelet pixel range. These wavelet pixel areas do not necessarily correspond to the same image of the document to be verified.
Beispielsweise ist es bei einer Ausführungsform möglich, die mehreren Abbildungen so zu erfassen, dass unterschiedliche Wellenlängenbereiche des an dem Dokument gestreuten/reflektierten und/oder emittierten Lichts erfasst werden. Licht kann beispielsweise aufgrund einer angeregten Lumineszenz von dem Dokument emittiert werden. So können beispielsweise unterschiedliche Farbauszüge des Dokuments als Abbildungen erfasst werden. Um dies zu erreichen, ist bei einer Ausführungsform vorgesehen, dass das Dokument mit monochromatischer und/oder polychromatischer elektromagnetischer Strahlung bestrahlt wird. Um unterschiedliche Farbauszüge als Abbildungen desselben Dokuments zu erhalten, kann das Dokument beispielsweise zeitlich nacheinander mit monochromatischem Licht unterschiedlicher Wellenlänge belichtet werden und jeweils eine Abbildung erfasst werden. Bei einer Ausführungsform umfasst die Erfassungsvorrichtung somit eine oder mehrere monochromatische Lichtquellen und gegebenenfalls zusätzlich eine polychromatische Lichtquelle. Alternativ kann die Erfassungsvorrichtung eine oder mehrere wellenlängenselektive Erfassungseinheit umfassen, die ein polychromatisch beleuchtetes Dokument erfassen.For example, in one embodiment it is possible to detect the multiple images so that different wavelength ranges of the light scattered / reflected and / or emitted on the document are detected. For example, light may be emitted from the document due to an excited luminescence. For example, different color separations of the document can be captured as images. To achieve this, it is provided in one embodiment that the document is irradiated with monochromatic and / or polychromatic electromagnetic radiation. For example, in order to obtain different color separations as images of the same document, the document can be exposed one after the other with monochromatic light of different wavelengths and one image each captured. In one embodiment, the detection device thus comprises one or more monochromatic light sources and optionally additionally a polychromatic light source. Alternatively, the detection device may comprise one or more wavelength-selective detection units that detect a polychromatically illuminated document.
Da es üblich ist, in Sicherheitsdokumente auch Farbmittel zu integrieren, welche bei geeigneter Anregung selbständig Licht emittieren, ist bei einigen Ausführungsformen vorgesehen, dass beispielsweise eine Anregung mit UV-Strahlung vor- oder während eines Erfassens einer Abbildung erfolgt. Ebenso können andere Ausführungsformen eine Anregung im IR-Wellenlängenbereich vorsehen. Eine Ausführungsform sieht somit vor, dass die mehreren Abbildungen so erfasst werden, dass das Dokument unterschiedlich beleuchtet und/oder angeregt wird. Insbesondere ist es vorgesehen, dass das Dokument mit IR-Strahlung, sichtbarer Strahlung und/oder UV-Strahlung angeregt/beleuchtet wird. Ebenso sind jedoch Ausführungsformen möglich, bei denen andere Anregungsmechanismen, beispielsweise in Form eines elektrischen Felds usw., verwendet werden, um beispielsweise eine Elektrolumineszenz anzuregen.Since it is customary to integrate in security documents also colorants which emit light independently with suitable excitation, it is provided in some embodiments that, for example, an excitation with UV radiation takes place before or during a capture of an image. Likewise, other embodiments may provide for excitation in the IR wavelength range. One embodiment thus provides that the multiple images are acquired in such a way that the document is illuminated and / or excited differently. In particular, it is provided that the document is excited / illuminated by IR radiation, visible radiation and / or UV radiation. However, embodiments are possible in which others Excitation mechanisms, for example in the form of an electric field, etc., may be used to excite electroluminescence, for example.
Da die unterschiedlichen Abbildungen möglicherweise mit unterschiedlicher Qualität und/oder Auflösung aufgenommen werden und unterschiedliche Informationsgehalte des Dokuments repräsentieren, ist es bei einer Ausführungsform vorgesehen, dass die Wavelet-Transformationen für die unterschiedliche Abbildungen desselben Dokuments mit unterschiedlichen Wavelet-Basisfunktionen und/oder Skalierungen ausgeführt werden. Es versteht sich, dass auch Abbildungen gleicher Qualität und Auflösung mit unterschiedlichen Wavelet-Basisfunktionen und/oder Skalierungen transformiert werden können.Since the different mappings may be captured with different quality and / or resolution and represent different information contents of the document, in one embodiment it is provided that the wavelet transforms for the different mappings of the same document are performed with different wavelet basis functions and / or scalings , It is understood that also images of the same quality and resolution with different wavelet basis functions and / or scaling can be transformed.
Um komplexe Merkmale zum Verifizieren von Sicherheitsdokumenten zu erhalten, ist bei einer Ausführungsform vorgesehen, dass zum Ableiten von Merkmalen mehrere Wavelet-Pixel in Kombination ausgewertet werden, die den unterschiedlichen Abbildungen desselben Sicherheitsdokuments und/oder unterschiedlichen Ortsbereichen und/oder unterschiedlichen Spektralbereichen jeweils desselben Sicherheitsdokuments zugeordnet sind.In order to obtain complex features for verifying security documents, it is provided in one embodiment that for deriving features, a plurality of wavelet pixels are evaluated in combination, which are assigned to the different images of the same security document and / or different location areas and / or different spectral areas respectively of the same security document are.
Die Merkmale lassen sich anhand der Wavelet-Pixel oder einer Gruppe von Wavelet-Pixeln, die in einem Wavelet-Pixelbereich zusammengefasst sind bzw. werden, einfach ableiten. Bei einer bevorzugten Ausführungsform wird mindestens eines der Merkmale durch eine statistische Auswertung abgeleitet. Unter einer statistischen Auswertung wird hier ein Ermitteln einer Häufigkeit einzelner Koeffizientenwerte, das Vorliegen von Koeffizientenwerten in einem bestimmten Koeffizientenbereich, aber auch das Ermitteln komplexerer statistischer Größen, wie beispielsweise eines Mittelwerts von Koeffizientenwerten, einer Standardabweichung und/oder einer Varianz oder anderer Größen, verstanden.The features can be easily derived from the wavelet pixels or a group of wavelet pixels that are combined in a wavelet pixel area. In a preferred embodiment, at least one of the features is derived by a statistical evaluation. A statistical evaluation here is understood as determining a frequency of individual coefficient values, the presence of coefficient values in a certain coefficient range, but also the determination of more complex statistical values, such as an average of coefficient values, a standard deviation and / or a variance or other variables.
