Title of Invention	METHOD AND APPARATUS FOR VERIFYING VALUE DOCUMENTS
Abstract	The invention relates to a method and a device for identifying forged value documents, for example composed forgeries consisting of parts of different value documents. According to the method, a characteristic of a first feature of the value document to be verified is determined in a first step and a characteristic of a second feature of the value document to be verified is determined in a second step. In an additional step, the characteristics of both features are compared and the difference is determined. Alternatively, in the additional step an anticipated characteristic for the second feature is compared to the characteristic of the second feature and the difference is determined. Verification then takes place as to whether the determined difference lies in a predetermined value range that contains target values which are defined as acceptable for said difference.

Title of Invention

METHOD AND APPARATUS FOR VERIFYING VALUE DOCUMENTS

Abstract

The invention relates to a method and a device for identifying forged value documents, for example composed forgeries consisting of parts of different value documents. According to the method, a characteristic of a first feature of the value document to be verified is determined in a first step and a characteristic of a second feature of the value document to be verified is determined in a second step. In an additional step, the characteristics of both features are compared and the difference is determined. Alternatively, in the additional step an anticipated characteristic for the second feature is compared to the characteristic of the second feature and the difference is determined. Verification then takes place as to whether the determined difference lies in a predetermined value range that contains target values which are defined as acceptable for said difference.

