Method and apparatus for inspecting value documents and method and apparatus for generating inspection parameters for use in a method of inspecting value documents

Abstract

The invention relates to a method for producing element templates for forming a template when checking value documents, in particular banknotes, of a predetermined value document type, wherein the value documents of the predetermined value document type have at least two predetermined manufacturing elements, in particular print layers and / or security elements, which partially overlap if necessary, and the element templates correspond to the manufacturing elements, wherein digital training images of training value documents of the predetermined value document type and digital reference images of reference value documents of the predetermined value document type are used, which each have pixels which are each assigned pixel data, the method having the following steps: for the training images, the positions of the manufacturing elements are each determined, and the pixels of the respective training images are assigned to the manufacturing elements using the reference images and taking into account the manufacturing element positions determined each time, for each manufacturing element, a first element template is produced which contains those pixels which are assigned to the respective manufacturing element in all training images, and for each manufacturing element, a second element template is produced which contains those pixels which are assigned to the respective manufacturing element in one or more training images. Furthermore, a checking method is described in which the element templates are used.

Description

Technical Field

[0001] The present invention relates to a method and apparatus for examining valuable documents, and a method and apparatus for generating examination parameters for use in the method of examining valuable documents. Background Technology

[0002] A valuable document is understood herein as a page-like item that, for example, represents monetary value or qualification and therefore should not be arbitrarily manufactured by unauthorized persons. Thus, a valuable document possesses the characteristic of not being easily manufactured, especially copied, and the presence of this characteristic is a sign of authenticity, i.e., manufactured by an authorized department. Valuable documents can have different types. Important examples of these types are coupons, vouchers, checks, and especially banknotes. In the case of banknotes, the type of valuable document can be further distinguished; within the scope of this invention, the type of valuable document can be given by the currency system and face value or denomination of the banknote, and, if necessary, by the period of issuance or, for example, public issuance by a central bank. While the following description pertains to banknotes, these descriptions correspondingly apply to each other type of valuable document.

[0003] Banknotes are typically manufactured by printing a substrate in multiple steps or production stages, where, in each step, a printed layer is applied to the banknote substrate. The applied printed layers should have a predetermined position relative to the banknote substrate and / or relative to each other, but this is often only roughly achieved because the printed layers may shift relative to each other during manufacturing. This results in variations in the appearance of the banknote.

[0004] When inspecting valuable documents of a predetermined type, such as verifying authenticity or condition, especially deterioration due to contamination or ink wear, a digital image of the document to be inspected is typically acquired and used for inspection. Within the scope of this disclosure, a digital image is understood to include pixels and pixel data associated with each pixel. Here, each pixel corresponds to a corresponding position in the image, and the pixel data is applied to said position. The position in the image corresponds to a position on the valuable document. Pixel data may, for example, display the brightness and color value of the corresponding pixel. During inspection, the digital image is examined to determine whether and to what extent it meets at least one predetermined inspection criterion, depending on the method used.

[0005] When verifying the standard, verification parameters are used to determine the preset characteristics of a valuable file of a preset valuable file type. In some verification methods, when verifying the standard, for example, whether a pixel or a group of pixels has the characteristics determined by the verification parameters is verified. Verification parameters may thus include, for example, a template for a valuable file of a preset valuable file type. Within the scope of this application, a template is understood as data comprising pixels of at least one segment of a digital image of a valuable file of a preset valuable file type and template pixel data respectively associated with those pixels. Here, each pixel corresponds to a corresponding position in the image, and the pixel data is applied to the image. Resolution, i.e., the number of pixels with respect to area, and the arrangement of pixels to each other, preferably correspond here to the arrangement of the digital image of the valuable file to be verified for the preset valuable file type. Here, the template pixel data may include at least one individual value and / or also include a range of values ​​considered permissible for the pixel data of the image of the valuable file to be verified, depending on the type of the template.

[0006] Therefore, when inspecting for deterioration in condition, such as contamination or ink wear, the pixel data or pixel value of the image used for valuable documents can be used as an inspection standard to check whether it is within a preset range of allowable values. If the value is below the range, contamination can be inferred; if the value is above the range, ink wear can be inferred. Otherwise, it is assumed that there is no deterioration for the pixel.

[0007] These templates are created for different types of valuable documents, in the case of banknotes, for example by currency system and value or denomination, and, if necessary, by issuance location.

[0008] Templates are typically created, for example, by averaging a large number of images from training valuable files of a predefined valuable file type. Therefore, the average value of the corresponding pixel data from the training valuable files is used as the template pixel data. In other templates, lower and upper limits for intensity can be determined for each pixel as template pixel data. These lower and upper limits can be given by the minimum and maximum intensity values ​​of the corresponding pixels in the set of training valuable files.

[0009] Therefore, the same template is used for any document of the same type. However, variations in the position of the printing layer result in inaccurate templates, especially when using interval boundaries.

[0010] When inspecting valuable documents of a preset valuable document type, the aforementioned positional changes of the printed layers cause the method to work less accurately because the template is not very accurate and / or must allow tolerances to compensate for the still permissible movement of the printed layers relative to each other and the resulting changes in pixel data as the movement changes.

[0011] Similar problems may arise when the relative positions of other manufacturing elements that at least partially determine the appearance of a valuable document are affected, for example, due to manufacturing variations. Within the scope of this application, a manufacturing element is understood to be an element of the valuable document that at least partially determines and / or influences the appearance of the valuable document in the visible or invisible (e.g., IR and / or UV) spectral range, particularly a printed layer, and is manufactured or applied independently of other elements, particularly manufacturing elements, and / or in separate manufacturing steps. Examples of manufacturing elements include printed layers applied to a substrate of the valuable document by gravure or offset printing, elements applied to a substrate by screen printing, such as images or markings with colors related to the viewing angle, or optically detectable anti-counterfeiting elements, such as watermarks or embedded security threads in the substrate, or films, if necessary, with holograms, applied to the substrate by thermoforming. For ease of readability, the following description may refer in part to printed layers. However, the description applies correspondingly to each type of manufacturing element, and particularly to the aforementioned examples of manufacturing elements. Summary of the Invention

[0012] The technical problem to be solved by this invention is to provide a method for inspecting a valuable document of a preset valuable document type having at least two manufacturing elements, and a method for providing inspection parameters for the inspection method, said inspection parameters allowing for simple and accurate inspection of such valuable documents. Furthermore, the technical problem to be solved by this invention is to provide means for performing said method.

[0013] This technical problem is solved by methods for examining valuable documents and methods for generating component templates for use in examining valuable documents, each having the features of a corresponding independent method claim, and by apparatus for examining valuable documents and apparatus for generating component templates for use in examining valuable documents, each having the features of a corresponding apparatus claim.

[0014] This technical problem is particularly solved by a method for verifying valuable documents of a preset valuable document type, each valuable document having at least two preset manufacturing elements, particularly a printed layer and / or anti-counterfeiting elements. The method is implemented using a component template for the manufacturing elements and comprises the following steps: providing a digital valuable document image of the valuable document to be verified of the preset valuable document type, the valuable document image including pixels respectively configured (or assigned) with pixel data; determining the positions of the manufacturing elements for the provided valuable document image; determining a template for the digital valuable document image using the component template for the manufacturing elements and the determined manufacturing element positions; and verifying the digital valuable document image using the determined template. A signal presenting the verification result can be generated based on the verification result. The method is hereinafter simply referred to as the verification method.

[0015] The inspection method can be performed with computer assistance. The component templates used in the inspection method are preferably previously provided, for example, stored in the memory of the data processing device or computer performing the inspection method.

[0016] The subject of this invention is also an apparatus for verifying valuable documents, particularly banknotes, of a predetermined type, wherein the valuable documents each have at least two predetermined manufacturing elements, particularly printing layers and / or anti-counterfeiting elements. The apparatus uses element templates for manufacturing the elements and includes an analysis device having at least one memory storing the element templates and an interface for providing digital images of the valuable documents. The analysis device is configured to execute the verification method according to this invention using the stored element templates and the digital valuable documents. This apparatus is also referred to below as an verification device.

[0017] Therefore, the analysis device preferably includes a processor connected to a memory via a data link, and a program memory connected to the processor via a data link. The program memory stores program instructions, and when the program is executed, at least one processor performs the verification method according to the invention using element templates and digital valuable document images stored in the memory. Within the scope of the invention, a processor is understood not only as a single processor having one or more cores, but also as a system of coupled processors. Here, the memory, interface, and processor can form part of the data processing unit of the analysis device.

