Method and device for checking a security element for a security document

Abstract

The invention relates to a method and a device for checking a security element for a security document. The method comprises the following: providing a security element with an image that has a first image element and a second image element, wherein the first image element (8) is visible when illuminated with light from a first wavelength range and when illuminated with light from a second wavelength range that is different from the first wavelength range, and wherein the second image element is visible when illuminated with light from the first wavelength range, but not when illuminated with light from the second wavelength range; capturing an image recording (20) of the image when the image is illuminated with light from the second wavelength range, and determining the second image element by evaluating digital image data that indicate the image recording (20).The following is further provided: dividing the image recording (20) into partial areas (22, 23), wherein one of the partial areas (22, 23) is formed to comprise the second image element; determining a respective texture for the partial areas (22, 23), wherein the texture of the partial area indicates a distribution of gray value intensities within the partial area of ​​the image recording (20); determining a respective texture measure for the partial areas (22, 23), which indicates the extent to which the partial area is covered by the texture or is free of such a texture; and determining the presence of the second image element in the image of the security element if it is found by evaluating the texture measures that one of the partial areas (22, 23) comprising the second image element is substantially texture-free.

Description

[0001] The invention relates to a method and a device for checking a security element for a security document. background

[0002] Various security features for security documents are known that are optically verifiable. Optically verifiable security features are those that can be verified by detecting light transmitted through, reflected from, and / or remitted by the security feature. Light is defined here as all electromagnetic radiation in the visible, infrared, or ultraviolet wavelength range.

[0003] In connection with security documents such as documents that identify a person (ID document), it can be provided to apply a printed image to the security document as a security element, which image shows, for example, the face of a person to whom the document is assigned. In this case, it can be provided to produce image elements of the printed image with different colors. For example, four-color printing is known. It can be provided that the color of an image element, for example a fifth color in addition to the four-color printing, is visible when illuminated with light in the visible wavelength range, but not when the same image element is illuminated with light from a different wavelength range, for example with infrared light.In this way, with the aid of a camera device, different images can be captured of the printed image when illuminated with light of different wavelengths. In these images, the image element is either visible or invisible, depending on the light used to illuminate the printed image. Such a valuable or security product is described, for example, in document DE 10 2016 201 709 A1. Checking for the presence of such an image element on the security document can be used to verify the authenticity of the security document.

[0004] If a security document is intended to include such a security element with a printed image, which has an image element that is visible or invisible depending on the lighting, it must be checked during the production of the security element for the security document whether the image element was correctly created during the manufacturing process, particularly during printing. For such a testing process, it has been proposed to examine an image of the security element with the printed image using image analysis when illuminated with light in the wavelength range in which the image element is not visible.As part of the image analysis, respective grayscale histograms are determined for the image acquisition on the one hand and a comparison image acquisition on the other, which shows the printed image illuminated with the light in which the image element is also visible. These histograms are compared to verify the correct production of the image element that is visible or invisible under different lighting conditions. Such a method is known, for example, from document DE 10 2019 129 491 A1.

[0005] However, the known testing method proves to be insufficiently reliable for printed images, especially facial images, if, for example, the image element that is supposed to be visible or not visible under different lighting conditions is incomplete or insufficiently invisible, i.e., it still "shines through" in the image captured and used for image analysis.

[0006] Document DE 10 2006 011 143 A1 concerns a security code for application to packaging that, for example, contains a deposit and is recognized via the so-called barcode. There is a risk that the barcode can be easily reproduced in order to manipulate the return deposit system and thus illegally acquire deposit money. This security code consists of an IR ink or a combination of a color in the visible range, preferably black, and the IR security ink. One color is visible to the human eye and the other color is invisible as an infrared color, or alternatively, another visible IR color is used. Furthermore, the security symbol consists of a custom-designed geometry, or the EAN code is created entirely or partially using the security ink. Summary

[0007] The object of the invention is to provide a method and a device for testing a security element for a security document, with which the security element can be determined or tested with improved reliability.

[0008] The object is achieved by a method and a device for testing a security element for a security document according to independent claims 1 and 11. Embodiments are the subject of dependent subclaims.

[0009] According to one aspect, a method for checking a security element for a security document is provided, the method comprising: providing a security element with an image having a first image element and a second image element, wherein the first image element is visible upon illumination with light from a first wavelength range and upon illumination with light from a second wavelength range which is different from the first wavelength range, and wherein the second image element is visible upon illumination with light from the first wavelength range, but not upon illumination with light from the second wavelength range; capturing an image of the image with a camera device upon illumination of the image with light from the second wavelength range, and determining the second image element by evaluating digital image data indicating the image capture.When determining the second image element, the following is provided: dividing the image recording into partial areas, wherein one of the partial areas is formed to comprise the second image element; determining a respective texture for the partial areas, wherein the texture of the partial area indicates a distribution of gray value intensities within the partial area of ​​the image recording; determining a respective texture measure for the partial areas, which indicates the extent to which the partial area is covered by the texture or is free of such a texture; and determining the presence of the second image element in the image of the security element if it is found by evaluating the texture measures that one of the partial areas comprising the second image element is substantially texture-free.