Bei einer Ausführungsform ist vorgesehen, dass das Ableiten eines Merkmals durch einen Vergleich ermittelter statistischer Eigenschaften oder Eigenschaftskombinationen mit für Merkmale typischen statistischen Vorgabeeigenschaften oder Vorgabeeigenschaftskombinationen erfolgt, wobei das Merkmal als abgeleitet gilt, wenn die ermittelte statistische Eigenschaft oder Eigenschaftskombination eine vorgegebene Übereinstimmung mit einer dem Merkmal zugeordneten typischen statistischen Vorgabeeigenschaft bzw. Vorgabeeigenschaftskombination aufweist. Beispielsweise kann für einen Wavelet-Pixelbereich ermittelt werden, wie viele Wavelet-Pixel einen Koeffizientenwert innerhalb eines bestimmten Wertebereichs aufweisen. Hierbei wird vorzugsweise eine Normierung auf die Gesamtzahl der betrachteten Wavelet-Pixel vorgenommen. Einzelne Merkmale von Sicherheitsdokumenten sind beispielsweise mit einem bestimmten relativen Anteil von Wavelet-Koeffizientenwerten in diesem Wertebereich korreliert. Dies bedeutet, dass das Merkmal eine Vorgabeeigenschaft aufweist, nämlich eine relative Häufigkeit, mit der in dem betrachteten Wavelet-Pixelbereich Koeffizientenwerte in dem bestimmten Wertebereich auftreten.In one embodiment it is provided that the derivation of a feature is performed by a comparison of determined statistical properties or property combinations with characteristic statistical default properties or default property combinations characteristic of features, where the feature is considered derived if the determined statistical property or property combination meets a predetermined match with the feature associated typical statistical Has default property or default property combination. For example, it can be determined for a wavelet pixel area how many wavelet pixels have a coefficient value within a certain value range. In this case, a normalization to the total number of considered wavelet pixels is preferably carried out. For example, individual features of security documents are correlated to a particular relative proportion of wavelet coefficient values in that range of values. This means that the feature has a default property, namely a relative frequency with which coefficient values in the considered wavelet pixel range occur in the particular value range.
Da bei einer Wavelet-Transformation die einzelnen Wavelet-Pixel immer noch einem Ortsbereich der ursprünglichen Abbildung bzw. des Dokuments zugeordnet werden können, können die Merkmale lokal aufgelöst, d.h. zugeordnet zu Ortsbereichen der Abbildung des Dokuments, ermittelt werden.Since, in a wavelet transformation, the individual wavelet pixels can still be assigned to a location area of the original image or the document, the features can be resolved locally, i. assigned to spatial areas of the image of the document.
Bei den abgeleiteten Merkmalen kann es sich beispielsweise um verwendete Druckverfahren handeln. Hierfür können einzelne Ortsbereiche oder auch das gesamte Dokument hinsichtlich einer Verwendung eines bestimmten Druckverfahrens untersucht werden. Es wurde herausgefunden, dass einzelne Druckverfahren, die zur Herstellung eines Dokuments verwendet sind, dazu führen, dass in bestimmten Spektralbereichen statistische Eigenschaften der Wavelet-Koeffizienten auftreten, die charakteristisch für das jeweilige Druckverfahren sind. Da bestimmte Druckverfahren einen hohen technischen Aufwand erfordern, liefert das Nachweisen eines bestimmten verwendeten Druckverfahrens in einfach gelagerten Fällen bereits eine ausreichende Sicherheit, um festzustellen, ob es sich um ein echtes oder gefälschtes Dokument handelt. Kann hingegen ein Druckverfahren identifiziert werden, welches für Sicherheitsdokumente in der Regel nicht verwendet wird, so stellt ein solches Merkmal ein Indiz dafür dar, dass dieses Dokument nicht echt oder zumindest verfälscht ist. Bei einer Ausführungsform ist somit vorgesehen, dass zum Ermitteln der statistischen Eigenschaften zum Ableiten eines Merkmals, welches ein Druckverfahren angibt, mit dem ein Ortsbereich des Dokuments bedruckt ist, eine relative Anzahl der Wavelet-Pixel des zugehörigen Ortsbereichs mindestens eines Spektralbereichs ermittelt wird, deren Koeffizientenwert in einem vorgegebenen Werteintervall liegt.The derived features may be, for example, printing methods used. For this purpose, individual local areas or even the entire document can be examined with regard to the use of a specific printing method. It has been found that individual printing methods used to produce a document cause statistical properties of the wavelet coefficients that are characteristic of the particular printing method to occur in certain spectral ranges. Since certain printing processes require a great deal of technical effort, the detection of a particular printing process used in simple storage already provides sufficient certainty to determine whether it is a genuine or forged document. If, on the other hand, a printing method can be identified which is not normally used for security documents, such a feature is an indication that this document is not genuine or at least falsified. In one embodiment, it is thus provided that for determining the statistical properties for deriving a feature which indicates a printing method with which a location area of the document is printed, a relative number of the wavelet pixels of the associated location area of at least one spectral range is determined whose coefficient value is within a given value interval.
Auf einfache Weise lassen sich auch Verfälschungen an Sicherheitsdokumenten nachweisen, zu deren Ausführung eine Rasur ausgeführt wurde. Hierzu ist es vorteilhaft, dass die Wavelet-Pixel oder Wavelet-Pixelgruppen auf statistische Abweichungen gegenüber typischen statistischen Voreingabeeigenschaften ausgewertet werden, um solche Verfälschungen zu ermitteln.In a simple way, it is also possible to detect adulterations on security documents for which a shaving was carried out. For this purpose, it is advantageous that the wavelet pixels or wavelet pixel groups are evaluated for statistical deviations from typical statistical input properties in order to determine such distortions.
Weitere einfach abzuleitende Merkmale umfassen: erkannte Kanten von gedruckten Zeichen oder Symbolen oder eine Erkennung einer Oberflächenbeschaffenheit, z.B. Relief, Struktur. Werden Wavelet-Pixelbereiche in Umgebungen von erkannten Kanten von gedruckten Zeichen oder Symbolen statistisch ausgewertet, um eine Oberflächenrauigkeit auf statische Abweichungen hin zu untersuchen, können beispielsweise wie oben erwähnt Rasuren erkannt werden, die eine Oberflächenbeschaffenheit des Dokuments charakteristisch verändern.Other readily derivable features include: recognized edges of printed characters or symbols, or surface texture recognition, e.g. Relief, structure. If wavelet pixel regions in environments of detected edges of printed characters or symbols are statistically evaluated to investigate surface roughness for static deviations, for example, as mentioned above, shaves that characteristically change a surface finish of the document may be recognized.
Anhand der abgeleiteten Merkmale findet bei einigen Ausführungsformen eine Klassifizierung und Identifizierung des Dokuments mittels eines neuronalen Netzwerks oder einer sogenannten Support-Vektor-Maschine oder eines anderen Klassifikationsverfahrens statt. Da ein Ableiten der Merkmale aus den Wavelet-Pixeln bzw. Wavelet-Pixelbereichen ein Ableiten allgemeiner Merkmale für Dokumente ermöglicht, beispielsweise das Vorhandensein mittels bestimmter Druckverfahren aufgedruckter Informationen, ein Vorhandensein eines Guillochendrucks, eine Bestimmung einer bestimmten Oberflächenrauigkeit usw., ist es möglich, anhand der ermittelten Merkmale sukzessive mögliche Dokumentidentifikationen auszuschließen. Während im Stand der Technik meist eine Vielzahl von Pixelbereichen zwischen dem zu verifizierenden Dokument und einem möglichen Referenzdokument zu vergleichen ist, um eine zuverlässige Identifikation zu ermöglichen, kann hier anhand eines Ableitens einer geringen Anzahl von Merkmalen eine eindeutige Identifikation eines Dokumenttyps herbeigeführt werden.Based on the derived features, in some embodiments, the document is classified and identified by means of a neural network or a so-called support vector machine or other classification method. Since deriving the features from the wavelet pixel areas enables derivation of general features for documents, such as the presence of information printed by certain printing methods, the presence of guilloche printing, a determination of a certain surface roughness, etc., it is possible to do so the identified features successively exclude possible document identifications. While a plurality of pixel areas between the document to be verified and a possible reference document is usually to be compared in the prior art in order to allow reliable identification, a clear identification of a document type can be brought about here by deriving a small number of features.