[0018] Therefore, the subject of this invention is also a computer program for execution by means of a data processing device having a processor, comprising program code, wherein when the program code is executed, the processor executes the verification method according to the invention.

[0019] The subject of this invention is also a computer-readable data carrier storing a computer program according to the invention, and in particular a computer program for execution by a data processing device having a processor, the computer program comprising program code, wherein, upon execution of the program code, the processor executes the verification method according to the invention.

[0020] In the aforementioned inspection method, when providing an image of a valuable document, an image acquisition device can be used to acquire a digital image of the valuable document to be inspected and store the digital image in a memory device or memory. For this purpose, the inspection equipment may also include an image acquisition device for acquiring a digital image of the valuable document to be inspected, the image acquisition device being connected via a signal connection to an interface for providing the digital valuable document.

[0021] This verification method, in particular, allows for real-time processing of the valuable documents while verifying each individual valuable document.

[0022] Therefore, the subject of this invention is also a method for processing, particularly inspecting and / or counting and / or classifying and / or destroying, valuable documents of a predetermined valuable document type, each having two predetermined manufacturing elements, particularly a printing layer and / or anti-counterfeiting elements, wherein the valuable documents are individually conveyed through an image acquisition device for acquiring digital images of the valuable documents, acquiring digital images of the valuable documents during the respective conveyance of the valuable documents, and subsequently, preferably in real time, inspecting the digital images of the valuable documents according to the inspection method of this invention. Based on the inspection results, the valuable documents can be classified and / or destroyed.

[0023] Therefore, the subject of this invention is also an apparatus for processing, particularly inspecting and / or counting and / or classifying and / or destroying valuable documents of a predetermined type, especially banknotes, said valuable documents having at least two predetermined manufacturing elements, particularly printing layers and / or anti-counterfeiting elements. The apparatus includes an input device for inputting individual or sorted valuable documents to be processed, an output device having at least one output section for accommodating the processed valuable documents, a conveying device for conveying the individual or sorted valuable documents from the input device to the output device, and an inspection apparatus according to the invention, wherein an image acquisition device of the inspection apparatus is arranged on the conveying path and configured such that a digital image of the valuable document to be inspected, conveyed through the image acquisition device, is acquired during conveyance and provided for use in the inspection apparatus. Preferably, the apparatus can be designed for real-time inspection of the conveyed valuable documents.

[0024] This technical problem is also solved by a method for generating component templates for forming templates when verifying valuable documents of a preset valuable document type, wherein the valuable document of the preset valuable document type has at least two preset manufacturing elements, particularly printing layers and / or anti-counterfeiting elements, wherein, in the method, digital training images of training valuable documents of the preset valuable document type and digital reference images of reference valuable documents of the preset valuable document type are used, the digital training images and digital reference images each having pixels, the pixels each configured with pixel data, the method comprising the following steps: for training images, determining the positions of manufacturing elements respectively, and configuring the pixels of the corresponding training images to the manufacturing elements using the reference images and taking into account the respectively determined positions of the manufacturing elements. The method further includes, for each manufacturing element, generating a first component template, the first component template containing those pixels that are configured to the corresponding manufacturing element in all training images. Furthermore, the method includes generating a second component template, the second component template containing those pixels that are configured to the corresponding manufacturing element in one or more training images. These component templates can then preferably be stored in memory or a computer-readable storage medium. For simplicity, this method is also referred to below as an adaptation method. Preferably, the valuable document can be banknotes. Preferably, the element template pixel data associated with a pixel is determined from the pixel data of corresponding pixels for at least a portion of the training valuable document or training image.

[0025] The digital training images of the training valuable files of the preset valuable file type and the digital reference images of the reference valuable files of the preset valuable file type can be provided in advance, for example, in memory.

[0026] The optimal adaptation method can be implemented with computer assistance.

[0027] The subject of this invention is also an apparatus for generating an element template for forming a template when verifying a valuable document of a predetermined valuable document type, wherein the valuable document of the predetermined valuable document type has at least two predetermined manufacturing elements, particularly a printed layer and / or anti-counterfeiting elements, which may partially overlap if necessary. The apparatus has a memory device for storing a digital training image of the predetermined valuable document type, a reference image of a reference valuable document of the predetermined valuable document type, and preferably the generated element template. The apparatus is designed to implement the adaptation method according to the invention when using the training image and the reference image, and preferably stores the generated element template in the memory device. This apparatus is also referred to below as an adaptation device.

[0028] Therefore, the adapter device may in particular include at least one processor connected to a memory device via a data connection, and having a program memory connected to the processor via a data connection, wherein program code is stored in the program memory, and when the program code is executed, the processor performs the adapter method according to the invention using training images stored in the memory device. The processor, program memory, and memory device may be part of the data processing apparatus of the adapter device.

[0029] Therefore, the subject of this invention is also a computer program for execution by a data processing device having a processor, the computer program comprising program code in which, upon execution of the program code, the processor implements the adaptation method according to the invention.

[0030] The subject of this invention is also a computer-readable data carrier storing a computer program according to the invention, and in particular a computer program for execution by a data processing device having a processor, the computer program comprising program code, wherein, upon execution of the program code, the processor performs the adaptation method according to the invention.

[0031] In the adaptation method, only images of valuable documents of a preset valuable document type are needed. Preferably, the training and reference valuable documents can include, or are preferably, finished and / or newly printed valuable documents of the preset valuable document type, especially those of the same currency and / or denomination. In this way, suitable component templates can be determined for manufacturing components, and these component templates can be used to determine the template when using finished banknotes, without having to provide, for example, separate printed materials of the corresponding printing layers first. Furthermore, difficulties that may arise due to the fact that these printing layers partially overlap in the finished valuable document but not in separate printed materials are avoided. Overall, this allows for simpler and more reliable template generation. The images used preferably show the complete valuable document, i.e., especially not damaged or incomplete valuable documents.

[0032] In the inspection method, component templates generated or provided by other methods can be used, which are suitable for inspecting valuable documents of a preset valuable document type. However, it is preferred to use component templates generated according to the adaptation method of the present invention.

[0033] Without using the results of other steps, the steps of the method according to the invention may be performed in whole or in part in parallel or alternately or in any order.

[0034] These methods use digital images of valuable files of a preset valuable file type. Preferably, these images have the same resolution and therefore the same number of pixels for a preset imaging area. Preferably, the images display the same segment of the corresponding valuable file, particularly preferably the entire valuable file, and most preferably only the entire valuable file. Here, these pixels can be identified by their position in the image or corresponding position data, or by type and / or sequence in memory. Pixel data may include data, in the case of an image, presenting brightness and / or color and / or intensity values ​​within a preset wavelength range, such as infrared.

[0035] A valuable file of a preset valuable file type has at least two preset manufacturing elements, which are at least partially visible and thus mapped or imaged in the image. The manufacturing elements may partially overlap if necessary. Valuable files may also have other manufacturing elements, but these are not explicitly considered in the method according to the invention.

[0036] In these methods, a component template is generated or used to manufacture an element, and the component template corresponds to the manufactured element. In the sense of this disclosure, a component template, like a template, contains pixels and component template pixel data respectively associated with the pixels. The description of the template applies accordingly to the pixels. Similar to a template, the component template pixel data may include at least one individual value and / or also include data defining a range of allowed values ​​for the pixel data of the pixels for the component template, depending on the type of component template. However, the component template used to manufacture an element preferably contains fewer pixels than the digital image of the entire valuable document, because the component template only needs to correspond to the manufactured element, which does not need to extend across the entire surface of the valuable document.

[0037] The component template and the templates generated from it must be applicable to the inspection methods used for individual valuable documents that employ the template.

[0038] Component templates are assigned to specific manufacturing components. Based on the position of the manufacturing component on the current value document to be inspected, determined in the inspection method, a separate template can be generated for each value document to be inspected, using the component templates, and this template can be used in further inspections. Therefore, the templates can be determined, in particular, in real-time and thus, arguably "dynamically," during the inspection process. This significantly reduces or eliminates inaccuracies caused by fluctuations in the position of the manufacturing components. Furthermore, when comparing a value document image with a separate template, methods known per se that use templates can be used, where the permissible tolerances can be reduced or minimized. However, this is not necessary.