Full Text	Method and apparatus for checking documents of value [001] The invention relates to a method for checking documents of value, in particular for recognizing forged documents of value, and an apparatus for carrying out the method. The forged documents of value to be recognized are composed forgeries which are composed of parts of different documents of value. The composed forgeries can be composed of parts of authentic and of forged documents of value, but composed forgeries that are composed exclusively from parts of authentic documents of value are also known. [002] From the prior art there are known various methods for recognizing forged bank notes. For recognizing composed forgeries, for example methods for detecting adhesive strips are employed by which the individual parts of the composed forgery are glued together. For the authenticity check, moreover, the bank notes for example are checked as to properties which distinguish the authentic bank note paper from normal paper, for example as to its fluorescence properties. Many composed forgeries partly consist of authentic and partly of forged paper which, however, has fluorescence properties similar to those of authentic bank notes. Furthermore, forgeries are composed which exclusively consist of parts of authentic bank notes. With the previous methods it is not possible to recognize those composed forgeries which deliver measuring signals, e.g. fluorescence signals, comparable to those of authentic bank notes. [003] It is therefore an object of the invention to provide a possibility for recognizing forged documents of value which are composed of a plurality of parts of authentic or of authentic and forged documents of value. [004] This problem is solved by the features of the independent claims. In claims dependent thereon advantageous embodiments and developments of the invention are specified. [005] With the method according to the invention in a first step a property of a first feature of the document of value to be checked and in a second step a property of a second feature of the document of value is determined. In a further step the properties of the first and second feature are compared to each other and therefrom a difference between the first and the second feature is ascertained. Preferably, the properties to be compared are the same properties, such as e.g. the color of the first feature and the color of the second feature. The property of the first and the one of the second feature are preferably determined with the same sensor. Then the ascertained difference is evaluated. In the evaluation it is examined whether the ascertained difference lies within a predetermined range which includes the set values set to be acceptable for the difference. This range is determined for example with the aid of standard data valid for the document of value. The limits of the range of the set values are given by one or a plurality of threshold values, such as e.g. maximum values or also minimum values for the ascertained difference. The standard data contain e.g. information about reference features and about maximum acceptable deviations from the reference features. [006] The further step, which serves for ascertaining a difference, can also consist in - as an alternative to the direct comparison of the properties of the two features - that a property to be expected for the second feature is compared to the property of the second feature. The property to be expected for the second feature can be determined by extrapolation, for example by an extrapolation from the property or the properties of the first feature and/ or from further measuring data of the document of value. The extrapolation can also be carried out using standard data valid for the bank note, for example starting out from the first feature. Thus, with printed image features having an individual structure the extrapolation of the position of a second printed image segment can be effected for example from the position of a first printed image segment with the aid of a reference printed image which was determined on the basis of a multitude of corresponding printed images of authentic bank notes. The extrapolation can also be effected from further measuring data of the bank note, e.g. from the measured spatial arrangement of further features with the aid of the standard data. [007] The property to be expected can also be determined with the aid of the standard data valid for the document of value. The difference ascertained in doing so in general differs from the difference ascertained with the first-mentioned variant of the method. For evaluating, the ascertained difference is compared to set values, the set values also being determined from the standard data for the document of value. These set values, too, in general differ from the set values of the first-mentioned variant of the method. In the evaluation it is examined whether the ascertained difference lies within the range of set values which contains e.g. the maximum acceptable deviations of the property of the second feature from the property to be expected for the second feature. In the evaluating step it is also possible to examine a match of the difference with the set values and to determine a corresponding match degree. [008] The standard data which form the basis for the range of set values and/or for the determination of the second feature's property to be expected can be ascertained with the help of a multitude of documents of value or could have been ascertained in the run-up to the check. Preferably, the standard data are ascertained with the help of documents of value of the same type as the document of value to be examined, in the case of bank notes for example with the help of bank notes of the same denomination. The standard data contain, for example, information about the properties of reference features and about the maximum acceptable deviations of the properties of the features checked with the method according to the invention from the properties of the reference features. For the different types of documents of value the respective specific standard data can be stored. The standard data can be chosen with the help of the type of the document of value, for example the currency and denomination of a bank note. For choosing the standard data valid for the document of value to be checked, the type of document of value is for this purpose identified for example before the document of value is checked. In the case of bank notes this identifying can be e.g. determining the denomination previous to the method according to the invention. [009] The first and second feature to be compared with the method according to the invention can be of the same type but can also be features of a different type. The method can be carried out for a single property of the two features, but it can also be carried out for a plurality of properties of the two features. Furthermore, the method can also be carried out for further first and second features. It is especially advantageous to carry out the method for both features of the front side and features of the back side of the document of value, i.e. at least one feature pair of the front side and at least one feature pair of the back side is checked. [0010]In an embodiment the first and/or the second feature is at least one segment of at least one predetermined printed image. The first and second feature can be for example segments of the same predetermined printed image. But they can also be segments of different predetermined printed images which preferably have been produced with the same printing method (e.g. offset printing, engraving, ...), in particular within the same printing step. The printed image areas produced with the same printing method, in particular within the same printing step, have a very well defined relative position to each other which can hardly be exactly reproduced when individual parts of the documents of value are subsequently joined together. [0011]The printed images to be compared as to their properties can have one or a plurality of segments of one or a plurality of design elements, such as e.g. symbols or geometric elements, and/or of one or a plurality of image motifs, such as e.g. pictures of persons, buildings, animals, objects or the like, or, likewise, one or a plurality of segments of patterns, writings or characters, such as e.g. graphic characters, numbers, letters. In a further embodiment the first and/or second feature, additionally or alternatively to the printed image or printed segment, has at least one segment of a security element or of an edge of the document of value. [0012]The property with respect to which the first and the second feature are compared preferably relates to the position and/ or the shape of the respective feature. In the embodiment with the alternative further step the property of the first and second feature as well as the property to be expected for the second feature relates to the position and/or the shape of the respective feature. Properties which relate to the shape of the respective feature are for example a color, a spectral and/ or spatial color distribution, a size or a form of the respective feature. Properties which relate to the position of the respective feature are for example an absolute position, a relative position, in particular a distance relative to a reference feature or an orientation relative to a reference feature. As a reference feature for example an edge of the document of value, a security element, a reference printed image or also a segment of the same can be used. The reference printed image can be e.g. a further segment of the predetermined printed image of the first and/ or of the second feature, but it can also be a segment of a further printed image which has been produced preferably with the same printing method, in particular within the same printing step, as the printed image of the first and/ or second feature. [0013]With authentic bank notes the relative positions and orientations of the features are very precisely defined, in particular in case of printed features which have been produced within the same printing process or even within the same printing step. Thus, the method according to the invention is preferably carried out in such a way that as features certain printed image areas are used from which a difference concerning the position is ascertained. [0014]For evaluating, the difference ascertained with the method according to the invention is compared with the particular set values, and it is checked whether the ascertained difference lies within the range of the particular set values. In the case of an authentic document of value the ascertained difference should lie within the range of the particular set values. In the case of a composed forgery, in contrast, there is a high probability that at least a number of features at least in some of their properties have differences which do not lie within the range of the particular set values. Due to the subsequent composing of the composed forgery it is to be expected that in doing so e.g. the difference of the absolute position of some printed image areas deviates by more than the acceptable deviation from the reference values of the authentic documents of value. [0015]The evaluation of the ascertained difference is effected for example for determining the authenticity of the document of value or also for the fitness check of the document of value, for instance with regard to the presence of tears or gluings. With the aid of the method according to the invention not only composed forgeries can be detected but also authentic documents of value which, however, for the purpose of „repairing" were partly or completely glued together. This gluing together may also lead to differences in the position and/ or the shape of features which do not correspond to the set values. For example, printed images through which runs a tear and which subsequently were joined together can be changed such that the differences in position ascertained in the printed image lie outside the range of the set values. With torn authentic documents of value the differences in the position of features may also deviate from those of intact authentic documents of value and can be detected with the method according to the invention. [0016]In general, a composed forgery can have a separating line or a plurality of separating lines at which it is composed. Though the position of these separating lines on the composed forgery are not foreseeable, but for recognizing composed forgeries it is advantageous to selectively use such features which with a high probability are disposed on different partial bank notes of the composed forgery. For this reason the method according to the invention is preferably carried out on the basis of the properties of such features which are disposed on the document of value as far apart from each other as possible. Preferably, the first and second feature are chosen such that the distance between these features in one direction, in particular in the direction along the connecting line of the features, extends over a substantial part of the document of value, but at least amounts to half of the extent of the document of value. Furthermore, it is advantageous to compare further first and further second features with each other whose connecting line encloses with the connecting line of the first and second feature a large angle, e.g. an angle of at least 20°. In a development of the method the position of the separating lines of a plurality of known composed forgeries can also be detected, and therefrom a probability for the position of the separating lines on the document of value to be checked can be determined. In dependence thereon e.g. the features to be used for the method according to the invention can be chosen such that most of the known separating lines extend between these features. [0017] A second method according to the invention for recognizing forged documents of value, in particular for recognizing composed forgeries, can run in a way as follows: For checking the bank note or the composed forgery, in a first step a feature of a document of value is identified, in particular a feature of a bank note or of a composed forgery. As a feature, preferably at least one segment of at least one printed image is chosen, e.g. at least one segment of a pattern, or of a writing, or of a character, in particular of a number, of a letter, or of a graphic character. For identifying the feature, a recognition of a character string can be carried out with the aid of an OCR method. Then the size and/ or form and/ or structure of the feature is determined, preferably with a higher accuracy than during the identification. For ascertaining a difference the determined size and/ or form and/ or structure of the feature is compared with at least one size and/ or form and/ or structure to be expected for the feature. This size and/ or form and/ or structure to be expected for the feature is determined with the aid of the standard data valid for the document of value. These standard data can be chosen with the help of the previously determined type of the document of value, for example the currency and denomination of a bank note, and contain for example information about the properties of reference features and about the maximum acceptable deviations therefrom. For evaluating, the ascertained difference is compared to set values, the set values also being determined from the standard data for the document of value. In doing so, for example, it is examined whether the ascertained difference lies within the range of the set values which contains the maximum deviations with respect to the size and/ or form and/ or structure of the feature, or whether it exceeds the maximum deviations. In the evaluation it is examined, for example, how good the match is between the measured properties of the feature and the set values therefor, and a corresponding match degree is determined. In the steps where the difference is ascertained and evaluated e.g. an OCV method can be employed. [0018]In the methods according to the invention the features are detected, for example, optically, in particular in the visible or in the not visible spectral region, or magnetically. A further aspect of the invention relates to the apparatus, which is used for carrying out the method. Such an apparatus can have a sensor for checking documents of value, in particular bank notes, e.g. an image sensor for detecting optical features of the document of value or a magnet sensor for detecting magnetic features. [0019]In the following the invention is described by way of example with reference to the accompanying Figures. [0020]Fig. 1a shows a schematic representation of an authentic bank note, [0021]Fig. 1b shows a schematic representation of a composed forgery which is composed of two parts, [0022]Fig. 2a shows a misalignment of a printed image in a composed forgery, [0023]Fig. 2b shows a segment from the denomination detail of a composed forgery, the separating line of which runs through the numbers of the denomination detail. [0024]Figure 1a schematically represents the back side of an authentic bank note 10. Here are shown, by way of example, some printed images 40, 60, 70 which have been produced e.g. by means of engraving, offset printing or further printing methods, as well as a serial number 30 printed at two positions onto the bank note 10. Of course, further printed images and security elements can be additionally contained on or in the banknote 10. [0025]Figure lb schematically shows a composed forgery 1 which is composed of two parts, for example of a left authentic partial bank note 1a and a right forged partial bank note lb. At a separating line 2 running approximately perpendicular through the composed forgery 1 the two partial bank notes la and lb are glued together. In this example, however, the gluing together of the composed forgery 1 did not succeed in disposing the two lower edges 5a and 5b such that they extend precisely in one line. The lower edges 5a and 5b are slightly inclined to each other and at the intersection point with the separating line 2 they can also be a bit misaligned in vertical direction. For a better illustration the inclination is drawn relatively strong in Figure 1b. For the method according to the invention, however, it is sufficient that the two partial bank notes la and lb are not quite precisely aligned to each other. A slight deviation is sufficient which does not have to be given with respect to the edges of the bank note but may be given with respect to the property of any feature which with a corresponding accuracy can be metrologically detected. [0026]For checking a bank note 10 e.g. an image sensor is used with which, among other things, the lower edge 50 of the bank note 10 is detected. From the image data can be determined for example the absolute position of a left section of the lower edge 50, which e.g. extends from the left corner over a certain distance to the right, and analogously the absolute position of a right section of the lower edge which extends from the right corner over a distance to