Bei einer Ausführungsform der Erfindung wird ein Verfahren geschaffen, bei denen vorgesehen ist, dass nach einem Ableiten einer statistischen Eigenschaft, welcher kein Merkmal in dem Expertenwissen zugeordnet ist oder werden kann, eine Antwort auf die Frage, ob (a) diese Eigenschaft ein neues authentisches Merkmal eines Dokuments oder (b) ein neues verfälschtes Merkmal eines Dokuments oder (c) ein neues eine Fälschung anzeigendes Merkmal eines Dokuments oder (d) keine
der vorgenannten Alternativen darstellt, erfasst wird und abhängig von der erfassten Antwort in den Fällen (a) bis (c) das Expertenwissen erweitert wird. Hierdurch kann ein selbstlernendes Verifikationsverfahren bzw. eine selbstlernende Verifikationsvorrichtung geschaffen werden. Eine Vorrichtung kann entsprechend ausgebildet werden.In one embodiment of the invention, a method is provided in which it is provided that after deriving a statistical property which is or can not be assigned a feature in the expert knowledge, a response to the question whether (a) this property is a new authentic Feature of a document or (b) a new forged feature of a document or (c) a new forgery-indicating feature of a document or (d) none
of the aforementioned alternatives, and depending on the detected answer in cases (a) to (c), the expert's knowledge is expanded. As a result, a self-learning verification method or a self-learning verification device can be created. A device can be designed accordingly.
Neben einem Abprüfen auf charakteristische Merkmale einzelner authentischer Dokumente ist es ebenso möglich, charakteristische Merkmale im Expertenwissen abzulegen, die eine Fälschung und/oder Verfälschung anzeigen. Hierdurch ist eine Prüfung von unbekannten Dokumenten im Hinblick darauf möglich, ob typische Fälschungsmerkmale vorhanden sind.In addition to checking for characteristic features of individual authentic documents, it is also possible to store characteristic features in expert knowledge that indicate a forgery and / or falsification. This makes it possible to check unknown documents with regard to whether typical counterfeiting features are present.
Bei einer Weiterbildung ist vorgesehen, dass eine Merkmalskombination eines Dokuments dem Expertenwissen zugeführt wird, wenn diese Merkmalskombination unbekannt ist und eine Verifikationsinformation erfasst wird, die das zugehörige Dokument als authentisches und/oder nicht authentisches Dokument identifiziert. Bei einem gemäß dem Expertenwissen nicht zu identifizierenden Dokumententyp wird bevorzugt eine Vielzahl von Merkmalen abgeleitet. Bei einer Ausführungsform der Erfindung ist es möglich, diese Merkmalskombinationen im Expertenwissen nach Dokumenttypen zu klassifizieren. Dadurch wird die Möglichkeit geschaffen, gemeinsame Merkmale zu extrahieren, die einen jeweiligen Dokumenttyp eindeutig identifizieren. Ist eine ausreichend große Anzahl von Dokumenten eines Dokumenttyps gemäß dem Verfahren oder von der Vorrichtung bearbeitet worden, so kann dieser Dokumenttyp einfacher und schneller zuverlässig identifiziert werden. Die Anzahl der hierfür benötigten Merkmale wird automatisch auf ein notwendiges und hinreichendes Maß reduziert.In a further development, it is provided that a feature combination of a document is supplied to expert knowledge if this combination of features is unknown and verification information is detected which identifies the associated document as an authentic and / or non-authentic document. In the case of a type of document that can not be identified according to expert knowledge, a plurality of features is preferably derived. In one embodiment of the invention, it is possible to classify these feature combinations in expert knowledge according to document types. This provides the ability to extract common features that uniquely identify a particular document type. If a sufficiently large number of documents of a document type have been processed according to the method or by the device, then this type of document can be reliably identified more easily and quickly. The number of required features is automatically reduced to a necessary and sufficient level.
Bei einer Weiterbildung des erfindungsgemäßen Verfahrens bzw. einer das erfindungsgemäße Verfahren umsetzenden Vorrichtung ist vorgesehen, dass das Expertenwissen Informationen über einzelne typische Merkmale von authentifizierten Dokumenten und/oder über typische gefälschte Merkmale umfasst. Diese sind nicht unmittelbar einzelnen Dokumenttypen zugeordnet. Bei einer solchen Ausführungsform erfolgt die Authentifizierung anhand des Vorliegens einzelner und/oder einer Kombination von Merkmalen, die identifiziert sind und die nicht explizit über das Expertenwissen einem Dokumententyp zugeordnet sind. Dieses Verfahren und eine entsprechende Vorrichtung können beispielsweise dort eingesetzt werden, wo Expertenwissen über eine konkrete Ausgestaltung einzelner Dokumenttypen nicht preisgegeben werden soll. Gegenüber einer gewöhnlichen Sichtkontrolle wird das
Auffinden nicht authentischer Dokumente durch einen Einsatz einer solchen Vorrichtung bzw. die Anwendung eines solchen Verfahrens bereits deutlich gesteigert.In a further development of the method according to the invention or a device implementing the method according to the invention, it is provided that the expert knowledge comprises information about individual typical features of authenticated documents and / or about typical forged features. These are not directly assigned to individual document types. In such an embodiment, the authentication is based on the presence of individual and / or a combination of features that are identified and that are not explicitly assigned to a document type via the expert knowledge. This method and a corresponding device can be used, for example, where expert knowledge about a specific embodiment of individual document types should not be revealed. Compared to an ordinary visual inspection is the
Finding non-authentic documents by using such a device or the application of such a method already significantly increased.
Bei einer Ausführungsform ist vorgesehen, dass Merkmalen und/oder Kombinationen von Merkmalen Gewichtungen zugeordnet sind und anhand der für ein Dokument abgeleiteten Merkmale und Merkmalskombinationen aus den Gewichtungen eine Gesamtgewichtung ermittelt wird, die ein Maß für ein Vorliegen eines authentischen Dokuments angibt. Hierbei können Merkmale, die auf eine Fälschung hinweisen, negativ berücksichtigt werden.In one embodiment, it is provided that features and / or combinations of features are assigned weightings, and based on the features and feature combinations derived from the weights for a document, an overall weighting is determined that indicates a measure of the presence of an authentic document. Here, features that indicate a forgery can be negatively considered.
Nachfolgend wird die Erfindung anhand bevorzugter Ausführungsbeispiele unter Bezugnahme auf eine Zeichnung näher erläutert. Hierbei zeigen:
- Fig. 1
- ein schematisches Ablaufdiagramm für eine Ausführungsform eines Verfahrens zur Verifikation von Dokumenten;
- Fig. 2
- eine schematische Darstellung eines Dokuments;
- Fig. 3
- eine schematische Darstellung der sich aus der Wavelet-Transformation ergebenden Spektralbereiche zu dem Dokument nach
Fig. 2 ; - Fig. 4, 5
- grafische Darstellungen statistischer Auswertungen für ein glattes (
Fig. 4 ) und ein raues (Fig. 5 ) Dokument; - Fig. 6
- eine schematische Darstellung einer Vorrichtung zum Verifizieren eines Dokuments;
- Fig. 7
- eine schematische Darstellung zur Erläuterung einer lokalisiert durchgeführten Verifikationsprüfung; und
- Fig. 8
- eine schematische Darstellung einer Verifikationsvorrichtung, die eine externe Datenbank mit Expertenwissen nutzt.