[0039] In the adaptation method, two component templates are generated for each manufacturing component. However, it is preferable to generate a component template for the manufacturing component only when the corresponding pixels are determined.

[0040] The first element template for the corresponding manufacturing element contains pixels that can be configured for the manufacturing element across all training images. The first element template preferably contains only these pixels. Therefore, these pixels can be considered to always correspond to at least one segment of the corresponding manufacturing element and always largely or completely define the appearance within the corresponding image region. Thus, these pixels can also be referred to as "dominant." Therefore, different first element templates do not contain the same pixels. In this respect, the pixels of the corresponding first element template can correspond to a mask through which pixels corresponding to the manufacturing element can be identified. The masks thus do not overlap.

[0041] The second element template contains pixels that are respectively assigned to the corresponding manufacturing elements in at least one training image. Therefore, the pixels of the first element template for a manufacturing element preferably form a subset of the pixels of the second element template for the same manufacturing element. However, it is possible that a pixel is included in the second element template for two different manufacturing elements. In this case, there is at least one such image or training file in which one of the manufacturing elements is offset relative to the same manufacturing element in another training file, such that it is assigned to the stated manufacturing element once and to the second manufacturing element another time. At these locations, the manufacturing elements may overlap. Because such a pixel cannot be assigned to only one of the manufacturing elements, it must be treated differently when creating a template using the element template.

[0042] The advantage of using two component templates for the same manufactured component is that, for example, in the case of a printed layer, sections of the printed layer that are always covered by other components and are invisible will not exist in the template. The following sections of the printed layer, which affect the appearance only based on the position of the printed layer in the training image, are considered in the second component template; those sections that are always present in the training image are considered in the first component template. Not only is it unnecessary to know exactly which printed layer was applied during the manufacturing process, but also portions that do not contribute to the appearance in the training image are not considered in the component template. Even a fully printed layer that only partially contributes to the appearance in the training image can be identified. This achieves more accurate component templates and therefore more accurate individual templates determined by these component templates.

[0043] The component templates generated by the method described above can preferably be used in the inspection method according to the present invention.

[0044] Therefore, in the inspection method, it is preferred that, for at least one preset manufacturing element, and preferably for multiple preset manufacturing elements, a first element template and a second element template are used. When determining the template, template pixel data is configured for pixels present in the first element template that moves corresponding to the position of the determined manufacturing element. This template pixel data is determined using the element template pixel data of the corresponding pixels in the first element template that moves corresponding to the position of the determined manufacturing element. For the remaining pixels present in at least one second element template that moves corresponding to the position of the determined manufacturing element, the template pixel data is determined using the element pixel data of the respective corresponding pixels in the second element template that moves corresponding to the position of the determined manufacturing element. Thus, if a pixel exists in the second element templates of two manufacturing elements, the template pixel data is determined based on the pixel data of the second element template. Preferably, the first element templates for different manufacturing elements do not contain the same pixels, while the second element templates for different manufacturing elements may contain the same pixels, but these pixels are configured with different element pixel data. The first and second element templates are preferably element templates determined according to the adaptation method of the present invention.

[0045] Preferably, if a pixel exists only in one of the moving element templates, the element template pixel data of the corresponding pixel in the corresponding moving first element template is used as the template pixel data. Conversely, if a pixel exists in more than one first element template, the template pixel data is determined using the element template pixel data of these moving first element templates. This method has at least two advantages: firstly, pixels contained in the first sub-templates whose characteristics are therefore determined by the corresponding manufacturing elements and are "dominant" to the appearance can be clearly identified, and corresponding, more accurate template pixel data can be configured for said pixels. This achieves a clear outline of these areas. Since, when creating individual templates, pixels or pixel data of the second element template are configured only for pixels not present in the first element templates, the template pixel data for these pixels can be determined based on the element template data of the element templates. This also achieves improved accuracy for individual templates.

[0046] Depending on the type of valuable file, manufacturing elements do not need to be extended across the entire valuable file, either individually or in combination. In adaptation methods, situations may arise where pixels cannot be assigned to one of the manufacturing elements for any training image. Consequently, even in combination, the element template may be insufficient to generate a single template for the entire valuable file.

[0047] Therefore, in the adaptation method, it is preferred that the method includes generating a background template as an additional step. The background template contains pixels that are not assigned to any of the manufacturing elements in any training image and background template pixel data respectively associated with those pixels, wherein the background template pixel data is preferably determined from the pixel data of the corresponding pixels associated with the training image. The background template is determined for a preset valuable document type and particularly contains information about surface areas of the valuable document that are not determined by the optical appearance of existing manufacturing elements such as printed layers or anti-counterfeiting elements; specifically, it is essentially background information, i.e., pixels and, if necessary, pixel data associated with those pixels. The background template can be stored like an element template.

[0048] Therefore, a background template can be preferably used in the inspection method. This background template includes pixels not included in the component template used to manufacture the component, and background template pixel data respectively associated with those pixels. When determining a template for a valuable digital document image, template pixel data can be determined separately for the remaining pixels present in the background template but not configured with a component template, using the background template pixel data of the corresponding pixels in the background template. The background template can be stored like a component template. Preferably, the background template matches the component template. Particularly preferably, the background template is generated by the adaptation method according to the present invention.

[0049] The advantage of this implementation is that the separate template for the valuable document image generated in this way also contains pixels with pixel data that are not included in any first and / or second element templates. It is particularly feasible to generate separate templates for the entire face of a valuable document.

[0050] In both methods, the location of the manufacturing element is determined. Any method can be used for this. For example, for each manufacturing element, at least one so-called anchor element and an anchor point representing its location can be set. Here, the anchor element is preferably an image segment that is representative of the manufacturing element, such as a symbol or other distinguishable printed image segment, existing in the image of a pre-defined valuable document type, i.e., particularly in training and reference images. These anchor elements can be selected automatically and / or manually. In the image, the location of the anchor element or anchor point can be determined by template matching or other related methods. To be able to determine the location of the manufacturing element, it is preferable to pre-define at least two anchor elements for each manufacturing element. Particularly preferred is that a first element template for the manufacturing elements respectively contains corresponding anchor elements for these manufacturing elements.

[0051] In principle, it is sufficient to determine, when determining the position of a corresponding manufacturing element in a corresponding training image and / or reference image, the position of the manufacturing element in the training image or reference image relative to a positional reference system, which can be particularly preferably determined by elements of a valuable document in the image, for example by the outline or edges of the valuable document in the corresponding training image or reference image, or by multiple edges or one edge and one corner point of the training image or reference image. Preferably, the determined position of the manufacturing element in the training image is relative to the position of the corresponding manufacturing element in the reference image. This is beneficial for generating templates when performing the inspection method according to the invention. As a result, position data can be generated, from which the position of the manufacturing element relative to a positional reference system, such as the previously mentioned positional reference system, can be determined. For example, these positions can be related to a predetermined corner of the valuable document and one or two adjacent edges. Element templates can preferably also implicitly or explicitly include position data, which presents the position of the manufacturing element in the reference image, especially relative to the positional reference system, for example, through their storage method. In the case where the element template does not include all pixels of the image of the valuable document, this can be achieved by determining predetermined pixels of the element template. However, the location of the anchor point can also be used.

[0052] In the inspection method, the determined manufacturing component positions are also used when generating the template. The explanation of position determination in the adaptation method applies accordingly. Therefore, when determining the position of the corresponding manufacturing component in the valuable document image, this position is preferably determined relative to a position reference system that corresponds to the position reference system used when generating the component template. The component template can then be offset or moved in accordance with the difference between the manufacturing component position in the valuable document image and the manufacturing component position based on the position data in the component template.

[0053] Understandably, whenever a background template is created or used, the description of the location also applies to the background template.

[0054] In the adaptation method, for each training image, an attempt is made to assign pixels to one of the manufacturing elements. Because their positions may change between different valuable files, the manufacturing element positions determined for the training images are taken into account during configuration.

[0055] Therefore, in a preferred embodiment, when configuring pixels of a corresponding training image to a manufacturing element, for a corresponding pixel of a corresponding training image, the degree of consistency between the pixel environment of the pixel and the pixel environment of the corresponding pixel in the reference image can be determined. This involves moving or shifting the corresponding pixel environments of the corresponding pixels in the training image and the reference image in relation to the corresponding relative positions of the manufacturing elements in the training image and the corresponding manufacturing elements in the reference image, and configuring the corresponding pixels of the training image to one of the manufacturing elements according to a preset configuration standard based on the determined degree of consistency. Within the scope of this disclosure, the pixel environment of a pixel is understood as a set of pixels, including the pixel and pixels within a preset distance relative to the pixel. The distance determines the size of the pixel environment, i.e., the number of pixels in the pixel environment.