[0010] According to a further aspect, a device for checking a security element for a security document is provided. The device comprises a camera device, an illumination device, and an evaluation device, which is formed, for example, with one or more processors for data processing and associated programming.The device is configured to capture an image of a security element using a camera device, the image having a first image element and a second image element, the first image element being visible when illuminated with light from a first wavelength range and when illuminated with light from a second wavelength range that is different from the first wavelength range, and the second image element being visible when illuminated with light from the first wavelength range but not when illuminated with light from the second wavelength range, and the image capture being capable of being carried out when the image is illuminated by the illumination device with light from the second wavelength range. Furthermore, the device is configured to determine the second image element by evaluating digital image data that indicate the image capture.The following is provided here: dividing the image recording into partial areas, wherein one of the partial areas is formed to comprise the second image element; determining a respective texture for the partial areas, wherein the texture of the partial area indicates a distribution of gray value intensities within the partial area of ​​the image recording; determining a respective texture measure for the partial areas, which indicates the extent to which the partial area is covered by the texture or is free of such a texture; and determining the presence of the second image element in the image of the security element if it is found by evaluating the texture measures that one of the partial areas comprising the second image element is substantially texture-free.

[0011] It is intended to use texture analysis to determine or check the security element with the image. The image comprises image elements produced using different colors, which has the result that at least a first image element is visible when the image is illuminated with light of different wavelength ranges, whereas at least a second image element, which is different from the first image element, is either visible or invisible when illuminated with light from different wavelength ranges. The image can, for example, be produced as a printed image or printed image, wherein different printing inks can be used in particular when producing the first and second image elements.Printing inks are known as such that are visible when illuminated with light from a first wavelength range, for example visible light, as well as when illuminated with light from a second wavelength range that is different from the first wavelength range, for example infrared or UV light. Printing inks are also known that are visible in both different wavelength ranges. In the case of a black and white image, black ink of different compositions can be used to achieve the corresponding behavior with regard to visibility in light from different wavelength ranges. Such colorants, in particular printing inks, are known as such in various forms.

[0012] The different visibility of the colours used to create the image makes it possible to create a security element with an image in which different image elements have different visibility depending on the lighting.

[0013] It is provided that an image recording for the image of the security element is captured by means of a camera device when the image is illuminated with light from the second wavelength range, which means that the second image element is then essentially invisible. Such non-visibility of the second image element can, for example, be due to the light of the second wavelength range being essentially completely reflected by the ink used to produce the second image element. Visibility due to the absorption of light by the ink used for the second image element is then eliminated, which is why the second image element is not visible in the image recording.

[0014] The image is then analyzed or determined. In this case, a texture analysis is used for partial areas of the image to determine a texture and a texture measure characterizing this texture for the respective partial area. The texture measure can be, for example, a value number, a value range and / or a threshold value. More complex texture measures are also possible, for example a matrix characterizing the texture of the assigned partial area or a tuple of numbers. Since non-visibility is expected for the second image element in the captured image, the presence of the second image element in a partial area comprising the second image element in the image of the security element is inferred if a missing texture (texture freedom) is determined for the partial area containing the second image element. This means that a texture provided by the second image element is missing in this partial area.

[0015] The light used for illumination in the different wavelength ranges can be narrowband, using essentially only one wavelength, or broadband, for example white light.

[0016] The texture of the sub-area can indicate a planar or spatial distribution of gray-value intensities within the sub-area. A gray-value intensity can be assigned to a single pixel or a group of pixels.

[0017] A digital camera can be used to capture the image of the security element with the image.

[0018] The image can be divided into sub-areas such that adjacent sub-areas abut each other at the edges of the sub-areas. This supports the ability to determine information regarding the transition of the texture dimension between adjacent sub-areas during texture analysis and to use it for inspection. For example, a minimum difference between the texture dimension for adjacent sub-areas of the image can indicate a transition between a visible image element in the image and a substantially invisible (second) image element.

[0019] In this or other embodiments, the partial areas formed when dividing the image can take on different geometric shapes, for example, square or round. Triangles, squares, or polygons can be provided when forming the partial areas. The edges of the partial areas can abut one another or be spaced apart for at least some of the partial areas.