- Fig. 1
- a schematic flow diagram for an embodiment of a method for verification of documents;
- Fig. 2
- a schematic representation of a document;
- Fig. 3
- a schematic representation of the resulting from the wavelet transform spectral regions to the document
Fig. 2 ; - Fig. 4, 5
- graphical representations of statistical evaluations for a smooth (
Fig. 4 ) and a rough (Fig. 5 ) Document; - Fig. 6
- a schematic representation of an apparatus for verifying a document;
- Fig. 7
- a schematic representation for explaining a localized conducted verification test; and
- Fig. 8
- a schematic representation of a verification device that uses an external database with expert knowledge.
In
In
In
Gemäß dem schematisch dargestellten Verfahrensablauf nach
Hierzu wird Expertenwissen verwendet, welches angibt, wie die Wavelet-Pixel einzeln oder in bestimmten Wavelet-Pixelgruppen, welche auch als Wavelet-Pixelbereiche bezeichnet werden, auszuwerten sind, um ein Merkmal abzuleiten. Die einzelnen Wavelet-Pixel eines Wavelet-Pixelbereichs müssen nicht räumlich benachbart oder aus demselben Spektralbereich oder aus derselben Abbildung des Dokuments stammen. Vielmehr können Wavelet-Pixel zu einem Wavelet-Pixelbereich beliebig kombiniert werden, solange sie aus Abbildungen ein und desselben Dokuments stammen.For this purpose, expert knowledge is used, which specifies how the wavelet pixels are to be evaluated individually or in particular wavelet pixel groups, which are also referred to as wavelet pixel areas, in order to derive a feature. The individual wavelet pixels of a wavelet pixel region need not be spatially adjacent or from the same spectral region or from the same image of the document. Rather, wavelet pixels can be arbitrarily combined into a wavelet pixel range as long as they come from images of one and the same document.
Ein Ableiten von Merkmalen, welche beispielsweise ein Druckverfahren, eine Oberflächenbeschaffenheit, eine Zeichengröße usw. sein können, kann am einfachsten durch eine statistische Auswertung einzelner Pixel oder von Pixelbereichen erfolgen. Beispielhaft wird dies anhand der
Es ergibt sich für den Fachmann, dass hier lediglich exemplarisch eine Ableitung eines möglichen Merkmals beschrieben ist. Eine Vielzahl weiterer Merkmale, welche beispielsweise eine Angabe umfassen, mit welchem Druckverfahren ein Bereich und/oder das gesamte Dokument bedruckt sind, eine Angabe über verwendete Farben und/oder Farbzusammensetzungen usw., abgeleitet werden können. Die benötigten Auswerteschritte, d.h. Rechenschritte, die zum Ableiten einzelner Merkmale aus den Wavelet-Pixeln oder Wavelet-Pixelbereichen notwendig sind, sind in der Regel nicht aufwendig. Um beispielsweise festzustellen, ob ein Dokument rau oder glatt ist, wäre es in dem oben gezeigten Beispiel vermutlich ebenfalls ausreichend, Koeffizientenwertbereiche im Bereich der Linienfüße auf ein Vorhandensein von Koeffizientenwerten zu untersuchen. Ein vollständiges Histogramm müsste somit nicht erstellt werden.It is clear to the person skilled in the art that only a derivation of a possible feature is described here by way of example. A multiplicity of further features, which for example comprise an indication, with which printing method an area and / or the entire document are printed, an indication of colors and / or color compositions used, etc., can be derived. The required evaluation steps, that is to say computing steps which are necessary for deriving individual features from the wavelet pixels or wavelet pixel areas, are generally not complicated. For example, to determine whether a document is rough or smooth, it would probably also be sufficient in the example shown above to examine coefficient value ranges in the region of the line feet for the presence of coefficient values. A complete histogram would therefore not have to be created.
Anhand der abgeleiteten Merkmale ist es möglich, unter Rückgriff auf Expertenwissen, welches eine Klassifizierung 10 zulässt, eine Identifizierung und/oder Verifikation des Dokuments vorzunehmen. Schließlich wird ein Ergebnis der Verifikation ausgegeben 11.Based on the derived characteristics it is possible to identify and / or verify the document by resorting to expert
In
Die Vorrichtung 41 umfasst ferner eine Transformationseinheit 47. Die Transformationseinheit 47 ist ausgebildet, die die Abbildungen repräsentierenden Daten einer Wavelet-Transformation zu unterziehen. Hierdurch werden für jede Abbildung Spektralbereiche erzeugt, die vorzugsweise in einem Speicher 48 für eine weitere Bearbeitung abgelegt werden. Die Transformationseinheit 47 kann bei einigen Ausführungsformen auch mittels einer programmgesteuerten Recheneinheit 55 ausgebildet sein. Die Vorrichtung 41 umfasst ferner eine Auswerteeinheit 49, die anhand der Wavelet-Pixel oder Wavelet-Pixelbereiche, die in den erzeugten Spektralbereichen enthalten sind, Merkmale des Dokuments ableitet. Die Auswerteeinheit 49 kann mittels einer oder mehrerer elektronischen Schaltungen ausgebildet sein. Alternativ kann die Auswerteeinheit 49 mittels der programmgesteuerten Recheneinheit 55 oder einer anderen Recheneinheit realisiert sein. Zur Ableitung der Merkmale können in der Speichereinheit 48 abgelegte Informationen, beispielsweise Programmmodule 56, abgerufen werden, die als eine Repräsentation von Expertenwissen für ein Ableiten der Merkmale angesehen werden können.The device 41 further comprises a
Die Vorrichtung 41 umfasst ferner eine Verifikationseinheit 50, die in der dargestellten Ausführungsform ebenfalls mittels der Recheneinheit 55 realisiert ist. Diese ist ausgebildet, anhand von in der Speichereinrichtung 48 abgelegten Experteninformationen eine Klassifikation des Dokuments 46 anhand der abgeleiteten Merkmale vorzunehmen.The device 41 further comprises a verification unit 50, which in the illustrated embodiment is also realized by means of the
Bei einer sehr einfachen Ausführungsform reicht es aus, zur Verifikation des Dokuments ein einziges typisches Merkmal abzuleiten. In einem solchen Fall besteht die Verifikationseinheit aus einer Abfrageeinrichtung, die prüft, ob das abgeleitete Merkmal mit dem erwarteten Merkmal übereinstimmt, d.h., ob das Merkmal abgeleitet werden konnte. Bei komplexeren Dokumenten werden jedoch mehrere Merkmale in Kombination ausgewertet, um das Dokument zu klassifizieren und einem Dokumenttyp zuzuordnen. Die Klassifizierung kann beispielsweise mittels eines neuronalen Netzwerks oder einer Support-Vektor-Maschine erfolgen. Das Expertenwissen über die Klassifikation ist dann zumindest teilweise in dem neuronalen Netzwerk bzw. der Support-Vektor-Maschine enthalten. Diese können programmgesteuert mittels eines Rechners ausgestaltet sein.In a very simple embodiment, it suffices to derive a single typical feature for verification of the document. In such a case, the verification unit consists of an interrogator which checks whether the derived feature matches the expected feature, i.e., whether the feature could be derived. However, for more complex documents, multiple features are evaluated in combination to classify the document and assign it to a document type. The classification can be done for example by means of a neural network or a support vector machine. The expert knowledge about the classification is then at least partially contained in the neural network or support vector machine. These can be configured programmatically by means of a computer.