[0056] Preferably, the configuration criteria can be preset such that the manufacturing component with the highest degree of consistency is configured. Particularly preferred is that configuration is only performed when the consistency of the manufacturing component with the highest consistency exceeds a preset minimum value. This minimum value is preferably preset such that weak and potentially random, inconspicuous consistency will not lead to configuration for the manufacturing component. This has the advantage of making the component template more accurate.

[0057] The element template pixel data can be configured with pixels determined from or corresponding to the pixel data of corresponding pixels in the training image. Specifically, the element template pixel data can be any pixel data function of the pixel data in the training image associated with the corresponding manufacturing element. For example, a pixel data point can be the average value of the pixel data, or in another embodiment, the pixel data can include the minimum and maximum values ​​of the pixel data as a lower and upper limit of an allowed range of values.

[0058] Therefore, in the preferred method of inspection, when determining the template, template pixel data is configured for the pixels of the template that exist in the component template that are moved or misaligned in accordance with the determined manufacturing component position. The template pixel data is determined based on the component template pixel data of the corresponding pixels of the component template that are moved in accordance with the determined positions for manufacturing the component.

[0059] In another embodiment of the adaptation method, the pixel configuration can be performed twice. More precisely, when configuring pixels of a corresponding training image to a manufacturing element, for a corresponding pixel of a corresponding training image, a first degree of consistency between the first pixel environment of that pixel and the first pixel environment of a pixel in a reference image can be determined, wherein the first pixel environments in the training image and the reference image are offset correspondingly to the corresponding relative positions of the corresponding manufacturing element in the training image relative to the corresponding manufacturing element in the reference image; then, in the first configuration, the corresponding pixel of the training image can be configured to one of the manufacturing elements according to a preset first configuration standard based on the determined first degree of consistency. Furthermore, for a corresponding pixel of a corresponding training image, a second degree of consistency between the second pixel environment of that pixel and the second pixel environment in the reference image can be determined, wherein the second pixel environments in the training image and the reference image are offset correspondingly to the corresponding relative positions of the corresponding manufacturing element in the training image relative to the corresponding manufacturing element in the reference image; then, in the second configuration, the corresponding pixel of the training image can be configured to one of the manufacturing elements according to a preset second configuration standard based on the determined second degree of consistency. Preferably, the second pixel environment includes more pixels than the first pixel environment, i.e., preferably larger than the first pixel environment.

[0060] As mentioned earlier, the degree of consistency can be determined in both cases. The configuration criteria can be selected as described above, but the minimum values ​​for the two criteria can be chosen to be the same or different.

[0061] If two configurations are performed, in the adaptation method, the first element template for the corresponding manufacturing element may contain only pixels that are assigned to the corresponding manufacturing element in the first configuration across all training images, and the second element template for the corresponding manufacturing element may preferably contain only pixels that are assigned to the corresponding manufacturing element in the second configuration across at least one training image. The advantage of this implementation is that the first element template contains, as precisely as possible, those pixels that are always visible on valuable files of a preset valuable file type.

[0062] The advantage of this is that the first element template corresponds fairly accurately to the dominant segment of the corresponding manufacturing element, i.e., the segment that is always visible in all training images. However, the second element template also includes such portions of the manufacturing element that are only visible at specific locations in the training images. For the recognition of these portions, it is more advantageous to use a larger pixel environment because this allows for more frequent configuration.

[0063] The component template not only contains individual pixels, but also component template pixel data associated with each pixel. Preferably, in the adaptation method, when generating the component template, first component template pixel data is configured for pixels included in a first component template, and second component template pixel data is configured for pixels included in a second component template, wherein the first and / or second component template pixel data are determined from or correspond to the pixel data of corresponding pixels in the training image. The component template pixel data can, in particular, be any pixel data function of the pixel data associated with the corresponding manufacturing component in the training image. For example, a pixel data can be the average value of the pixel data, or in another embodiment, the pixel data can include the minimum and maximum values ​​of the pixel data of the pixel as a lower and upper limit of an allowed range of values. This is correspondingly applied to the pixel data of the background template. The advantage of this method is that the pixel data of the component template is determined from multiple training files and is therefore more accurate.

[0064] Therefore, in the inspection method, the first element template and the second element template can preferably be used for at least one manufacturing element, wherein, when determining the template, the element template pixel data of the first element template that is moved in accordance with the determined manufacturing element position is configured as template pixel data for the pixels of the template that exist in the first element template that is moved in accordance with the determined manufacturing element position. For the remaining pixels that exist in at least one second element template that is moved in accordance with the determined manufacturing element position, the template pixel data is determined by using the element template pixel data of the corresponding pixels of the at least one second element template that is moved in accordance with the determined manufacturing element position. Attached Figure Description

[0065] The invention is further illustrated below with reference to the accompanying drawings. In the drawings:

[0066] Figure 1 This diagram illustrates a valuable document processing device, specifically a banknote sorting device in this example.

[0067] Figure 2 A general schematic diagram of the apparatus used to generate component templates is shown.

[0068] Figure 3A Figure B shows a general schematic diagram of valuable document images with different manufacturing component locations.

[0069] Figure 4 This shows an example of an image acquired from a valuable file of a preset valuable file type.

[0070] Figure 5The images show the pixel settings in a smaller (left) and larger (right) pixel environment, respectively. Figure 4 An example of the configuration of pixels in the valuable file.

[0071] Figure 6 Show respectively for such Figure 4 An example of the first element template of the first printed layer (left) and the second printed layer (right) of a valuable document of the valuable document type.

[0072] Figure 7 Show respectively for such Figure 4 An example of a second element template for the first printed layer (left) and the second printed layer (right) of a valuable document of the valuable document type.

[0073] Figure 8 This shows an example of a single template composed of component templates, which is used as follows: Figure 4 Valuable files of the valuable file type in the document,

[0074] Figure 9 The diagram shows a general schematic flowchart illustrating an example of an adaptation method for generating component templates.

[0075] Figure 10 The flowchart illustrates a general example of a verification method for a predefined valuable document type, where verification is performed based on... Figure 9 The method generates the component template, and

[0076] Figure 11 A rough schematic flowchart shows another example of an adaptation method for generating component templates. Detailed Implementation

[0077] Figure 1 The valuable document processing device 10 in this example is a device for processing valuable documents 12 of a preset valuable document type in the form of banknotes. It is designed to classify the valuable documents 12 according to the status determined by the valuable document processing device 10 and the authenticity of the processed valuable documents verified by the valuable document processing device 10.

[0078] The valuable document processing device has an input device 14 for inputting valuable documents 12, an output device 16 for outputting or receiving processed, i.e., classified valuable documents, and a conveying device 18 for conveying the sorted valuable documents from the input device 14 to the output device 16.

[0079] In this example, the input device 14 includes an input bin 20 for stacking valuable documents and a sorter 22 for sorting the valuable documents 12 from the stack of valuable documents in the input bin 20 and inputting the sorted valuable documents into the conveying device 18.

[0080] In this example, output device 16 includes three output sections 24, 25, and 26, into which processed valuable documents can be categorized based on intermediate processing results (in this example, verification). Each section includes a storage bin and a stacking wheel (not shown) for storing the input valuable documents. In other embodiments, an output section could be used instead of a banknote destruction device.

[0081] The conveying device 18 has at least two, in this example three, branches 28, 29 and 30, with one of output sections 24, 25 or 26 arranged at the end of each branch, and the conveying device has switches 32 and 34 at each branch that can be controlled by an execution signal, through which valuable documents can be input into the branches 28 to 30 and thus into the output sections 24 to 26 according to the execution signal.

[0082] Sensor device 38 is arranged on a transport path 36 defined by transport device 18, located between input device 14, more precisely sorter 22 in this example, and a first switch 32 located after sorter 22 in the transport direction T. Sensor device 38 acquires physical characteristics of the valuable document during its transport and generates sensor signals representing the acquisition results, which constitute sensor data. In this example, sensor device 38 includes image acquisition device 40 with an optical reflection sensor that acquires a reflected color image of the valuable document, and other sensors 42 that symbolically represent the physical characteristics of the valuable document, shown only by boxes.