the left. In doing so, in a composed forgery 1 e.g. the absolute positions of the lower edge sections 5a, 5b are determined. Then a difference is ascertained by comparing these absolute positions, for example by subtraction of the coordinates of the two corner points or of a plurality of single image points of the two lower edge sections or of two center points of the two lower edge sections ascertained from the image data. For evaluating the difference, the difference is compared with set values which result from standard data valid for the document of value which for example have been determined in the run-up to the process with the help of authentic bank notes of the same denomination. The ascertained difference of the absolute positions for an authentic bank note 10 lies with a very high probability within the range of the set values for this position difference. In a composed forgery 1 whose lower edge sections 5a, 5b do not extend in a straight line the difference exceeds - in the case of a correspondingly large deviation - the limits of the range of the set values for this position difference. The result of the evaluation of the difference can be used in this way for checking a bank note, in particular for recognizing a composed forgery. The evaluation result, however, cannot only be used as an authenticity criterion but also for checking the fitness of the bank note, for example for identifying a torn bank note or a bank note glued together with adhesive strips. In these bank notes, too, a displacement of the positions of features may occur, when e.g. bank note parts have been glued together a bit obliquely or when the parts of a torn bank note are a bit displaced against each other during the transport of the bank note. [0027] In an alternative embodiment a difference of the absolute positions of the right and left lower edge section can also be ascertained in a way as follows: From a first of the absolute positions of the two lower edge sections, e.g. of the left lower edge section 5a, the absolute position to be expected for the second lower edge section, 5b, is determined. This can be effected for example by extrapolating the course of the first lower edge section 5a. Then a difference between the extrapolated position to be expected of the second lower edge section 5b and the actual position of the second lower edge section 5b is ascertained and compared with set values for the difference. Since here the match between an actual property and the set value for this property is examined, the difference and the set value in this embodiment are relatively small compared to those of the previously described embodiment. [0028]The deviation from the set values, which results from the composing of the partial bank notes la, 1b, could be recognized by many features of the composed forgery 1. With the method according to the invention in the first and second procedure step there can be compared e.g. the properties described in the following of the following features: The serial numbers 3a and 3b have a relative position and a relative orientation to each other which deviates from the relative position and orientation of the two serial numbers 30 of the authentic bank note. Since the separating line 2 extends through between the numbers of the denomination ,,1000" of the bank note, the printed images of the numbers 4a and 4b also have a deviating relative position and orientation compared to the respective characters of the authentic bank note 10. The separating line 2 furthermore extends through the printed images shown above the denomination detail, so that the partial printed images 6a and 6b or 7a and 7b, too, have a deviating relative position and orientation compared to the authentic bank note 10 (misalignment of printed image). [0029]The property to be determined according to the invention can also be the size of the features, e.g. the size of the serial numbers 3a and 3b, of the numbers of the denomination detail or also of the printed images. Furthermore, as a property there can also be determined the distance of the respective feature, e.g. the serial numbers 3a and 3b, to a reference feature, for instance the distance to an edge of the document of value or to a further security element on the bank note (not shown). The reference feature can also be a reference printed image to which the distance of a first and of a second segment of a further printed image is determined. Reference printed image could be e.g. the denomination number „1", to which the distances of the right outer edge of the printed image 7b and of the left outer edge of the printed image 7a could be determined. Furthermore, with the method according to the invention there can also be determined the color of the printed images 6a and 6b or 7a and 7b. The color of the printed images of the two partial bank notes 1a, 1b e.g. can be different, while the color within the printed images 60 or 70 of the authentic bank note 10 is uniform. For example, there can be determined properties such as the average color of the printed image, or the averaged color or spectral color distribution over one or a plurality of specific areas of the printed image, or also their spatial distribution over the printed image. If the features of the composed forgery 1 do not reproduce precisely enough those of the authentic bank note 10, the difference ascertained from the properties of the two features lies outside the set values. Thus, a recognition of a forgery is possible. [0030] For illustrating the misalignment of the printed image of the composed forgery 1, in Figure 2a is shown the printed image composed of the printed images 6a, 7a of the authentic partial bank note 1a and the printed images 6b, 7b of the forged partial bank note lb (continuous line). Additionally, the two printed images 6a' and 7a' extrapolated from the printed images 6a, 7a are sketched (dotted line), which have the absolute position to be expected for the printed images 6b, 7b. The absolute position of the printed image 6a' can be extrapolated e.g. from the straight course of the printed image 6a. Additionally, here standard data of the (authentic) printed image 60 could be used, for example for determining the length and the exact form of the extrapolated printed image 6a'. So as to be able to extrapolate the absolute position of the printed image 7a', it is necessary to exactly know the properties of the (authentic) printed image 70. The absolute position of the printed image 7a', which has the absolute position to be expected for the printed image 7b, is determined, starting out from the printed image 7a, with the aid of the standard data of the (authentic) printed image 70 of the bank note. Then a difference is ascertained by comparing the absolute position of the printed image 6a' or the printed image 7a' with the (actual) position of the printed image 6b or 7b, respectively, for example by comparing the coordinates of certain details of the printed image. The respective difference is compared with set values, which this difference is allowed to take on in the case of authentic bank notes. The set values, which for example are valid for the difference of the coordinates of the right lower corner of the printed images 6a' and 6b or 7a' and 7b, can lie e.g. within a range including the zero, the range extending [sic] in a fluctuation range around zero. This fluctuation range can be determined from the standard data of the bank note and extend e.g. symmetrically or asymmetrically around zero. In the case of the printed images 6a, 6b or 7a, 7b of the composed forgery shown in Figure 2a the difference in the y-coordinates of the lower corner is relatively large and with a high probability lies outside the range of the set values valid for this difference. [0031]The mentioned examples represent only a small selection. For the method according to the invention there can be used any properties of all sorts of features of the document of value. [0032]In Figure 2b is shown a further example for a misalignment of a printed image in a different composed forgery, with the help of which the second method according to the invention is represented. The two characters „1" and „0" belong e.g. to a denomination detail through which extends a separating line 8. The character „1" is formed by a printed image 9, which in this case is composed of two partial printed images 9a and 9b. For checking the composed forgery, in a first step a feature of the bank note and/or of the composed forgery is identified. This step e.g. can consist in recognizing the characters of the denomination detail, in particular of the printed image 9 as character „1", with the aid of an OCR method. Then for example the form and the size of the printed image 9 is determined. By comparing the form and size of the printed image 9 with the form and size to be expected in the case of an authentic bank note a difference is ascertained. The form and size to be expected is determined with the aid of standard data which are valid for authentic bank notes of the denomination of the composed forgery. The standard data contain for example reference characters which are chosen from a set of reference characters on the basis of the previously determined currency and denomination of the bank note or composed forgery to be checked. For ascertaining the difference, for example, intensity values at various image points of the printed image 9 or of the reference character „1" can be subtracted from each other. For evaluating, the ascertained difference is compared with set values. In the evaluation it is examined for example with the aid of an OCV method how good the match is between the printed image 9 and the reference character „1" with respect to its form and size. It is examined whether the ascertained difference lies within the range of set values or whether it exceeds the maximum acceptable deviations of the character form and character size. Patent Claims 1. A method for checking documents of value (1, 10), in particular for recognizing forged documents of value (1, 10), with the steps: determining at least one property of a first feature (3a, 4a, 5a, 6a, 7a) of a document of value (1, 10), determining at least one property of a second feature (3b, 4b, 5b, 6b, 7b) of the document of value (1, 10), ascertaining a difference by comparing at least one of the properties of the first feature (3a, 4a, 5a, 6a, 7a) with at least one of the properties of the second feature (3b, 4b, 5b, 6b, 7b), and/ or by comparing at least one property to be expected for the second feature (3b, 4b, 5b, 6b, 7b) with at least one of the properties of the second feature (3b, 4b, 5b, 6b, 7b), evaluating the difference by comparing the difference with set values, the set values being determined from standard data which are valid for the document of value (1, 10). 2. The method according to claim 1, characterized in that the property to be expected for the second feature (3b, 4b, 5b, 6b, 7b) is determined with the aid of the standard data, in particular starting out from the first feature (3a, 4a, 5a, 6a, 7a). 3. The method according to one or more of the previous claims, characterized in that the property to be expected for the second feature (3b, 4b, 5b, 6b, 7b) is determined by extrapolating, in particular by extrapolating from at least one of the properties of the first feature (3a, 4a, 5a, 6a, 7a) and/ or from further measuring data of the document of value (1, 10). 4. The method according to one or more of the previous claims, characterized in that the properties of the first feature (3a, 4a, 5a, 6a, 7a) and of the second feature (3b, 4b, 5b, 6b, 7b) are determined with the same sensor. 5. The method according to one or more of the previous claims, characterized in that the first feature (3a, 4a, 5a, 6a, 7a) has at least one segment of a security element of the document of value (1, 10) or of an edge of the document of value (1, 10) and/ or the second feature (3b, 4b, 5b, 6b, 7b) has at least one segment of a security element of the document of value (1, 10) or of an edge of the document of value (1, 10). 6. The method according to one or more of the previous claims, characterized in that the properties of the first feature (3a, 4a, 5a, 6a, 7a) and of the second feature (3b, 4b, 5b, 6b, 7b) and/ or the properties of the first feature (3a, 4a, 5a, 6a, 7a) and of the second feature (3b, 4b, 5b, 6b, 7b) and the property to be expected for the second feature (3b, 4b, 5b, 6b, 7b) relate to the position and/or the shape of the respective feature. 7. The method according to one or more of the previous claims, but at least according to claim 6, characterized in that the shape is a color and/ or a spectral and/ or spatial color distribution and/or a size and/or a form of the respective feature. 8. The method according to one or more of claims 6 to 7, characterized in that the position is an absolute position and/ or a relative position, in particular a distance of the respective feature relative to a reference feature and/ or an orientation of the respective feature relative to a reference feature. 9. The method according to claim 8, characterized in that the reference feature is at least one edge of the document of value (1, 10) and/ or at least one security element and/ or at least one segment of at least one reference printed image. 10. The method according to one or more of the previous claims, characterized in that the first feature (3a, 4a, 6a, 7a) is at least one segment of at least one predetermined printed image and/or the second feature (3b, 4b, 6b, 7b) is at least one segment of at least one predetermined printed image. 11. The method according to claim 10, characterized in that the first feature (3a, 4a, 6a, 7a) and the second feature (3b, 4b, 6b, 7b) are segments of the same predetermined printed image or of different predetermined printed images which have preferably been produced with the same printing method, in particular within the same printing step. 12. The method according to claim 10 or 11, characterized in that at least one of the predetermined printed images has at least one segment of a design element or of an image motif or of a pattern or of a writing or of a character, in particular of a number, of a letter or of a graphic character. 13. The method according to claim 9 as well as according to one or more of claims 10 to 12, characterized in that the reference printed image is a further segment of the predetermined printed image of the first feature (3 a, 4a, 6a, 7a) and/ or of the second feature (3b, 4b, 6b, 7b) or that the reference printed image is a segment of a further printed image which has preferably been produced with the same printing method, in particular within the same printing step, as the predetermined printed image of the first feature (3a, 4a, 6a, 7a) and/or of the second feature (3b, 4b, 6b, 7b). 14. The method according to one or more of the previous claims, characterized in that the method is carried out for at least one further property of the first feature (3a, 4a, 5a, 6a, 7a) and of the second feature (3b, 4b, 5b, 6b, 7b). 15. The method according to one or more of the previous claims, characterized in that the distance between the first feature (3a, 4a, 5a, 6a, 7a) and the second feature (3b, 4b, 5b, 6b, 7b) in a direction amounts to at least half of the extent of the document of value (1, 10) in this direction. 16. The method according to one or more of the previous claims, characterized in that the method is carried out for at least one further first feature (3a, 4a, 5a, 6a, 7a) and at least one further second feature (3b, 4b, 5b, 6b, 7b) of the document of value (1, 10), the method preferably being carried out for both features of the front side of the document of value (1, 10) and features of the back side of the document of value (1, 10). 17. A method for checking documents of value (1, 10), in particular for recognizing forged documents of value (1, 10), with the steps: identifying a feature (9) of a document of value (1, 10), determining a size and/ or a form and/ or a structure of the feature (9), ascertaining a difference by comparing the size and/ or the form and/ or the structure of the feature (9) with at least one size and/ or form and/ or structure to be expected for the feature (9), evaluating the difference by comparing the difference with set values, the set values being determined from standard data which are valid for the document of value (1, 10). 18. The method according to claim 17, characterized in that the size and/or form and/ or structure to be expected for the feature (9) is determined with the aid of the standard data. 19. The method according to one or more of claims 17 to 18, characterized in that the feature (9) is at least one segment of at least one printed image, in particular at least one segment of a pattern or of a writing or of a character, in particular of a number, of a letter or of a graphic character. 20. The method according to one or more of the previous claims, characterized in that the standard data are ascertained or have been ascertained with the help of a multitude of documents of value (1, 10), in particular with the help of a multitude of documents of value (1, 10) of the type of the document of value (1,10). 21. The method according to one or more of the previous claims, characterized in that in the evaluation the match of the difference with the set values is examined, and a corresponding match degree is determined. 22. The method according to one or more of the previous claims, characterized in that the method is used for checking the fitness of the document of value (1, 10), in particular for checking as to gluings or tears, and/or for checking the authenticity of the document of value (1, 10), in particular for checking as to the presence of separating lines (2, 8) at which the document of value (1, 10) is composed. 23. The method according to one or more of the previous claims, characterized in that the respective feature or the respective features are optically detected, in particular in the visible and/ or in the not-visible spectral region, and/ or are magnetically detected. 24. An apparatus for carrying out the method according to one or more of the previous claims, wherein the apparatus preferably has at least one sensor for checking documents of value (1, 10). The invention relates to a method and a device for identifying forged value documents, for example composed forgeries consisting of parts of different value documents. According to the method, a characteristic of a first feature of the value document to be verified is determined in a first step and a characteristic of a second feature of the value document to be verified is determined in a second step. In an additional step, the characteristics of both features are compared and the difference is determined. Alternatively, in the additional step an anticipated characteristic for the second feature is compared to the characteristic of the second feature and the difference is determined. Verification then takes place as to whether the determined difference lies in a predetermined value range that contains target values which are defined as acceptable for said difference.