Es ergibt sich für den Fachmann, dass die Informationseinheit, die Auswerteeinheit und/oder die Verifikationseinheit, sofern sie ganz oder teilweise mittels eines programmgesteuerten Rechners realisiert sind, mittels ein und desselben Rechners bzw. Mikroprozessors realisiert werden können.It results for a person skilled in the art that the information unit, the evaluation unit and / or the verification unit, insofar as they are completely or partially realized by means of a program-controlled computer, can be realized by means of one and the same computer or microprocessor.
Die dargestellte Ausführungsform umfasst ferner eine Ausgabeeinheit 51, um ein Verifikationsergebnis auszugeben. Dieses kann beispielsweise eine Angabe darüber sein, ob das Dokument echt ist, eine Fälschung ist oder verfälscht ist. Bei anderen Ausführungsformen kann zusätzlich oder alternativ der Dokumententyp ausgegeben werden, als welches das Dokument identifiziert worden ist.The illustrated embodiment further includes an
Bei der Ausgabeeinheit kann es sich um eine beliebige Einheit handeln, die in der Lage ist, Informationen auszugeben. Beispielsweise kann es sich um eine Anzeigeeinrichtung handeln. Bei anderen Ausführungsformen kann es sich jedoch um eine Schnittstelle handeln, die die Ausgabe in Form von elektronischen Signalen vornimmt, die von anderen elektronischen Einrichtungen ausgewertet und gegebenenfalls in eine für einen Nutzer wahrnehmbare Form übersetzt werden können.The output unit may be any unit capable of outputting information. For example, it may be a display device. In other embodiments, however, it may be an interface that performs the output in the form of electronic signals that may be evaluated by other electronic devices and, where appropriate, translated into a form perceptible to a user.
Die Vorrichtung 41 umfasst ferner eine Eingabeeinheit 52, über die Nutzereingaben erfasst werden können. Hierdurch ist es möglich, beispielsweise das Expertenwissen zu erweitern. Bei der dargestellten Ausführungsform ist die Speichereinrichtung 48 Bestandteil der Vorrichtung 41. Bei anderen Ausführungsformen kann es vorgesehen sein, dass zumindest ein Teil des Expertenwissens in einer externen Datenbank abgelegt ist. Eine solche Ausführungsform ist beispielsweise in
Bei einigen Ausführungsformen ist vorgesehen, dass die Verifizierung anhand einzelner Merkmale oder Merkmalskombinationen vorgenommen wird, die Dokumenttypen oder Klassen von Dokumenttypen zugeordnet sind. Bei einer anderen Ausführungsform ist vorgesehen, dass ein oder mehrere Merkmale abgeleitet werden, die unterschiedliche Ausprägungen, beispielsweise unterschiedliche Werte, annehmen können. Anhand der unterschiedlichen ermittelten Ausprägungen der einzelnen Merkmale für ein Dokument wird dann ein Konfidenzwert gebildet, der ein Maß dafür ist, ob es sich bei dem untersuchten Dokument um ein echtes oder ein gefälschtes oder verfälschtes Dokument handelt. Die einzelnen abgeleiteten Merkmale können mit einem Wichtungsfaktor versehen sein. Eine Verifikation anhand dieses Verfahrens bietet sich insbesondere dann an, wenn eine Klasse von Dokumenten hinsichtlich einzelner Merkmale eine gewisse Variationsbreite aufweist. Schwankt beispielsweise eine Rauigkeit eines verwendeten Dokumentmaterials bei den Reisepässen eines Staates, so kann als ein Merkmal die Rauigkeit des Dokuments ermittelt werden. Bestimmt man die Rauigkeitswerte für eine große Anzahl von Dokumenten, so kann man ein Histogramm erstellen, durch welches für jeden Rauigkeitswert eine Häufigkeit angegeben werden kann, mit der dieser Rauigkeitswert auftritt. Führt man eine solche Analyse für unterschiedliche Merkmale aus und gewichtet die einzelnen Merkmale, so kann man hieraus auf einfache Weise einen Konfidenzwert dafür ableiten, dass es sich bei dem Dokument um ein echtes Dokument handelt. Weist das Dokument in vielen Einzelmerkmalen eine hohe Abweichung von dem statistischen Mittelwert für das jeweilige Merkmal auf, so wird ein resultierender Konfidenzwert sich stark von dem Konfidenzwert derjenigen Dokumente unterscheiden, deren Einzelmerkmale jeweils nahe an den entsprechenden Mittelwerten liegen. Dieses Verfahren ermöglicht es, anhand der Ermittlung einfach festzustellender Merkmale für Dokumente eine zuverlässige Echtheitsprüfung und gegebenenfalls auch eine Identifizierung vorzunehmen. Hierbei wird angemerkt, dass eine Identifizierung nicht nur möglich ist, indem Dokumente mit einer Variationsbreite in den einzelnen Merkmalen untersucht werden. Vielmehr ist es ebenso möglich, Merkmale, die unterschiedliche Ausprägungen haben, auf diese Weise miteinander zu kombinieren, um jedem Dokument einen charakteristischen Wert, welcher beispielsweise als Erkennungswert bezeichnet wird, zuzuordnen. Die Erkennungswerte von echten Dokumenten eines Dokumententyps, beispielsweise eines Reisepasses einer Nation, werden alle einen ähnlichen Wert aufweisen. Anhand der Erkennungswerte ist somit auch eine Klassifikation der Dokumente möglich. Hierbei wird herausgestellt, dass keine genaue Zuordnung einer Merkmalskombination mit bestimmten Ausprägungen zu einem bestimmten Dokumenttyp in dem Expertenwissen gespeichert ist.In some embodiments, verification is made based on individual features or feature combinations associated with document types or classes of document types. In another embodiment, it is provided that one or more features are derived, which may assume different characteristics, for example different values. Based on the different determined characteristics of the individual features for a document, a confidence value is then formed, which is a measure of whether the document examined is a genuine or a falsified or falsified document. The individual derived features may be provided with a weighting factor. Verification by means of this method is particularly suitable when a class of documents has a certain range of variation with respect to individual features. If, for example, a roughness of a documentary material used varies with the passports of a state, the roughness of the document can be determined as a feature. If one determines the roughness values for a large number of documents, one can create a histogram by which a frequency can be given for each roughness value with which this roughness value occurs. If one carries out such an analysis for different characteristics and weights the individual characteristics, one can derive from this in a simple way a confidence value for the fact that the document is a genuine document. If the document has a large deviation from the statistical mean value for the respective feature in many individual features, then a resulting confidence value will differ greatly from the confidence value of those documents whose individual features are each close to the corresponding mean values. This method makes it possible to perform a reliable authenticity check and, if necessary, also an identification based on the determination of easily determinable features for documents. It should be noted that identification is not only possible by examining documents with a range of variation in the individual features. Rather, it is also possible to combine features that have different characteristics in this way in order to assign each document a characteristic value, which is for example referred to as a recognition value. The recognition values of genuine documents of a document type, for example a passport of a nation, will all have a similar value. Based on the recognition values, a classification of the documents is thus possible. Here it is pointed out that no exact assignment of a Characteristic combination with certain characteristics for a specific document type in which expert knowledge is stored.