[0083] The control and analysis device 46 is connected via signal connections to the sensor device 38 and the conveying device 18, particularly switches 32 and 34. In conjunction with the sensor device 38, the control and analysis device classifies valuable documents into one of preset classification levels based on signals or sensor data from the sensor device 38. For example, these classification levels can be preset based on status values ​​determined by the sensor data and authenticity values ​​also determined by the sensor data. For example, status values ​​could be "circulating" or "not circulating," and authenticity values ​​could be "forged," "suspected forgery," or "authentic." Based on the determined classification level, the control and analysis device manipulates the conveying device 18, more precisely, the switches 32 or 34, by issuing execution signals, such that the valuable documents are output to the output section of the output device 16 corresponding to the classification level determined during classification. Here, the configuration or classification to one of the preset classification levels is achieved according to preset criteria for status assessment and authenticity assessment, which depend on at least a portion of the sensor data.

[0084] To this end, the control and analysis device 46, in particular, has, in addition to having at least one corresponding interface 44 for the sensor device 38 or its sensors, especially the image acquisition device 40, a processor 48 and a memory 50 connected to the processor 48, in which at least one computer program with program code is stored. When the computer program is executed, the processor 48 controls the device and analyzes the sensor signals of the sensor device 38, especially for determining the classification level of the valuable documents being processed. Furthermore, the processor stores program code, which, when executed, controls the device and manipulates the transport device 18 in accordance with the analysis.

[0085] Interface 44, processor 50, and memory 48, or a segment of memory 48 storing corresponding computer programs and method parameters, form the analysis apparatus 47 in the sense of this disclosure. In this example, the analysis apparatus 47 analyzes the signal from the reflection sensor 40 separately from signals from other sensors. Furthermore, other segments of processor 50 and memory 48 can also perform other functions, such as controlling the valuable document processing device 10.

[0086] The reflection sensor 40 is designed to acquire an RGB reflection image of the valuable document during the transport of the valuable document by means of the transport device 18, thereby generating a digital image, which is then analyzed by the analysis device 47.

[0087] Based on the characteristics of the valuable document, the control and analysis device 46, or more precisely, the analysis device 47, determines, in partial analysis and when using sensor data from different sensors, whether the determined characteristics of the valuable document represent an indication of the status or authenticity of the valuable document. Therefore, the corresponding data can be stored in the control and analysis device 46, such as in the memory 50, for later use. Then, based on the partial analysis, the control and analysis device 46 determines a classification level as the overall result of the inspection according to a preset overall standard, and forms a classification or execution signal for the delivery device 18 based on the determined classification level.

[0088] To process valuable documents 12, the valuable documents 12, either stacked or individually placed in the input bin 20, are sorted by the sorter 22 and individually input into the conveyor 18, which transports the sorted valuable documents 12 through the sensor device 38. The sensor device collects characteristics of the valuable documents 12, forming sensor signals that reflect or represent the characteristics of the corresponding valuable documents. The control and analysis device 46 collects the sensor signals or data, determines the classification level of the corresponding valuable document based on these signals or data (in this example, a combination of authenticity level and status level), and manipulates the switch accordingly to transport the valuable document to the output section corresponding to the determined classification level.

[0089] The analysis device 47 and the image acquisition device 40 together form an example of an inspection device for inspecting valuable documents of a preset valuable document type, wherein the valuable documents have two preset manufacturing elements, particularly a printing layer and / or anti-counterfeiting elements. Therefore, the computer program contains instructions to execute a method for inspecting valuable documents of a preset valuable document type, particularly banknotes, wherein the valuable documents have two preset manufacturing elements, particularly a printing layer and / or anti-counterfeiting elements, the method using element templates for manufacturing elements, which are generated, in particular, by the adaptation method described below. In the inspection method, a digital image of the valuable document to be inspected is acquired by the reflection sensor 40, and the digital image is provided in a corresponding segment of the memory 50 in the analysis device 47. For the provided image of the valuable document, the position of the manufacturing element is determined, and a template for the digital image of the valuable document is determined in real time using at least two element templates for manufacturing elements and the determined position of the manufacturing element, and in this example, a background template. After this, the digital image of the valuable document is inspected using the determined template.

[0090] exist Figure 2The adapter device, schematically shown, is a device 70 for generating component templates to form templates when verifying valuable documents of a preset valuable document type. This device provides component templates, wherein the valuable document of the preset valuable document type has at least two preset manufacturing elements, particularly printed layers and / or anti-counterfeiting elements, which may partially overlap. The device is a data processing apparatus having a memory device 72 for storing digital training images of the preset valuable document type, reference images of reference valuable documents of the preset valuable document type, and preferably generated component templates. The device 70 is designed to perform the adapter method described below using the training images and reference images, and to store the generated component template in the memory device 72. For this purpose, the device may have at least one processor 74 connected to the memory device 72 via a data link, and a program memory 76 connected to the processor 74 via a data link, storing program code. During the execution of the program code, the device, via the processor 74, performs the adapter method described below using the training images stored in the memory device 72. In other embodiments, the program memory 76 may also be formed from segments of the memory device 72. In addition, the adapter device 70 may have a data interface not shown in the figure, such as a network card, through which the generated component template stored in the memory device 72 can be transmitted to other devices.

[0091] exist Figure 3A The image schematically illustrates an example of a digital image 60 of a pre-defined valuable document type 12, which has two pre-defined manufacturing elements 62 and 64 in the form of printed layers. The valuable document serves as a reference valuable document; therefore, the valuable document image is a reference image.

[0092] Digital image 60 contains pixels 66, which in this example are arranged on a square grid and represent positions in the digital image, and thus positions on the imaged valuable document. Here, the image is preprocessed such that only the valuable document 12 is displayed on the entire surface, i.e., the edges of the valuable document in the image extend along the corresponding edge pixels. In this example, this preprocessing is performed on all images, thus ensuring that each image contains a corresponding view.

[0093] Two regions 62 and 64, characterized by different patterns, are shown. These regions show different manufacturing elements, in this example, printed layers, which do not overlap in this example.

[0094] Figure 3BA digital image 60T of another valuable document of a preset valuable document type is shown schematically. In the valuable document, the manufacturing element 64, or the corresponding printed layer in this example, is positioned relative to the edge of the valuable document and therefore relative to the image. Figure 3A The valuable file in the document is shifted by a vector V. Therefore, manufacturing element 64 is also related to manufacturing element 62. Figure 3A They are positioned differently in the middle. Therefore, those areas that show or correspond to manufacturing elements 62 and 64 are relative to... Figure 3A The areas are misaligned. Furthermore, the printed layers overlap, so corresponding pixels 68 show different appearances, symbolized by different shading lines in the accompanying drawings.

[0095] Figure 4 The example shown is a training image of a valuable document in the form of a banknote, more specifically, a 5-euro banknote. In this example, the first printing layer (mainly "5 Euro" and "BCE ECB EZB...") produced by intaglio printing and the second printing layer (mainly stars and small rings) produced by offset printing can be seen in the acquired image.

[0096] To generate a component template, the following exemplary adaptation method is used. Figure 9 The text provides a general illustration of this adaptation method.

[0097] In the adaptation method, digital training images of training valuable files of a preset valuable file type and digital reference images of reference valuable files of a preset valuable file type are used. For this purpose, in this example, clean, preferably freshly printed valuable files of a preset valuable file type are used, which preferably have variations in the location of the manufactured element. Preferably, the valuable files also include those that differ significantly in the location of the manufactured element.

[0098] For example, training and reference images can be acquired by the processing device 10, particularly the reflection sensor 40, for which digital images input to the analysis device 47 are stored. These digital images can be transmitted to the adapter device 70 via a storage medium or through a data connection not shown, and therein are stored in the memory device 72 and thus provided. The digital images each have the same number of pixels and pixel arrangement structure, and the entire valuable file is displayed.

[0099] In step S10, the location of the manufacturing element is determined for each training image and preferably for a reference image.

[0100] The positions are determined relative to the same positional reference system, which is given by the edges of the valuable document in the image, or by the edges of the image or the corresponding axis since the image only displays the entire valuable document.