Full Text

Method and apparatus for checking documents of value
[001] The invention relates to a method for checking documents of value, in
particular for recognizing forged documents of value, and an apparatus for carrying
out the method. The forged documents of value to be recognized are composed
forgeries which are composed of parts of different documents of value. The composed
forgeries can be composed of parts of authentic and of forged documents of value, but
composed forgeries that are composed exclusively from parts of authentic documents
of value are also known.
[002] From the prior art there are known various methods for recognizing forged
bank notes. For recognizing composed forgeries, for example methods for detecting
adhesive strips are employed by which the individual parts of the composed forgery
are glued together. For the authenticity check, moreover, the bank notes for example
are checked as to properties which distinguish the authentic bank note paper from
normal paper, for example as to its fluorescence properties. Many composed forgeries
partly consist of authentic and partly of forged paper which, however, has
fluorescence properties similar to those of authentic bank notes. Furthermore,
forgeries are composed which exclusively consist of parts of authentic bank notes.
With the previous methods it is not possible to recognize those composed forgeries
which deliver measuring signals, e.g. fluorescence signals, comparable to those of
authentic bank notes.
[003] It is therefore an object of the invention to provide a possibility for recognizing
forged documents of value which are composed of a plurality of parts of authentic or
of authentic and forged documents of value.
[004] This problem is solved by the features of the independent claims. In claims
dependent thereon advantageous embodiments and developments of the invention are
specified.