Ein solches Verfahren eignet sich ferner, um ein selbstlernendes Verifikationssystem aufzubauen. Werden beispielsweise sämtliche Dokumente hinsichtlich einer Vielzahl von Merkmalen untersucht, so können die ermittelten Erkennungswerte in einem Speicher abgelegt werden. Zumindest während einer Lernphase ist es notwendig, dass eine auf andere Weise ausgeführte Überprüfung des Dokuments, beispielsweise über eine Prüfperson oder eine andere Verifikationsvorrichtung, vorgenommen wird, und zu jedem Dokument eine Verifikationsentscheidung oder eine andere Verifikationsinformation erfasst wird. Im einfachsten Fall wird lediglich erfasst, ob es sich um ein echtes oder um ein gefälschtes oder verfälschtes Dokument handelt. Somit ist es möglich, einzelnen Erkennungswerten eine Verifikationsinformation zuzuordnen. Sobald eine ausreichend große Anzahl von Dokumenten des gleichen Typs erfasst ist, können Dokumente dieses Typs zuverlässig als echt, gefälscht oder verfälscht erkannt werden.Such a method is also suitable for constructing a self-learning verification system. If, for example, all documents are examined with respect to a plurality of features, then the determined recognition values can be stored in a memory. At least during a learning phase, it is necessary to have a different way of checking the document, for example, by a reviewer or other verification device, and to capture a verification decision or other verification information for each document. In the simplest case, it only records whether it is a genuine or a forged or falsified document. Thus, it is possible to assign verification information to individual recognition values. Once a sufficiently large number of documents of the same type is captured, documents of this type can be reliably identified as being genuine, falsified or falsified.
Bei einer anderen Ausführungsform eines selbstlernenden Verifikationsverfahrens bzw. einer entsprechenden Vorrichtung ist vorgesehen, dass für die einzelnen erfassten Dokumente in einer Lernphase die konkreten Merkmalsausprägungen der abgeleiteten Merkmale abgespeichert werden. Hierdurch werden differenzierte Verifikationsanalysen erleichtert. Soll nicht nur eine Prüfung auf Echtheit, sondern zusätzlich eine Klassifikation ausgeführt werden, so ist dieses bei einer großen Anzahl von möglichen Dokumenten zuverlässiger möglich, wenn für die einzelnen Dokumenttypen charakteristische Merkmalskombinationen oder Merkmalsausprägungen gespeichert werden.In another embodiment of a self-learning verification method or a corresponding device, it is provided that the concrete characteristic values of the derived features are stored for the individual documents recorded in a learning phase. This facilitates differentiated verification analyzes. If not only an authenticity check but also a classification is to be carried out, this is more reliably possible with a large number of possible documents if characteristic feature combinations or characteristic values are stored for the individual document types.
Bei einer Ausführungsform ist vorgesehen, dass in der Lernphase zunächst für alle Dokumente dieselben Merkmale abgeleitet und deren Ausprägungen gespeichert werden. Zusätzlich wird eine Verifikationsinformation, die beispielsweise das Dokument identifiziert, erfasst. Sind eine Vielzahl von Dokumenten desselben Dokumenttyps erfasst worden, so lässt es sich ermitteln, in welchen Merkmalen bzw. mit welchen Merkmalsausprägungen die Dokumente eines Dokumententyps übereinstimmen bzw. sich von anderen Dokumenttypen abgrenzen lassen. Auf diese Weise kann es erreicht werden, dass nach der Lernphase nur noch eine begrenzte Anzahl von Merkmalen abzuleiten ist, um eine Klassifizierung eines Dokuments vorzunehmen. Ebenso ist es selbstverständlich möglich, Merkmale zu erfassen, die eindeutig gefälschte und/oder verfälschte Dokumente kennzeichnen. Häufig ist es sinnvoll, die Dokumente ebenfalls auf ein Vorliegen solcher Merkmale zu untersuchen, um eine Verifikationsentscheidung zu treffen. Dies gilt insbesondere dann, wenn ein Dokument anhand von Merkmalen identifizierbar ist, die beim Verfälschen des Dokuments nicht verändert werden.In one embodiment, it is provided that in the learning phase, the same features are first derived for all documents and their characteristics are stored. In addition, verification information identifying, for example, the document is detected. If a large number of documents of the same document type has been recorded, it is possible to determine in which characteristics or with which characteristic values the documents of a document type match or can be delimited from other document types. In this way, it can be achieved that only a limited number of features have to be derived after the learning phase in order to classify a document. Likewise, it is of course possible to detect features that are clearly fake and / or Mark falsified documents. It often makes sense to also examine the documents for the presence of such features in order to make a verification decision. This applies in particular if a document can be identified on the basis of features that are not changed when the document is falsified.
Bei einer Ausführungsform ist vorgesehen, dass eine Verifikation einzelner Merkmale, insbesondere lokal aufgelöst, vorgenommen wird, nachdem der Dokumenttyp identifiziert ist. Hierdurch ist es möglich, insbesondere solche Merkmale auf Verfälschungen zu untersuchen, die für eine Individualisierung und/oder Personalisierung eines Dokuments verantwortlich sind. Dies sind beispielsweise biometrische Daten in Form eines Passbilds oder alphanumerische Aufdrucke, die einen Namen, eine Anschrift, ein Geburtsdatum, eine Seriennummer usw. umfassen.In one embodiment, it is provided that a verification of individual features, in particular locally resolved, is performed after the document type has been identified. This makes it possible, in particular to examine such features for adulteration, which are responsible for an individualization and / or personalization of a document. These are, for example, biometric data in the form of a passport photograph or alphanumeric imprints comprising a name, an address, a date of birth, a serial number, etc.
Anhand von
Es ergibt sich für den Fachmann, dass hier lediglich beispielhafte Ausführungsformen beschrieben sind. Die einzelnen im Zusammenhang mit den verschiedenen Ausführungsformen beschriebenen Merkmale können beliebig kombiniert werden, um ein Verifikationsverfahren oder eine Verifikationsvorrichtung auszubilden.It will be apparent to those skilled in the art that only exemplary embodiments are described herein. The individual features described in connection with the various embodiments can be arbitrarily combined to form a verification method or a verification device.
- 11
- Erfassen von AbbildungenCapturing pictures
- 22
- Ausführen einer Wavelet-TransformationPerform a wavelet transform
- 33
- AbbildungIllustration
- 4-n4-n
- Frequenzbereichfrequency range
- 4-n-a4-n-a
- Spektralbereich zur Richtung aSpectral range to the direction a
- 55
- Längskantelongitudinal edge
- 66
- kurze Seitenkanteshort side edge
- 77
- Diagonalediagonal
- 88th
- Sicherheitsmerkmalsafety feature
- 99
- Ableiten von MerkmalenDeriving characteristics
- 1010
- Klassifikation/VerifizierenClassification / Verify
- 1111
- Ausgeben eines ErgebnissesOutput a result
- 2222
- Abszisseabscissa
- 2323
- Ordinateordinate
- 25-2725-27
- Linienlines
- 28-3028-30
- Breite der LinienWidth of the lines
- 4141
- Vorrichtung zur VerifikationDevice for verification
- 4242
- Erfassungseinrichtungdetector
- 4343
- grafische Erfassungseinheitgraphical registration unit
- 44, 4544, 45
- Beleuchtungseinrichtungenlighting equipment
- 4646
- Dokumentdocument
- 4747
- Transformationseinheittransformation unit
- 4848
- Speichereinrichtungmemory device
- 4949
- Auswerteeinheitevaluation
- 5050
- Verifikationseinheitverification unit
- 5151
- Ausgabeeinheitoutput unit
- 5252
- Eingabeeinheitinput unit
- 5353
- Kommunikationsnetzcommunication network
- 5454
- DatenbankDatabase
- 5555
- Recheneinheitcomputer unit
- 5656
- Programmmoduleprogram modules
- 6161
- Dokumentdocument
- 62-1 ... 62-1062-1 ... 62-10
- lokale Bereichelocal areas
Claims (15)
- Method for verifying documents (46; 61), comprising the steps:detecting a plurality of different images (3) of the same document (46; 61) under different recording conditions by means of a detection device (42);deriving of features from the images (3) by means of an evaluation unit (49);comparing the features to expert knowledge by means of a verification unit (50),verifying the document (46; 61) using a result of the comparison, wherein the images (3) are subjected in each case by means of a transformation unit (47) to a wavelet transformation, and the features are derived from individual wavelet pixels or wavelet pixel areas, wherein each wavelet transformation produces for each frequency range (4-n) three spectral ranges (4-n-x), which are allocated accordingly to a spatial direction of a basis vector of a co-ordinate system and to a diagonal (7) in the co-ordinate system of the local region of the image (3).