[0101] To determine these locations, anchor regions 62A and 64A are used in this example. These anchor regions have been previously identified for valuable files of a preset valuable file type and are characterizing and, in particular, always visible for the manufacturing element. In this embodiment, only one anchor region is used for each; in other embodiments, it is preferable to use at least two or more anchor regions for each manufacturing element. The anchor regions typically contain multiple pixels, but for simplicity, they are shown in a generally schematic manner. Figure 3A and 3B The image is represented by only one pixel. Anchor regions can be searched for in the digital image using known methods. In this example, the median value of the pattern position can be used as the location of the anchor region to determine the position.

[0102] exist Figure 4 In the illustrated image of the valuable document, anchor points or related areas are marked by two overlapping rectangular shapes. To illustrate their relevance to the corresponding printed layers, anchor points in the first printed layer are surrounded by dotted lines, and anchor points in the second printed layer are surrounded by dashed lines.

[0103] In this example, the position is described by a position vector in a Cartesian coordinate system having an axis parallel to the edge of the valuable document, wherein the first component of the position vector is the x-coordinate of a pixel and the second component is the y-coordinate of a pixel.

[0104] In the next step S12, for the training image, pixels are assigned to manufacturing elements or manufacturing elements are assigned to pixels, wherein a reference image is used and the position of the manufacturing elements in the training image is taken into account.

[0105] In this example, for each training image, the following sub-steps are performed for each preset manufacturing element:

[0106] To take position into account, a movement vector is determined for the corresponding manufacturing element, which is equivalent to the difference between the determined position of the corresponding manufacturing element in the training image and the position of the corresponding manufacturing element in the reference image, or the difference in position vectors.

[0107] To configure pixels, the preset pixel environment of the corresponding pixel in the corresponding training image is compared with the preset pixel environment of the corresponding pixel in the reference image, taking into account the position of the corresponding fabrication element. The pixel environment for each pixel is defined identically.

[0108] This is accomplished in this example by comparing the pixel environment surrounding the corresponding pixel in the training image with the corresponding pixel environment in the reference image, assuming that the corresponding pixel in the training image belongs to one of the manufacturing elements. This means comparing the pixel environment of a pixel in the training image with the pixel environment of the corresponding pixel in the reference image, where the difference in position between the pixel environments in the training image and the reference image—that is, the difference between the position of the corresponding manufacturing element in the training image and the position of the corresponding manufacturing element in the reference image—corresponds to the translation vector.

[0109] Regarding the training image, this means that for a given manufacturing element in the manufacturing element, the training image is moved or shifted relative to the reference image by the movement vector, but in the opposite direction, i.e., in both direction and distance, the movement vector or distance being given by the difference between the position of the manufacturing element in the training image and the position of the manufacturing element in the reference image, and comparing the pixel environment and pixels in the moved training image and the reference image.

[0110] This comparison is achieved by determining the degree of consistency of pixel characteristics of pixels within a pixel environment. For each pixel environment of a given pixel, such as the 5x5 pixel environment of the given pixel including the pixel itself, the degree of consistency of the pixel data of the pixel environment in the moving training image with the pixel data in the reference image is determined; in this example, the degree of consistency is determined by the 2D correlation between the pixel data of the pixel environment and the pixel.

[0111] exist Figure 3A and 3B In the example, the anchor point 64A used to manufacture element 64 is moved two pixels to the left in the x-direction, and the movement vector V has only one non-zero component. Therefore, in the comparison, for example, Figure 3B The pixel 65T in the training image and its pixel environment are compared with the pixel and its pixel environment in the reference image, which is moved in the opposite direction to the movement vector V, i.e., moved by -V; this is in Figure 3A In the middle is pixel 65R. If these pixels belong to the same manufacturing element, a very good or perfect consistency or match is produced, but otherwise it is not: for example, the pixel characteristics of pixel 67T, which moves in the opposite direction to the movement vector V, and its surrounding pixel characteristics are inconsistent with those of pixel 67R and its surrounding pixel characteristics in the reference image.

[0112] After determining these consistency levels for each preset manufacturing element, a configuration criterion is used to decide whether a pixel can be assigned to one of the manufacturing elements based on the determined consistency level, and if so, which manufacturing element it can be assigned to. In this example, the configuration criterion is to configure the manufacturing element with the highest consistency level. However, this configuration only occurs when the consistency level exceeds a preset minimum value. If this is not the case, no manufacturing element is configured.

[0113] In the next step S14, a first component template and a second component template are generated for each manufacturing component.

[0114] This is accomplished in such a way that the first element template for the corresponding manufacturing element contains those pixels that are configured for the corresponding manufacturing element in all training images. The element template pixel data for said pixels is determined based on the corresponding pixel data of the training images. For example, for contamination detection, a lower limit and an upper limit of allowed pixel data values ​​can be determined as pixel data, the lower limit and the upper limit being formed by the minimum and maximum values ​​of the corresponding pixel data of the corresponding pixels in the training images, respectively.

[0115] For each manufacturing element, a second element template is generated in such a way that it contains the pixels that are configured for the corresponding manufacturing element in one or more training images. Here, the element template pixel data of the pixels can also be determined based on the pixel data of the corresponding pixels in those training images where the configuration is found. For example, in contamination testing, lower and upper limits of allowable pixel data values ​​can be determined as pixel data, said lower and upper limits being determined by the minimum and maximum values ​​of the corresponding pixel data of the pixels in the training images.

[0116] In step S16, a background template is generated, which includes pixels that are not included in any element template, particularly not in the first or second element template. These pixels are configured with pixel data determined using pixel data of corresponding pixels in the training image. For example, in contamination testing, lower and upper limits of allowed pixel data values ​​can be determined as pixel data, formed by the minimum and maximum values ​​of the corresponding pixel data of pixels in the training image.

[0117] In step S18, in this example, the component template and background template, or the corresponding data, are stored in memory 72. They can also be transferred to other devices via mobile memory or data connection, which preferably store and particularly preferably use the component template and background template, or the corresponding data.

[0118] During the operation of device 10, in this example, an inspection method for inspecting valuable documents of a preset valuable document type is executed. For this purpose, a component template, in this example determined by the previously described method, is stored in memory 48 within analysis device 47. In this embodiment, instructions for a computer program are stored in the same memory 48, and during the execution of the computer program, the inspection method is executed by the analysis device or its processor. For the inspection method, the determined component template is stored in memory 48. The inspection method is executed in real time, wherein valuable documents are transmitted through image acquisition device 40 at a rate exceeding 25 valuable documents per second, preferably exceeding 30 valuable documents per second.

[0119] In the inspection method, two component templates and one background template are used for manufacturing the components. These component templates and background templates are generated using the adaptation method described above.

[0120] The first component template used to manufacture the component, in this example, is a printed layer, determines which pixels of the image of the valuable document 12 to be inspected need to be explicitly assigned to this manufacturing component, in this example, this printed layer, and which component template pixel data need to be used for the corresponding pixels of the individual template to be created.

[0121] The second component template used to manufacture the component, in this example a printed layer, determines which pixels of the image of the valuable document 12 to be inspected need to be considered as belonging to this manufacturing component or this printed layer according to preset criteria, and which component template pixel data need to be used to determine the template pixel data for the corresponding pixels of the valuable document image.

[0122] The background template is assigned to the valuable file 12 as a whole and determines which template pixel data is needed for such pixels that do not need to be assigned to or belong to one of the manufacturing elements or one of the printing layers and are therefore not included in any element template. Therefore, such a background template exists only once for a valuable file type.

[0123] exist Figure 10 The above provides a general schematic illustration of the testing method.

[0124] As mentioned at the beginning, the valuable document 12 to be examined is conveyed by the conveying device 18 through the sensor device 38 and thus through the reflection sensor 40.

[0125] In step S20, for the valuable document conveyed through the reflection sensor 40, a digital image of the valuable document, i.e., the image of the valuable document, is acquired and, if necessary, transmitted to the analysis device 47 after preprocessing. The digital image is then provided in the memory 48 for verification. In this preprocessing, if the valuable document is not perfectly aligned with the reflection sensor and its longitudinal and lateral edges do not extend parallel to the corresponding axes of the coordinate system when the image of the valuable document is acquired, then, for example, the actual image of the valuable document can be aligned.

[0126] In step S22, the location of the manufacturing element is determined for the provided valuable file image, which is performed in accordance with step S10.

[0127] Then, in step S24, a separate template for a valuable digital file image is generated using the component template for manufacturing the component and the determined manufacturing component location and background template.