[0020] The image recording can be divided into partial areas such that one of the partial areas substantially completely overlaps with the second image element. In this embodiment, the partial area formed as part of the subdivision of the image recording and containing the second image element substantially completely overlaps with the area occupied by the second image element in the image of the security element. In an alternative embodiment, the second image element can be arranged in a larger or smaller partial area, so that there is no complete overlap.

[0021] The security element can be provided with a printed image comprising the first and second image elements. Different printing inks can be used to create the different behavior of the image elements under different lighting conditions (visibility vs. non-visibility). Printing inks that are visible when illuminated with light from different wavelength ranges or invisible under light of a specific wavelength, in particular due to their differing reflection and / or absorption behavior for light of different wavelengths, are known as such in various designs.

[0022] It may be provided to provide a threshold texture measure, and the presence of the second image element in the image of the security element can be determined if, upon evaluation of the texture measures, it is found that the texture measure for one of the partial areas comprising the second image element is less than or equal to the threshold texture measure. This makes it possible, for example, to use the threshold texture measure to determine a probability for the essentially given texture-free nature of the partial area with the second image element in an application-dependent manner, for example, if the non-visibility of the second image element in the image recording is only partially given, in particular because the second image element still "shines through" in the image recording, which may, for example, be the result of incomplete reflection of the incident light by the color of the second image element.In the testing process for the security element, threshold or limit values ​​for the texture freedom to be tested or the probability of texture freedom can be applied individually.

[0023] Descriptive data can be provided that indicate properties of the second image element, such as edge length, size, shape, area, and / or area contour, and the descriptive data can be evaluated to form one of the sub-areas when dividing the image into sub-areas such that it encompasses the second image element. Based on the descriptive data, the sub-area of ​​the image can be divided into (image) sub-areas in an application-dependent manner, for example, such that the second image element in the sub-area encompassing the image overlaps the second image element entirely or only in certain sections.

[0024] In one example, the following may be provided: the security element is provided with an image comprising a further second image element; the further second image element is visible when illuminated with light from the first wavelength range, but not when illuminated with light from the second wavelength range; the image recording is divided into partial areas such that another of the partial areas comprises the further second image element; and the presence of the further second image element in the image of the security element is determined or positively assessed if, by evaluating the texture measures, it is found that the further partial area comprising the further second image element is substantially texture-free. Testing or analysis of the further second image element can be carried out in the same way as for the second image element or in parts differently.In one example, correct production of the image of the security element can only be determined if the substantially given texture freedom is found for the second and the further second image element. Otherwise, it can be determined that the image of the security element has not been correctly produced, in particular if the texture freedom or the necessary probability for texture freedom is determined only for the second or only the further second image element.

[0025] Similarly, it may be provided that the authenticity of the security element or of a security document containing the security element is only determined if both second image elements are categorised or classified as substantially texture-free.

[0026] The texture measure for the partial areas can be determined using a texture filter according to Laws. Within the framework of the texture analysis, various methods can be used to determine the texture and the texture measure for the partial areas, which are known per se for texture analysis of image data. The methods each provide a characterization of properties or features of the texture of the assigned partial area, from which a texture measure can then be derived, which indicates the extent or degree to which the partial area has a texture or is essentially free of it. In one embodiment, texture filters or masks according to Laws are used for the analysis of the partial areas of the image recording (KI Laws: Textured Image Segmentation, Department of Electrical Engineering, Image Processing Institute, University of California, January 1980). ).Alternatively or additionally, texture analysis can be applied using co-occurrence matrices (see, for example, Nanni et al.: Different approaches for extracting information from the co-occurrence matrix, PLOS ONE, December 2013, Vol. 8, Issue 12, e83554). In particular, matrices characterizing the texture of the partial area are determined. The entire matrix can then be used as a texture measure, for example. Alternatively, the trace of the matrix can be determined and used as a texture measure.

[0027] It may be provided to check for correct production of the image of the security element, wherein correct production is determined if at least the presence of the second image element in the image of the security element is determined. The method for determining correct production of the image of the security element can be applied, for example, in the context of the production of ID documents, for example after the printed image of the security element has been applied to a document body, such as a plastic card. If it is found that the image of the security element was not produced correctly, the manufactured product must be classified as reject and is either corrected or removed from the further production process.

[0028] It may be provided that the method for checking a security element for a security document is used to check a security document, in particular to determine its authenticity, wherein optionally further security features of the document can be evaluated, as is known per se.

[0029] For texture analysis, artificial intelligence, for example in the form of machine learning, can be applied using appropriate software in the evaluation device. In one example, to determine the texture freedom in the partial area containing the second image element, a neural network can be used in such a way that the elements of a gray value matrix are passed to the input layer of the neural network. Alternatively, a suitable Support Vector Machinebe used to classify the texture measure; for example, a number tuple is used as input, which is formed from several gray value matrices from the traces or is calculated using several Laws filters.