- Method according to any one of the preceding claims, characterised in that one of the evaluated wavelet pixel areas comprises wavelet pixels which are not allocated to a cohesive local region of a spectral range (4-n-x).
- Method according to any one of the preceding claims, characterised in that the wavelet transformations are carried out for the plurality of different images (3) of the same document (46; 61) with different wavelet basis functions and/or scalings.
- Method according to any one of the preceding claims, characterised in that, for the derivation of features, a plurality of wavelet pixels are evaluated in combination, which are allocated to a plurality of different images (3) of the same security document and/or different local regions and/or different spectral ranges (4-n-x).
- Method according to any one of the preceding claims, characterised in that wavelet pixels or wavelet pixel areas are evaluated statistically, in order to derive at least one of the features for the document (46; 61).
- Method according to any one of the preceding claims, characterised in that the wavelet pixels or wavelet pixel groups are evaluated for statistical deviations in comparison with typical statistical default properties, in order to detect any falsifications or corruptions.
- Method according to any one of the preceding claims, characterised in that the features comprise identified edges of printed characters or symbols, and wavelet pixel areas are statistically evaluated in the surrounding areas of identified edges of printed characters or symbols, in order to investigate a surface roughness for statistical deviations.
- Method according to any one of the preceding claims, characterised in that, on the basis of the derived features, by means of a neural network or a support vector machine, a classification of the document (46; 61) is carried out, wherein the expert knowledge is stored in the form of the self-learning neural network or the support vector machine.
- Method according to any one of the preceding claims, characterised in that first features are derived from the wavelet pixels, and, on the basis of the first features, a classification takes place of the document (46; 61) into a document type;
from a databank, information regarding second features of the document type detected is called up, and
on the basis of the information regarding the second features, the second features of the document (46; 61) are examined in respect of a presence and/or an absence of falsification or corruption by means of an evaluation of wavelet pixels or wavelet pixel groups. - Method according to any one of the preceding claims, characterised in that, after a derivation of a statistical property which is not or cannot be allocated to any feature in the expert knowledge, an answer is determined to the question as to whether this property represents:a) a new authentic feature of a document (46; 61), orb) a new falsified feature of a document (46; 61), orc) a new feature, displaying a falsification, of a document (46; 61), ord) none of the foregoing alternatives,and, depending on the answer determined in the cases a) to c), the expert knowledge is expanded.
- Method according to any one of the preceding claims, characterised in that feature combinations of documents (46; 61) are added to the expert knowledge if this feature combination is unknown, and an item of verification information is detected which identifies the pertinent document (46; 61) is identified as an authentic or not authentic document (46;61).
- Method according to any one of the preceding claims, characterised in that the expert knowledge comprises information regarding individual typical features of authentic documents (46; 61) and/or regarding typical falsified features, which are not allocated directly to individual document types.
- Method according to any one of the preceding claims, characterised in that an authentication is carried out on the basis of a presence of individual features or a combination of features which are identified, which are not explicitly allocated by the expert knowledge to a document type.
- Method according to any one of the preceding claims, characterised in that features and/or combinations of features are allocated to weightings, and, on the basis of the features and feature combinations derived for a document (46; 61) from the weightings, an overall weighting is determined, which provides a measure for a presence of an authentic document (46; 61).
- Device (41) for verifying documents (46; 61), wherein the device (41) is configured so as to carry out the method according to any one of claims 1 to 14, with a detection device (42) for detecting a plurality of different images (3) of the same document (46; 61) under different recording conditions;
an evaluation device (49) for deriving features from the images (3);
a verification unit for comparing the features with expert knowledge; and
a transformation unit (47), in order to subject each of the plurality of images (3) in each case to a wavelet transformation, wherein each wavelet transformation produces for each frequency range (4-n) three spectral ranges (4-n-x), which are allocated accordingly to a spatial direction of a basis vector of a co-ordinate system and to a diagonal (7) in the co-ordinate system of the local region of the image (3),
wherein the evaluation unit (49) is configured to derive the features from individual wavelet pixels or wavelet pixel areas, and the device (41) is configured so as to verify the document (46; 61) on the basis of the comparison, and comprises an output unit (51), in order to output the verification result.
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
DE102009007588A DE102009007588A1 (en) | 2009-02-05 | 2009-02-05 | Method and apparatus for verifying documents using a wavelet transform |
PCT/EP2010/000956 WO2010089157A1 (en) | 2009-02-05 | 2010-02-03 | Method and device for verifying document using a wavelet transformation |
Publications (2)
Publication Number | Publication Date |
---|---|
EP2394250A1 EP2394250A1 (en) | 2011-12-14 |
EP2394250B1 true EP2394250B1 (en) | 2019-05-08 |
Family
ID=42154526
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
EP10705107.0A Active EP2394250B1 (en) | 2009-02-05 | 2010-02-03 | Method and device for verifying document using a wavelet transformation |
Country Status (3)
Country | Link |
---|---|
EP (1) | EP2394250B1 (en) |
DE (1) | DE102009007588A1 (en) |
WO (1) | WO2010089157A1 (en) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE102015003480A1 (en) * | 2015-03-18 | 2016-03-10 | Giesecke & Devrient Gmbh | Apparatus and method for checking value documents, in particular banknotes, as well as value document processing system |
DE102019105985A1 (en) * | 2019-03-08 | 2020-09-10 | Bundesdruckerei Gmbh | Method and device for automatic quality assessment |
DE102020120962B4 (en) | 2020-08-07 | 2023-03-30 | ThePeople.de GmbH | Recognition method and recognition system for clearly recognizing an object |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20030099379A1 (en) | 2001-11-26 | 2003-05-29 | Monk Bruce C. | Validation and verification apparatus and method |
US6785405B2 (en) | 2002-10-23 | 2004-08-31 | Assuretec Systems, Inc. | Apparatus and method for document reading and authentication |
KR101001691B1 (en) | 2006-03-13 | 2010-12-15 | 노틸러스효성 주식회사 | Recognizing the Denomination of a Note Using Wavelet transform |
EP2000992A1 (en) | 2007-06-01 | 2008-12-10 | Kba-Giori S.A. | Authentication of security documents, in particular of banknotes |
-
2009
- 2009-02-05 DE DE102009007588A patent/DE102009007588A1/en not_active Withdrawn
-
2010
- 2010-02-03 EP EP10705107.0A patent/EP2394250B1/en active Active
- 2010-02-03 WO PCT/EP2010/000956 patent/WO2010089157A1/en active Application Filing
Non-Patent Citations (1)
Title |
---|
None * |
Also Published As
Publication number | Publication date |
---|---|
DE102009007588A1 (en) | 2010-08-19 |
EP2394250A1 (en) | 2011-12-14 |
WO2010089157A1 (en) | 2010-08-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
DE112007001791B4 (en) | Method and device for comparing document features by means of texture analysis | |
EP2095341B1 (en) | Method for identifying soiling and/or colour fading in the region of colour transitions on documents of value, and means for carrying out the method | |
DE69128457T2 (en) | High speed document authentication system | |
DE112007001793B4 (en) | Method and device for comparing documents by means of a cross-level comparison | |
EP2625673B1 (en) | Method for checking an optical security feature of a valuable document | |
DE112007001792T5 (en) | Method and device for comparing document features by means of pattern recognition | |
EP3746992B1 (en) | Method for checking the authenticity and/or integrity of a security document having a printed security feature, security feature and arrangement for verification | |
DE102014108578A1 (en) | Method for checking a security document | |
DE102008041944A1 (en) | Data processing method for generation of data file and for testing authenticity of e.g. passport, involves inputting value set and associated information to learning module, and outputting classifier by learning module | |