[0128] Therefore, the component template used for manufacturing the component moves accordingly to the separately determined positions of the corresponding manufactured component. Figure 3B In the example of a valuable file image, used for Figure 3A The component template of manufacturing component 64 will be from Figure 3A The position movement vector V in the middle.

[0129] First, a first component template is used. Template pixel data is configured for template pixels present in the first component template that has moved corresponding to the determined manufacturing component position. This template pixel data is determined based on the component template pixel data of the first component template that has moved corresponding to the determined manufacturing component position. More precisely, if the pixel exists only in one of the moving component templates, the corresponding component template pixel data is used as the template pixel data for that pixel. If a pixel in a valuable file can be found in both component templates that have moved corresponding to the position, the template pixel data is determined from the corresponding component template pixel data. For example, in the example of contamination inspection, the minimum value of the component template pixel data for the corresponding pixel can be determined as the lower limit and the maximum value as the upper limit.

[0130] For the remaining pixels of the template present in at least one second element template that has moved corresponding to the determined manufacturing element position, template pixel data is determined using element template pixel data of the respective corresponding pixels of the at least one second element template that has moved corresponding to the determined manufacturing element position. The method is not fundamentally different from the method used for the first element template.

[0131] Finally, for the remaining pixels in the template that are not present in the first or second element template but only in the background template, the template pixel data is determined separately using the background template pixel data of the corresponding pixels in the background template. Therefore, a complete template exists for the entire valuable file.

[0132] Then, in step S26, the digital valuable document image is examined using the determined template. This can be done using known template-based methods. For example, in contamination detection, it can be checked whether the pixel data of the pixels in the valuable document image falls within a certain range, the boundaries of which are given by the template pixel data of the corresponding pixels in the template. If the pixel data falls below this range, contamination is present.

[0133] As a result, a signal reflecting the test results can be output in step S28.

[0134] The second embodiment differs from the first embodiment in that the configuration of pixels and manufacturing elements, as well as the subsequent creation of element templates, are modified in the adaptation method. Here, steps S12 and S14 are replaced by steps S12' and S14'. Figure 11 The text provides a general illustration of the adaptation method.

[0135] In step S12', two configurations are performed in this example, differing in the size of the pixel environment and the configuration criteria. The determination of the corresponding degree of consistency is implemented similarly to the first embodiment.

[0136] More precisely, taking into account the corresponding positions of the manufacturing elements, and based on a determined first degree of consistency and according to a preset first configuration standard, the corresponding pixels of the training image are configured for the first time, and the pixels are assigned to one of the manufacturing elements. For example, the pixel environment is a 3x3 pixel environment including the pixel.

[0137] For the second configuration or second configuration process, taking into account the separately determined manufacturing component positions, for each pixel of the training image, a second degree of consistency between the second pixel environment of the corresponding pixel in the training image and the corresponding pixel environment of the corresponding pixel in the reference image is determined. The second pixel environment can be, for example, a 5x5 pixel environment.

[0138] The second configuration process is implemented similarly to the first configuration process, except that the difference between the second and first configuration criteria lies in the selection of the minimum value. However, the minimum value can also be the same.

[0139] In the modified step S14', first and second element templates are generated based on these configuration processes.

[0140] More precisely, for each manufacturing element, a first element template is generated using a corresponding first configuration process. This first element template contains pixels that were configured to the corresponding manufacturing element in all training images during the initial configuration. The average pixel data of the corresponding pixels in the training images is used as the element template pixel data. In other embodiments, the element template pixel data may further include a lower and upper limit of an allowed value range, which can be given by the minimum and maximum values ​​of the pixel data of the corresponding pixels in the training images.

[0141] Furthermore, for each manufacturing element, a second element template is generated using the corresponding second configuration process. The second element template includes pixels that were configured to the corresponding manufacturing element in at least one training image during the second configuration. The average pixel data of the corresponding pixels in the training images is used as the element template pixel data. In other embodiments, the element template pixel data may also include a lower limit and an upper limit of an allowed value range, which can be given by the minimum and maximum values ​​of the pixel data of the corresponding pixels in the training images.

[0142] The other steps are unchanged from the first embodiment.

[0143] The corresponding testing methods remain unchanged, except that different component templates are used.

[0144] This variant also includes Figure 4 The example of a valuable file type in the document will be used for illustration.

[0145] The pixels contained in the first element template and the element template pixel data associated with these pixels characterize the dominant position of the corresponding printed layer on the training valuable file, that is, the corresponding printed layer covers all other possible printed layers at this position, so that the other printed layers are almost invisible.

[0146] Figure 5 (Left) shows an example of configuring pixels to either the first printed layer (dark gray) or the second printed layer (light gray), the configuration for Figure 4 The training data in the image is localized within a pixel environment of ±2 pixels. Black areas cannot be explicitly assigned to the printed layer.

[0147] Figure 5 (Right) shows an example of configuring pixels to either the first printed layer (dark gray) or the second printed layer (light gray), the configuration for... Figure 4 A portion of the training data is processed within a pixel environment of ±10 pixels. Black areas cannot be explicitly assigned to the printed layer.

[0148] Figure 6 They were displayed respectively Figure 4 The example shown is a first element template (shown in light gray) for the first printed layer (left) and the second printed layer (right) in a segment of a training valuable file of the valuable file type. The area shown in black is not a portion of the corresponding first element template.

[0149] Figure 7 They were displayed respectively Figure 4 An example of a second element template for the first printed layer (left) and the second printed layer (right) in a segment of a training valuable file of a valuable file type. The white area is not occupied by pixels of the corresponding second element template and therefore not by pixels of the first element template; this area belongs to the background template.

[0150] Figure 8 An example of a separate template for a valuable document or an image of a valuable document is shown. The separate template consists of a first and a second element template and a background template, which are preferably generated in the previously described method for generating the template.

[0151] Other embodiments differ from those described previously in that they use these values ​​as element template pixel data or background template pixel data. These values ​​are functions of the pixel data of corresponding pixels in a training image, in which the values ​​are assigned to the manufacturing element or background. For example, these values ​​could be the average of the pixel data of pixels. These templates are applicable to different types of contamination testing, but also to authenticity testing. In testing, possible acceptable tolerances can be preset by the testing method.

[0152] In other embodiments, instead of interval boundaries, preset points of the interval, such as its center and its length, can also be used as pixel data.

[0153] Other embodiments may differ from the previously described implementations in that a larger pixel environment is selected. This selection may depend on the resolution of the digital image and the size of the visible structures on the valuable file of a preset valuable file type.

[0154] Other embodiments differ from the previously described embodiments in that the control and analysis device 46 is divided into two parts: an analysis device corresponding to the analysis device 47 and a control device separated from the analysis device, which receives signals from the analysis device and uses them for control.

[0155] In other embodiments, the adapter device may be part of a valuable document processing device. For example, the analysis apparatus of the device in the first embodiment may have a corresponding computer program that executes in an adapter operation mode of the valuable document processing device that does not process valuable documents.

Claims

1. A method for generating a component template, the component template being used to form a template when inspecting valuable files of a preset valuable file type, wherein, A valuable file of a preset valuable file type has at least two preset manufacturing elements, and the element template corresponds to the manufacturing element. The method uses a digital training image of a training valuable file of the preset valuable file type and a digital reference image of a reference valuable file of the preset valuable file type. The digital training image and the digital reference image each have pixels, and each pixel is configured with pixel data. The method includes the following steps: - For the training images, the positions of the manufactured elements are determined respectively, and When using a reference image, pixels from the corresponding training image are assigned to the manufacturing element. To assign pixels, a preset pixel environment for the corresponding pixel in the corresponding training image is compared with a preset pixel environment for the corresponding pixel in the reference image. The position of the corresponding manufacturing element is taken into account. To include the position of the corresponding manufacturing element, a movement vector is determined, corresponding to the difference between the determined position of the corresponding manufacturing element in the training image and its position in the reference image. - For each manufacturing element, a first element template is generated, which contains the pixels that have been assigned to the corresponding manufacturing element in all training images, and - For each manufacturing element, a second element template is generated, which contains the pixels that were configured for the corresponding manufacturing element in one or more training images.

2. The method according to claim 1, wherein, The manufacturing element is a printed layer and / or an anti-counterfeiting element.