[005] With the method according to the invention in a first step a property of a first
feature of the document of value to be checked and in a second step a property of a
second feature of the document of value is determined. In a further step the properties
of the first and second feature are compared to each other and therefrom a difference
between the first and the second feature is ascertained. Preferably, the properties to be
compared are the same properties, such as e.g. the color of the first feature and the
color of the second feature. The property of the first and the one of the second feature
are preferably determined with the same sensor. Then the ascertained difference is
evaluated. In the evaluation it is examined whether the ascertained difference lies
within a predetermined range which includes the set values set to be acceptable for the
difference. This range is determined for example with the aid of standard data valid
for the document of value. The limits of the range of the set values are given by one or
a plurality of threshold values, such as e.g. maximum values or also minimum values
for the ascertained difference. The standard data contain e.g. information about
reference features and about maximum acceptable deviations from the reference
features.
[006] The further step, which serves for ascertaining a difference, can also consist in
- as an alternative to the direct comparison of the properties of the two features - that a
property to be expected for the second feature is compared to the property of the
second feature. The property to be expected for the second feature can be determined
by extrapolation, for example by an extrapolation from the property or the properties
of the first feature and/ or from further measuring data of the document of value. The
extrapolation can also be carried out using standard data valid for the bank note, for
example starting out from the first feature. Thus, with printed image features having
an individual structure the extrapolation of the position of a second printed image
segment can be effected for example from the position of a first printed image
segment with the aid of a reference printed image which was determined on the basis
of a multitude of corresponding printed images of authentic bank notes. The
extrapolation can also be effected from further measuring data of the bank note, e.g.
from the measured spatial arrangement of further features with the aid of the standard
data.

[007] The property to be expected can also be determined with the aid of the standard
data valid for the document of value. The difference ascertained in doing so in general
differs from the difference ascertained with the first-mentioned variant of the method.
For evaluating, the ascertained difference is compared to set values, the set values also
being determined from the standard data for the document of value. These set values,
too, in general differ from the set values of the first-mentioned variant of the method.
In the evaluation it is examined whether the ascertained difference lies within the
range of set values which contains e.g. the maximum acceptable deviations of the
property of the second feature from the property to be expected for the second feature.
In the evaluating step it is also possible to examine a match of the difference with the
set values and to determine a corresponding match degree.
[008] The standard data which form the basis for the range of set values and/or for
the determination of the second feature's property to be expected can be ascertained
with the help of a multitude of documents of value or could have been ascertained in
the run-up to the check. Preferably, the standard data are ascertained with the help of
documents of value of the same type as the document of value to be examined, in the
case of bank notes for example with the help of bank notes of the same denomination.
The standard data contain, for example, information about the properties of reference
features and about the maximum acceptable deviations of the properties of the
features checked with the method according to the invention from the properties of the
reference features. For the different types of documents of value the respective
specific standard data can be stored. The standard data can be chosen with the help of
the type of the document of value, for example the currency and denomination of a
bank note. For choosing the standard data valid for the document of value to be
checked, the type of document of value is for this purpose identified for example
before the document of value is checked. In the case of bank notes this identifying can
be e.g. determining the denomination previous to the method according to the
invention.
[009] The first and second feature to be compared with the method according to the
invention can be of the same type but can also be features of a different type. The

method can be carried out for a single property of the two features, but it can also be
carried out for a plurality of properties of the two features. Furthermore, the method
can also be carried out for further first and second features. It is especially
advantageous to carry out the method for both features of the front side and features
of the back side of the document of value, i.e. at least one feature pair of the front side
and at least one feature pair of the back side is checked.
[0010]In an embodiment the first and/or the second feature is at least one segment of
at least one predetermined printed image. The first and second feature can be for
example segments of the same predetermined printed image. But they can also be
segments of different predetermined printed images which preferably have been
produced with the same printing method (e.g. offset printing, engraving, ...), in
particular within the same printing step. The printed image areas produced with the
same printing method, in particular within the same printing step, have a very well
defined relative position to each other which can hardly be exactly reproduced when
individual parts of the documents of value are subsequently joined together.
[0011]The printed images to be compared as to their properties can have one or a
plurality of segments of one or a plurality of design elements, such as e.g. symbols or
geometric elements, and/or of one or a plurality of image motifs, such as e.g. pictures
of persons, buildings, animals, objects or the like, or, likewise, one or a plurality of
segments of patterns, writings or characters, such as e.g. graphic characters, numbers,
letters. In a further embodiment the first and/or second feature, additionally or
alternatively to the printed image or printed segment, has at least one segment of a
security element or of an edge of the document of value.
[0012]The property with respect to which the first and the second feature are
compared preferably relates to the position and/ or the shape of the respective feature.
In the embodiment with the alternative further step the property of the first and second
feature as well as the property to be expected for the second feature relates to the
position and/or the shape of the respective feature. Properties which relate to the
shape of the respective feature are for example a color, a spectral and/ or spatial color
distribution, a size or a form of the respective feature. Properties which relate to the

position of the respective feature are for example an absolute position, a relative
position, in particular a distance relative to a reference feature or an orientation
relative to a reference feature. As a reference feature for example an edge of the
document of value, a security element, a reference printed image or also a segment of
the same can be used. The reference printed image can be e.g. a further segment of the
predetermined printed image of the first and/ or of the second feature, but it can also
be a segment of a further printed image which has been produced preferably with the
same printing method, in particular within the same printing step, as the printed image
of the first and/ or second feature.
[0013]With authentic bank notes the relative positions and orientations of the features
are very precisely defined, in particular in case of printed features which have been
produced within the same printing process or even within the same printing step.
Thus, the method according to the invention is preferably carried out in such a way
that as features certain printed image areas are used from which a difference
concerning the position is ascertained.
[0014]For evaluating, the difference ascertained with the method according to the
invention is compared with the particular set values, and it is checked whether the
ascertained difference lies within the range of the particular set values. In the case of
an authentic document of value the ascertained difference should lie within the range
of the particular set values. In the case of a composed forgery, in contrast, there is a
high probability that at least a number of features at least in some of their properties
have differences which do not lie within the range of the particular set values. Due to
the subsequent composing of the composed forgery it is to be expected that in doing
so e.g. the difference of the absolute position of some printed image areas deviates by
more than the acceptable deviation from the reference values of the authentic
documents of value.
[0015]The evaluation of the ascertained difference is effected for example for
determining the authenticity of the document of value or also for the fitness check of
the document of value, for instance with regard to the presence of tears or gluings.
With the aid of the method according to the invention not only composed forgeries