EP2761604B1 (en) | Method for examining the production quality of an optical security feature of a valuable document | |
EP1815444B1 (en) | Device and method for the visual representation of measured values | |
CH710713B1 (en) | Authentication method using surface paper texture. | |
EP2394250B1 (en) | Method and device for verifying document using a wavelet transformation | |
EP3889922B1 (en) | Integrity check of a document with personal data | |
WO2011082792A1 (en) | Method for testing value documents | |
DE10105273A1 (en) | Method for checking a franking applied to a mail item and device for carrying out the method | |
EP2724323B1 (en) | Method and device for creating a document reference data set on the basis of a document | |
DE102008041113A1 (en) | Method of verifying the authenticity of a document, computer program product, testing device and data processing system | |
EP4205091B1 (en) | Device for generating a digital identifier of a copy of a printed object, said copy having at least one printed image, smartphone or tablet comprising said device, and method for using said device | |
DE102017126207A1 (en) | Sensor system for testing hand vein patterns | |
WO2011082793A1 (en) | Method for testing value documents | |
WO2022029321A1 (en) | Producing document templates having security features | |
EP3308348B1 (en) | Method for producing a security element with secretly coded information in a graphical image, security element and method for verification | |
DE102020004284A1 (en) | Method and device for checking documents of value and method and device for generating checking parameters for use in a method for checking documents of value | |
WO2014029476A1 (en) | Method and apparatus for checking valuable documents |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PUAI | Public reference made under article 153(3) epc to a published international application that has entered the european phase |
Free format text: ORIGINAL CODE: 0009012 |
|
17P | Request for examination filed |
Effective date: 20110905 |
|
AK | Designated contracting states |
Kind code of ref document: A1 Designated state(s): AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO SE SI SK SM TR |
|
DAX | Request for extension of the european patent (deleted) | ||
17Q | First examination report despatched |
Effective date: 20141111 |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: EXAMINATION IS IN PROGRESS |
|
GRAP | Despatch of communication of intention to grant a patent |
Free format text: ORIGINAL CODE: EPIDOSNIGR1 |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: GRANT OF PATENT IS INTENDED |
|
INTG | Intention to grant announced |
Effective date: 20181123 |
|
GRAS | Grant fee paid |
Free format text: ORIGINAL CODE: EPIDOSNIGR3 |
|
GRAJ | Information related to disapproval of communication of intention to grant by the applicant or resumption of examination proceedings by the epo deleted |
Free format text: ORIGINAL CODE: EPIDOSDIGR1 |
|
GRAL | Information related to payment of fee for publishing/printing deleted |
Free format text: ORIGINAL CODE: EPIDOSDIGR3 |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: EXAMINATION IS IN PROGRESS |
|
GRAR | Information related to intention to grant a patent recorded |
Free format text: ORIGINAL CODE: EPIDOSNIGR71 |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: GRANT OF PATENT IS INTENDED |
|
GRAA | (expected) grant |
Free format text: ORIGINAL CODE: 0009210 |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: THE PATENT HAS BEEN GRANTED |
|
INTC | Intention to grant announced (deleted) | ||
AK | Designated contracting states |
Kind code of ref document: B1 Designated state(s): AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO SE SI SK SM TR |
|
INTG | Intention to grant announced |
Effective date: 20190401 |
|
REG | Reference to a national code |
Ref country code: GB Ref legal event code: FG4D Free format text: NOT ENGLISH |
|
REG | Reference to a national code |
Ref country code: CH Ref legal event code: EP Ref country code: AT Ref legal event code: REF Ref document number: 1131407 Country of ref document: AT Kind code of ref document: T Effective date: 20190515 |
|
REG | Reference to a national code |
Ref country code: DE Ref legal event code: R096 Ref document number: 502010015981 Country of ref document: DE Ref country code: IE Ref legal event code: FG4D Free format text: LANGUAGE OF EP DOCUMENT: GERMAN |
|
REG | Reference to a national code |
Ref country code: NL Ref legal event code: MP Effective date: 20190508 |
|
REG | Reference to a national code |
Ref country code: LT Ref legal event code: MG4D |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: HR Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20190508 Ref country code: LT Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20190508 Ref country code: NL Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20190508 Ref country code: PT Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20190908 Ref country code: FI Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20190508 Ref country code: SE Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20190508 Ref country code: NO Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20190808 Ref country code: ES Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20190508 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: GR Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20190809 Ref country code: BG Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20190808 Ref country code: LV Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20190508 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: EE Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20190508 Ref country code: SK Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20190508 Ref country code: DK Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20190508 Ref country code: RO Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20190508 Ref country code: CZ Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20190508 |
|
REG | Reference to a national code |
Ref country code: DE Ref legal event code: R097 Ref document number: 502010015981 Country of ref document: DE |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: SM Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20190508 Ref country code: IT Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20190508 |
|
PLBE | No opposition filed within time limit |
Free format text: ORIGINAL CODE: 0009261 |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: NO OPPOSITION FILED WITHIN TIME LIMIT |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: TR Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20190508 |
|
26N | No opposition filed |
Effective date: 20200211 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: PL Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20190508 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: SI Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20190508 |
|
REG | Reference to a national code |
Ref country code: CH Ref legal event code: PL |
|
REG | Reference to a national code |
Ref country code: BE Ref legal event code: MM Effective date: 20200229 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: MC Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20190508 Ref country code: LU Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES Effective date: 20200203 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: LI Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES Effective date: 20200229 Ref country code: CH Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES Effective date: 20200229 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: IE Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES Effective date: 20200203 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: BE Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES Effective date: 20200229 |
|
REG | Reference to a national code |
Ref country code: AT Ref legal event code: MM01 Ref document number: 1131407 Country of ref document: AT Kind code of ref document: T Effective date: 20200203 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: AT Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES Effective date: 20200203 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: MT Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20190508 Ref country code: CY Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20190508 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: MK Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20190508 Ref country code: IS Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20190908 |
|
P01 | Opt-out of the competence of the unified patent court (upc) registered |
Effective date: 20230526 |
|
PGFP | Annual fee paid to national office [announced via postgrant information from national office to epo] |
Ref country code: DE Payment date: 20240216 Year of fee payment: 15 Ref country code: GB Payment date: 20240222 Year of fee payment: 15 |
|
PGFP | Annual fee paid to national office [announced via postgrant information from national office to epo] |
Ref country code: FR Payment date: 20240221 Year of fee payment: 15 |