3. The method according to claim 1, wherein, The manufacturing elements partially overlap.

4. The method according to claim 1, wherein, The training and reference valuable documents include valuable documents of a preset valuable document type that are either already manufactured or just printed.

5. The method according to claim 4, wherein, The manufactured and / or freshly printed pre-set valuable document type valuable documents are valuable documents of the same currency and / or face value.

6. The method according to claim 1, wherein the method comprises the further step: - Generate a background template, which includes pixels that were not configured for one of the manufacturing elements in the training image and background template pixel data respectively associated with said pixels.

7. The method according to claim 6, wherein, The background template pixel data is determined from the pixel data of the corresponding pixels belonging to the training image.

8. The method according to claim 1, wherein, When determining the position of a corresponding manufacturing element in a corresponding training image, the position is determined with respect to the same position reference frame or relative to at least one manufacturing element in a preset training image or relative to at least one manufacturing element in a reference image.

9. The method according to claim 1, wherein, When configuring the pixels of the corresponding training image to the manufacturing element, for the corresponding pixel of the corresponding training image, the consistency degree between the pixel environment of the pixel and the pixel environment of the corresponding pixel in the reference image is determined. In this process, the corresponding pixel environment of the corresponding pixel in the training image and the reference image is shifted in accordance with the corresponding relative position of the corresponding manufacturing element in the training image relative to the corresponding manufacturing element in the reference image. And according to the determined consistency degree, the corresponding pixel of the training image is configured to one of the manufacturing elements according to a preset configuration standard.

10. The method according to any one of claims 1 to 8, wherein, When configuring pixels of the corresponding training image to manufacturing elements, for the corresponding pixel of the corresponding training image, a first degree of consistency between the first pixel environment of the pixel and the first pixel environment of the pixel in the reference image is determined. This involves correspondingly shifting the first pixel environments in the training image and the reference image relative to the corresponding manufacturing element in the training image with respect to the corresponding manufacturing element in the reference image. Furthermore, in the first configuration, the corresponding pixels of the training image are configured to one of the manufacturing elements according to the determined first degree of consistency and a preset first configuration standard. For a corresponding pixel in the corresponding training image, a second degree of consistency between the second pixel environment of the pixel and the second pixel environment in the reference image is determined, wherein the second pixel environment in the training image and the reference image are correspondingly shifted relative to the corresponding manufacturing element in the training image with respect to the corresponding manufacturing element in the reference image, and in the second configuration, the corresponding pixel in the training image is configured to one of the manufacturing elements according to the determined second degree of consistency and a preset second configuration standard.

11. The method according to claim 10, wherein, The second pixel environment includes more pixels than the first pixel environment.

12. The method according to claim 10, wherein, The first element template for the corresponding manufacturing element contains only pixels that are assigned to the corresponding manufacturing element during the initial configuration in all training images, and The second element template for the corresponding manufacturing element includes pixels that are assigned to the corresponding manufacturing element in at least one training image during a second configuration.

13. The method according to claim 1, wherein, When generating a component template, first component template pixel data is configured for pixels included in the first component template, and second component template pixel data is configured for pixels included in the second component template, wherein the first and / or second component template pixel data... - Determine from the pixel data of the corresponding pixels assigned to the training image or - Corresponds to the pixel data of the corresponding pixel in the training image.

14. The method according to claim 1, wherein, To provide training data, digital images of multiple training valuable files of a preset valuable file type are acquired, and / or to provide reference images, digital images of reference valuable files of a preset valuable file type are acquired.

15. A computer program having program code means for performing the method according to any one of claims 1 to 14 when the program is executed on a computer.

16. A computer-readable data carrier having program code executable by a computer, thereby enabling the computer to implement the method according to any one of claims 1 to 14.

17. A method for verifying valuable documents of a preset valuable document type, wherein each valuable document has two preset manufacturing elements. Using a component template for manufacturing components, generated by means of the method according to any one of claims 1 to 14, the method comprising the steps of: - Provides a digital image of a valuable document to be verified, representing a preset valuable document type. The valuable document image includes pixels, each configured with pixel data. - Based on the provided valuable file images, determine the location of the manufactured components. - Determine a template for a valuable digital file image using a component template for manufacturing components and the determined location of the manufactured components. - Examine valuable digital document images using the established template. In order to configure pixels, the preset pixel environment of the corresponding pixel in the corresponding training image is compared with the preset pixel environment of the corresponding pixel in the reference image, wherein the position of the corresponding manufacturing element is taken into account, and in order to take the position of the corresponding manufacturing element into account, a movement vector is determined, the movement vector corresponding to the difference between the determined position of the corresponding manufacturing element in the training image and the position of the corresponding manufacturing element in the reference image.

18. The method according to claim 17, wherein, When determining the template, template pixel data is configured for each pixel of the template that exists in the component template that moves in accordance with the determined manufacturing component position. The template pixel data is determined based on the component template pixel data of the corresponding pixels of the component template for manufacturing the component that moves in accordance with the determined position.

19. The method according to claim 17, wherein, The first component template and the second component template are used to manufacture at least one component. Specifically, when determining the template, template pixel data is configured for the pixels of the template that exist in the first element template that moves corresponding to the determined manufacturing element position. This template pixel data is determined using the element template pixel data of the corresponding pixels in the first element template that moves corresponding to the determined manufacturing element position. For the remaining pixels that exist in at least one second element template that moves in correspondence with the determined manufacturing element position, template pixel data is determined using element template pixel data of the pixels respectively corresponding to the pixels of the second element template that moves in correspondence with the determined manufacturing element position.

20. The method according to any one of claims 17 to 19, wherein, A background template is used, which contains pixels that are not included in the component template for manufacturing the component, and wherein, when determining the template of a valuable file image for the numbers of the remaining pixels present in the background template, the template pixel data is determined separately using the component template pixel data of the respective corresponding pixels of the background template.

21. The method according to any one of claims 17 to 19, wherein, When providing images of valuable documents, an image acquisition device is used to acquire digital images of the valuable documents to be examined and the digital images of the valuable documents are stored in a memory device.

22. A method for processing valuable files of a preset valuable file type, wherein each valuable file has two preset manufacturing elements, wherein, The valuable document is transported separately through an image acquisition device for acquiring digital images of the valuable document, acquiring digital images of the valuable document during the transport of the respective valuable document, and subsequently examining the digital images of the valuable document by means of the method according to any one of claims 17 to 21.

23. The method of claim 22, wherein the processing is inspection and / or counting and / or sorting and / or destruction.

24. A computer program having program code means for performing the method according to any one of claims 17 to 23 when the program is executed on a computer.

25. A computer-readable data carrier having program code executable by a computer, thereby enabling the computer to implement the method according to any one of claims 17 to 23.

26. An apparatus for generating a component template for forming a template when verifying a valuable document of a preset valuable document type, wherein, The valuable file of the preset valuable file type has at least two preset manufacturing elements, and the device has a memory device for storing digital training images of the preset valuable file type and digital reference images of reference valuable files of the preset valuable file type, wherein the device is designed to implement the method according to any one of claims 1 to 14 when using training images and reference images.

27. The device according to claim 26, wherein the memory device is further configured to store the generated component template, the device being designed to store the generated component template in the memory device.

28. An apparatus for inspecting valuable documents, each of the valuable documents having at least two pre-defined manufacturing elements, the apparatus using a component template for manufacturing elements generated in the method according to any one of claims 1 to 14, the apparatus comprising: Analysis apparatus, the analysis apparatus having at least one memory storing component templates and an interface for providing digital valuable document images, wherein, The analytical apparatus is configured to perform the method according to any one of claims 17 to 23.

29. The apparatus according to claim 28, further comprising an image acquisition device for acquiring digital images of a valuable document to be examined, the image acquisition device being connected via a signal connection to an interface for providing digital valuable documents.

30. An apparatus for processing valuable documents of a preset valuable document type, wherein each valuable document has at least two preset manufacturing elements, the apparatus having An input device for inputting individual or sorted valuable documents to be processed. An output device having at least one output section for receiving the processed valuable file. A conveying device, wherein the conveying device is used to convey individual or sorted valuable documents from an input device to an output device, and The testing equipment according to claim 28 or 29 in, The image acquisition device of the inspection equipment is arranged on the transport path and configured such that digital images of the valuable documents to be inspected, which are transported through the image acquisition device, are acquired during transport and provided for use in the inspection equipment.

Images