can be detected but also authentic documents of value which, however, for the
purpose of „repairing" were partly or completely glued together. This gluing together
may also lead to differences in the position and/ or the shape of features which do not
correspond to the set values. For example, printed images through which runs a tear
and which subsequently were joined together can be changed such that the differences
in position ascertained in the printed image lie outside the range of the set values.
With torn authentic documents of value the differences in the position of features may
also deviate from those of intact authentic documents of value and can be detected
with the method according to the invention.
[0016]In general, a composed forgery can have a separating line or a plurality of
separating lines at which it is composed. Though the position of these separating lines
on the composed forgery are not foreseeable, but for recognizing composed forgeries
it is advantageous to selectively use such features which with a high probability are
disposed on different partial bank notes of the composed forgery. For this reason the
method according to the invention is preferably carried out on the basis of the
properties of such features which are disposed on the document of value as far apart
from each other as possible. Preferably, the first and second feature are chosen such
that the distance between these features in one direction, in particular in the direction
along the connecting line of the features, extends over a substantial part of the
document of value, but at least amounts to half of the extent of the document of value.
Furthermore, it is advantageous to compare further first and further second features
with each other whose connecting line encloses with the connecting line of the first
and second feature a large angle, e.g. an angle of at least 20°. In a development of the
method the position of the separating lines of a plurality of known composed forgeries
can also be detected, and therefrom a probability for the position of the separating
lines on the document of value to be checked can be determined. In dependence
thereon e.g. the features to be used for the method according to the invention can be
chosen such that most of the known separating lines extend between these features.
[0017] A second method according to the invention for recognizing forged documents
of value, in particular for recognizing composed forgeries, can run in a way as

follows: For checking the bank note or the composed forgery, in a first step a feature
of a document of value is identified, in particular a feature of a bank note or of a
composed forgery. As a feature, preferably at least one segment of at least one printed
image is chosen, e.g. at least one segment of a pattern, or of a writing, or of a
character, in particular of a number, of a letter, or of a graphic character. For
identifying the feature, a recognition of a character string can be carried out with the
aid of an OCR method. Then the size and/ or form and/ or structure of the feature is
determined, preferably with a higher accuracy than during the identification. For
ascertaining a difference the determined size and/ or form and/ or structure of the
feature is compared with at least one size and/ or form and/ or structure to be expected
for the feature. This size and/ or form and/ or structure to be expected for the feature
is determined with the aid of the standard data valid for the document of value. These
standard data can be chosen with the help of the previously determined type of the
document of value, for example the currency and denomination of a bank note, and
contain for example information about the properties of reference features and about
the maximum acceptable deviations therefrom. For evaluating, the ascertained
difference is compared to set values, the set values also being determined from the
standard data for the document of value. In doing so, for example, it is examined
whether the ascertained difference lies within the range of the set values which
contains the maximum deviations with respect to the size and/ or form and/ or
structure of the feature, or whether it exceeds the maximum deviations. In the
evaluation it is examined, for example, how good the match is between the measured
properties of the feature and the set values therefor, and a corresponding match degree
is determined. In the steps where the difference is ascertained and evaluated e.g. an
OCV method can be employed.
[0018]In the methods according to the invention the features are detected, for
example, optically, in particular in the visible or in the not visible spectral region, or
magnetically. A further aspect of the invention relates to the apparatus, which is used
for carrying out the method. Such an apparatus can have a sensor for checking
documents of value, in particular bank notes, e.g. an image sensor for detecting

optical features of the document of value or a magnet sensor for detecting magnetic
features.
[0019]In the following the invention is described by way of example with reference to
the accompanying Figures.
[0020]Fig. 1a shows a schematic representation of an authentic bank note,
[0021]Fig. 1b shows a schematic representation of a composed forgery which is
composed of two parts,
[0022]Fig. 2a shows a misalignment of a printed image in a composed forgery,
[0023]Fig. 2b shows a segment from the denomination detail of a composed
forgery, the separating line of which runs through the numbers of
the denomination detail.
[0024]Figure 1a schematically represents the back side of an authentic bank note 10.
Here are shown, by way of example, some printed images 40, 60, 70 which have been
produced e.g. by means of engraving, offset printing or further printing methods, as
well as a serial number 30 printed at two positions onto the bank note 10. Of course,
further printed images and security elements can be additionally contained on or in the
banknote 10.
[0025]Figure lb schematically shows a composed forgery 1 which is composed of two
parts, for example of a left authentic partial bank note 1a and a right forged partial
bank note lb. At a separating line 2 running approximately perpendicular through the
composed forgery 1 the two partial bank notes la and lb are glued together. In this
example, however, the gluing together of the composed forgery 1 did not succeed in
disposing the two lower edges 5a and 5b such that they extend precisely in one line.
The lower edges 5a and 5b are slightly inclined to each other and at the intersection
point with the separating line 2 they can also be a bit misaligned in vertical direction.
For a better illustration the inclination is drawn relatively strong in Figure 1b. For the
method according to the invention, however, it is sufficient that the two partial bank
notes la and lb are not quite precisely aligned to each other. A slight deviation is

sufficient which does not have to be given with respect to the edges of the bank note
but may be given with respect to the property of any feature which with a
corresponding accuracy can be metrologically detected.
[0026]For checking a bank note 10 e.g. an image sensor is used with which, among
other things, the lower edge 50 of the bank note 10 is detected. From the image data
can be determined for example the absolute position of a left section of the lower edge
50, which e.g. extends from the left corner over a certain distance to the right, and
analogously the absolute position of a right section of the lower edge which extends
from the right corner over a distance to the left. In doing so, in a composed forgery 1
e.g. the absolute positions of the lower edge sections 5a, 5b are determined. Then a
difference is ascertained by comparing these absolute positions, for example by
subtraction of the coordinates of the two corner points or of a plurality of single image
points of the two lower edge sections or of two center points of the two lower edge
sections ascertained from the image data. For evaluating the difference, the difference
is compared with set values which result from standard data valid for the document of
value which for example have been determined in the run-up to the process with the
help of authentic bank notes of the same denomination. The ascertained difference of
the absolute positions for an authentic bank note 10 lies with a very high probability
within the range of the set values for this position difference. In a composed forgery 1
whose lower edge sections 5a, 5b do not extend in a straight line the difference
exceeds - in the case of a correspondingly large deviation - the limits of the range of
the set values for this position difference. The result of the evaluation of the
difference can be used in this way for checking a bank note, in particular for
recognizing a composed forgery. The evaluation result, however, cannot only be used
as an authenticity criterion but also for checking the fitness of the bank note, for
example for identifying a torn bank note or a bank note glued together with adhesive
strips. In these bank notes, too, a displacement of the positions of features may occur,
when e.g. bank note parts have been glued together a bit obliquely or when the parts
of a torn bank note are a bit displaced against each other during the transport of the
bank note.

[0027] In an alternative embodiment a difference of the absolute positions of the right
and left lower edge section can also be ascertained in a way as follows: From a first of
the absolute positions of the two lower edge sections, e.g. of the left lower edge
section 5a, the absolute position to be expected for the second lower edge section, 5b,
is determined. This can be effected for example by extrapolating the course of the first
lower edge section 5a. Then a difference between the extrapolated position to be
expected of the second lower edge section 5b and the actual position of the second
lower edge section 5b is ascertained and compared with set values for the difference.
Since here the match between an actual property and the set value for this property is
examined, the difference and the set value in this embodiment are relatively small
compared to those of the previously described embodiment.
[0028]The deviation from the set values, which results from the composing of the
partial bank notes la, 1b, could be recognized by many features of the composed
forgery 1. With the method according to the invention in the first and second
procedure step there can be compared e.g. the properties described in the following of
the following features: The serial numbers 3a and 3b have a relative position and a
relative orientation to each other which deviates from the relative position and
orientation of the two serial numbers 30 of the authentic bank note. Since the
separating line 2 extends through between the numbers of the denomination ,,1000" of
the bank note, the printed images of the numbers 4a and 4b also have a deviating
relative position and orientation compared to the respective characters of the authentic
bank note 10. The separating line 2 furthermore extends through the printed images
shown above the denomination detail, so that the partial printed images 6a and 6b or
7a and 7b, too, have a deviating relative position and orientation compared to the
authentic bank note 10 (misalignment of printed image).
[0029]The property to be determined according to the invention can also be the size of
the features, e.g. the size of the serial numbers 3a and 3b, of the numbers of the
denomination detail or also of the printed images. Furthermore, as a property there
can also be determined the distance of the respective feature, e.g. the serial numbers
3a and 3b, to a reference feature, for instance the distance to an edge of the document

of value or to a further security element on the bank note (not shown). The reference
feature can also be a reference printed image to which the distance of a first and of a
second segment of a further printed image is determined. Reference printed image
could be e.g. the denomination number „1", to which the distances of the right outer
edge of the printed image 7b and of the left outer edge of the printed image 7a could
be determined. Furthermore, with the method according to the invention there can also
be determined the color of the printed images 6a and 6b or 7a and 7b. The color of the
printed images of the two partial bank notes 1a, 1b e.g. can be different, while the
color within the printed images 60 or 70 of the authentic bank note 10 is uniform. For
example, there can be determined properties such as the average color of the printed
image, or the averaged color or spectral color distribution over one or a plurality of
specific areas of the printed image, or also their spatial distribution over the printed
image. If the features of the composed forgery 1 do not reproduce precisely enough
those of the authentic bank note 10, the difference ascertained from the properties of
the two features lies outside the set values. Thus, a recognition of a forgery is
possible.
[0030] For illustrating the misalignment of the printed image of the composed forgery
1, in Figure 2a is shown the printed image composed of the printed images 6a, 7a of
the authentic partial bank note 1a and the printed images 6b, 7b of the forged partial
bank note lb (continuous line). Additionally, the two printed images 6a' and 7a'
extrapolated from the printed images 6a, 7a are sketched (dotted line), which have the
absolute position to be expected for the printed images 6b, 7b. The absolute position
of the printed image 6a' can be extrapolated e.g. from the straight course of the printed
image 6a. Additionally, here standard data of the (authentic) printed image 60 could
be used, for example for determining the length and the exact form of the extrapolated
printed image 6a'. So as to be able to extrapolate the absolute position of the printed
image 7a', it is necessary to exactly know the properties of the (authentic) printed
image 70. The absolute position of the printed image 7a', which has the absolute
position to be expected for the printed image 7b, is determined, starting out from the
printed image 7a, with the aid of the standard data of the (authentic) printed image 70
of the bank note. Then a difference is ascertained by comparing the absolute position

of the printed image 6a' or the printed image 7a' with the (actual) position of the
printed image 6b or 7b, respectively, for example by comparing the coordinates of
certain details of the printed image. The respective difference is compared with set
values, which this difference is allowed to take on in the case of authentic bank notes.
The set values, which for example are valid for the difference of the coordinates of the
right lower corner of the printed images 6a' and 6b or 7a' and 7b, can lie e.g. within a
range including the zero, the range extending [sic] in a fluctuation range around zero.
This fluctuation range can be determined from the standard data of the bank note and
extend e.g. symmetrically or asymmetrically around zero. In the case of the printed
images 6a, 6b or 7a, 7b of the composed forgery shown in Figure 2a the difference in
the y-coordinates of the lower corner is relatively large and with a high probability
lies outside the range of the set values valid for this difference.
[0031]The mentioned examples represent only a small selection. For the method
according to the invention there can be used any properties of all sorts of features of
the document of value.
[0032]In Figure 2b is shown a further example for a misalignment of a printed image
in a different composed forgery, with the help of which the second method according
to the invention is represented. The two characters „1" and „0" belong e.g. to a
denomination detail through which extends a separating line 8. The character „1" is
formed by a printed image 9, which in this case is composed of two partial printed
images 9a and 9b. For checking the composed forgery, in a first step a feature of the
bank note and/or of the composed forgery is identified. This step e.g. can consist in
recognizing the characters of the denomination detail, in particular of the printed
image 9 as character „1", with the aid of an OCR method. Then for example the form
and the size of the printed image 9 is determined. By comparing the form and size of
the printed image 9 with the form and size to be expected in the case of an authentic
bank note a difference is ascertained. The form and size to be expected is determined
with the aid of standard data which are valid for authentic bank notes of the
denomination of the composed forgery. The standard data contain for example
reference characters which are chosen from a set of reference characters on the basis

of the previously determined currency and denomination of the bank note or
composed forgery to be checked. For ascertaining the difference, for example,
intensity values at various image points of the printed image 9 or of the reference
character „1" can be subtracted from each other. For evaluating, the ascertained
difference is compared with set values. In the evaluation it is examined for example
with the aid of an OCV method how good the match is between the printed image 9
and the reference character „1" with respect to its form and size. It is examined
whether the ascertained difference lies within the range of set values or whether it
exceeds the maximum acceptable deviations of the character form and character size.

Patent Claims
1. A method for checking documents of value (1, 10), in particular for recognizing
forged documents of value (1, 10), with the steps:
determining at least one property of a first feature
(3a, 4a, 5a, 6a, 7a) of a document of value (1, 10),
determining at least one property of a second feature
(3b, 4b, 5b, 6b, 7b) of the document of value (1, 10),
ascertaining a difference by comparing at least one
of the properties of the first feature (3a, 4a, 5a, 6a, 7a) with at least one
of the properties of the second feature (3b, 4b, 5b, 6b, 7b), and/ or by
comparing at least one property to be expected for the second feature
(3b, 4b, 5b, 6b, 7b) with at least one of the properties of the second
feature (3b, 4b, 5b, 6b, 7b),
evaluating the difference by comparing the
difference with set values, the set values being determined from standard
data which are valid for the document of value (1, 10).
2. The method according to claim 1, characterized in that the property to be
expected for the second feature (3b, 4b, 5b, 6b, 7b) is determined with the
aid of the standard data, in particular starting out from the first feature (3a,
4a, 5a, 6a, 7a).
3. The method according to one or more of the previous claims, characterized in
that the property to be expected for the second feature (3b, 4b, 5b, 6b, 7b) is
determined by extrapolating, in particular by extrapolating from at least one
of the properties of the first feature (3a, 4a, 5a, 6a, 7a) and/ or from further
measuring data of the document of value (1, 10).

4. The method according to one or more of the previous claims, characterized in
that the properties of the first feature (3a, 4a, 5a, 6a, 7a) and of the second
feature (3b, 4b, 5b, 6b, 7b) are determined with the same sensor.
5. The method according to one or more of the previous claims, characterized in
that the first feature (3a, 4a, 5a, 6a, 7a) has at least one segment of a security
element of the document of value (1, 10) or of an edge of the document of
value (1, 10) and/ or the second feature (3b, 4b, 5b, 6b, 7b) has at least one
segment of a security element of the document of value (1, 10) or of an edge
of the document of value (1, 10).
6. The method according to one or more of the previous claims, characterized in
that the properties of the first feature (3a, 4a, 5a, 6a, 7a) and of the second
feature (3b, 4b, 5b, 6b, 7b) and/ or the properties of the first feature (3a, 4a,
5a, 6a, 7a) and of the second feature (3b, 4b, 5b, 6b, 7b) and the property to
be expected for the second feature (3b, 4b, 5b, 6b, 7b) relate to the position
and/or the shape of the respective feature.
7. The method according to one or more of the previous claims, but at least
according to claim 6, characterized in that the shape is a color and/ or a
spectral and/ or spatial color distribution and/or a size and/or a form of the
respective feature.
8. The method according to one or more of claims 6 to 7, characterized in that the
position is an absolute position and/ or a relative position, in particular a
distance of the respective feature relative to a reference feature and/ or an
orientation of the respective feature relative to a reference feature.
9. The method according to claim 8, characterized in that the reference feature is
at least one edge of the document of value (1, 10) and/ or at least one
security element and/ or at least one segment of at least one reference printed
image.

10. The method according to one or more of the previous claims, characterized in
that the first feature (3a, 4a, 6a, 7a) is at least one segment of at least one
predetermined printed image and/or the second feature (3b, 4b, 6b, 7b) is at
least one segment of at least one predetermined printed image.
11. The method according to claim 10, characterized in that the first feature (3a, 4a,
6a, 7a) and the second feature (3b, 4b, 6b, 7b) are segments of the same
predetermined printed image or of different predetermined printed images
which have preferably been produced with the same printing method, in
particular within the same printing step.
12. The method according to claim 10 or 11, characterized in that at least one of
the predetermined printed images has at least one segment of a design
element or of an image motif or of a pattern or of a writing or of a character,
in particular of a number, of a letter or of a graphic character.
13. The method according to claim 9 as well as according to one or more of claims
10 to 12, characterized in that the reference printed image is a further
segment of the predetermined printed image of the first feature (3 a, 4a, 6a,
7a) and/ or of the second feature (3b, 4b, 6b, 7b) or that the reference printed
image is a segment of a further printed image which has preferably been
produced with the same printing method, in particular within the same
printing step, as the predetermined printed image of the first feature (3a, 4a,
6a, 7a) and/or of the second feature (3b, 4b, 6b, 7b).
14. The method according to one or more of the previous claims, characterized in
that the method is carried out for at least one further property of the first
feature (3a, 4a, 5a, 6a, 7a) and of the second feature (3b, 4b, 5b, 6b, 7b).
15. The method according to one or more of the previous claims, characterized in
that the distance between the first feature (3a, 4a, 5a, 6a, 7a) and the second
feature (3b, 4b, 5b, 6b, 7b) in a direction amounts to at least half of the
extent of the document of value (1, 10) in this direction.

16. The method according to one or more of the previous claims, characterized in
that the method is carried out for at least one further first feature (3a, 4a, 5a,
6a, 7a) and at least one further second feature (3b, 4b, 5b, 6b, 7b) of the
document of value (1, 10), the method preferably being carried out for both
features of the front side of the document of value (1, 10) and features of the
back side of the document of value (1, 10).
17. A method for checking documents of value (1, 10), in particular for recognizing
forged documents of value (1, 10), with the steps:
identifying a feature (9) of a document of value (1,
10),
determining a size and/ or a form and/ or a structure
of the feature (9),
ascertaining a difference by comparing the size and/
or the form and/ or the structure of the feature (9) with at least one size
and/ or form and/ or structure to be expected for the feature (9),
evaluating the difference by comparing the
difference with set values, the set values being determined from standard
data which are valid for the document of value (1, 10).
18. The method according to claim 17, characterized in that the size and/or form
and/ or structure to be expected for the feature (9) is determined with the aid
of the standard data.
19. The method according to one or more of claims 17 to 18, characterized in that
the feature (9) is at least one segment of at least one printed image, in
particular at least one segment of a pattern or of a writing or of a character,
in particular of a number, of a letter or of a graphic character.
20. The method according to one or more of the previous claims, characterized in
that the standard data are ascertained or have been ascertained with the help

of a multitude of documents of value (1, 10), in particular with the help of a
multitude of documents of value (1, 10) of the type of the document of value
(1,10).
21. The method according to one or more of the previous claims, characterized in
that in the evaluation the match of the difference with the set values is
examined, and a corresponding match degree is determined.
22. The method according to one or more of the previous claims, characterized in
that the method is used for checking the fitness of the document of value (1,
10), in particular for checking as to gluings or tears, and/or for checking the
authenticity of the document of value (1, 10), in particular for checking as to
the presence of separating lines (2, 8) at which the document of value (1, 10)
is composed.
23. The method according to one or more of the previous claims, characterized in
that the respective feature or the respective features are optically detected, in
particular in the visible and/ or in the not-visible spectral region, and/ or are
magnetically detected.
24. An apparatus for carrying out the method according to one or more of the
previous claims, wherein the apparatus preferably has at least one sensor for
checking documents of value (1, 10).

The invention relates to a method and a device for identifying forged value documents, for example composed
forgeries consisting of parts of different value documents. According to the method, a characteristic of a first feature of the value
document to be verified is determined in a first step and a characteristic of a second feature of the value document to be verified is
determined in a second step. In an additional step, the characteristics of both features are compared and the difference is determined.
Alternatively, in the additional step an anticipated characteristic for the second feature is compared to the characteristic of the second
feature and the difference is determined. Verification then takes place as to whether the determined difference lies in a predetermined
value range that contains target values which are defined as acceptable for said difference.

Documents:

http://ipindiaonline.gov.in/patentsearch/GrantedSearch/viewdoc.aspx?id=0mq3F3ZzfzlgnJ5Zg6sMNA==&loc=wDBSZCsAt7zoiVrqcFJsRw==

« Previous Patent

Next Patent »

Patent Number

271706

Indian Patent Application Number

3223/KOLNP/2009

PG Journal Number

10/2016

Publication Date

04-Mar-2016

Grant Date

01-Mar-2016

Date of Filing

10-Sep-2009

Name of Patentee

GIESECKE & DEVRIENT GMBH

Applicant Address

PRINZREGENTENSTRASSE 159, 81677 MÜNCHEN GERMANY

Inventors:

#	Inventor's Name	Inventor's Address
1	STEIDL, HELMUT	RINGSTRASSE 32, 81375 MÜNCHEN GERMANY
2	SCHÜTZMANN, JÜRGEN	KARL-SHWAIGER-STRASSE 5A, 85276 PFAFFENHOFEN GERMANY
3	HOLI, NORBERT	AMSELWEG 13A, 82110 GERMERING GERMANY
4	SU, SHANCHUAN	BLUMENWEG 26, 85579 NEUBIBERG GERMANY

PCT International Classification Number

G07D7/00; G07D7/00

PCT International Application Number

PCT/EP2008/002537

PCT International Filing date

2008-03-31

PCT Conventions:

#	PCT Application Number	Date of Convention	Priority Country
1	10 2007 015 484.6	2007-03-30	Germany