[0030] Security documents include, for example, ID documents, passports, identity cards, identification cards, driver's licenses, vehicle registration documents, vehicle registration certificates, access authorization cards, company ID cards, access cards, bank cards, credit cards, visas, but also anti-counterfeit labels, admission tickets, banknotes, postage stamps, securities, or the like. Such security documents that embody value, such as postage stamps, securities, or banknotes, are also referred to as valuable documents. A clear distinction between valuable and security documents is not always possible and, with regard to the present disclosure, is also not essential. For the purposes of the subject matter described here, valuable documents are always understood as security documents. Description of implementation examples

[0031] Further embodiments are explained below with reference to the figures of a drawing. Herein: Fig. 1 is a schematic representation of a device for checking a security element of a security document; Fig. 2 is a schematic representation of an image recording for a facial image, wherein a second image element is not visible; Fig. 3 is a schematic representation of an image recording for the facial image from Fig. 2, whereby the second image element is also visible due to different lighting; Fig. 4a to 4d schematic representations in connection with a texture analysis for an image recording with a facial image using texture filters according to Laws; Fig. 5a and 5b schematic representations in connection with a further texture analysis for various image recordings with a facial image using texture filters according to Laws; Fig. 6 a schematic representation in connection with a texture analysis of an image recording with a facial image using texture filters according to Laws and Fig. 7 a schematic representation in connection with a texture analysis of another image recording with a facial image using texture filters according to Laws.

[0032] Fig. 1shows a schematic representation of an arrangement with a device 1 for checking or determining an image 2 of a security element 3 on a security document 4, which is, for example, a document for identifying a person (ID document). The security document 4 is formed with a document body 5, which is, for example, a plastic card. In other embodiments, the security document 4 is designed, for example, as an ID card or passport.

[0033] The device 1 comprises a camera device 6, for example a digital camera, which is configured to capture an image of the image 2. For this purpose, the security element 3 is illuminated with the image 2 by means of an illumination device 7.

[0034] In the embodiment shown, image 2 has a first image element 8 with an upper half of the face and a second image element 9 with a lower half of the face.

[0035] In the example shown, the first and second image elements 8, 9 abut one another. In other embodiments, a gap may be formed between the first and second image elements 8, 9.

[0036] In the exemplary embodiment, image 2 is embodied as a printed image. The first image element 8 and the second image element 9 were produced using different printing inks. Image 2 can be a color image or a black and white image. In the case of a black and white image, different black inks are used to produce the first and second image elements 8, 9. The color of the first image element 8 can be selected such that the first image element 8 is visible when illuminated with light from a first wavelength range as well as when illuminated with light from a second wavelength range that is different from the first wavelength range, in particular for the camera device 6. The visibility is based, for example, on at least partial absorption of the light from the respective wavelength range.

[0037] In contrast, the second image element 9 is only visible when illuminated with light from the first wavelength range, but not when illuminated with light from the second wavelength range (non-visibility). The light from the respective wavelength range can be narrowband or broadband. In the case of narrowband light, the illumination only comprises light of a few wavelengths or essentially only one wavelength, whereas with broadband light a plurality of wavelengths contributes to the illumination, as is the case with white light, for example. In one embodiment, the light from the first wavelength range is white light, whereas the light from the second wavelength range is infrared or UV light. For example, the second image element 9 is not visible when illuminated with infrared or UV light.In the context of the present disclosure, non-visibility means that the second image element 9 is not visible at all or only very dimly ("translucent") when captured with the camera device 6. In contrast, the first image element 8 is visible in an image captured by the camera device 6 not only when illuminated with light from the first wavelength range, but also when illuminated with light from the second wavelength range.

[0038] With the help of the Fig. 1The device 1 shown can be used for various purposes to check whether the image 2 correctly displays the second image element 9. For this purpose, digital image data representing the image captured by the camera device 9 are evaluated with the aid of an evaluation device 10. The evaluation device 10 is equipped for this purpose with hardware and corresponding software, in particular programs for digital image analysis. The programs can be configured to apply artificial intelligence to the image analysis.

[0039] The verification by means of image analysis can be used to check the correct production of image 2 in the manufacturing process, for example after the image 2 has been applied as a printed image on the document body 5. Alternatively or additionally, the analysis or verification can be used to check the authenticity of the security document 4 later in use.

[0040] When checking the image 2 of the security element 3, an image recording 20 of the image 2 is made by means of the camera device 6 according to Fig. 2 Here, the security element 3 is illuminated with light from the second wavelength range by means of the illumination device 7, with the aim of capturing an image by means of the camera device 6 in which the first image element 8 is visible, but not the second image element 9 with the lower half of the face and neck (cf. Fig. 2). Fig. 3 For comparison, Fig. 2 shows another image 21 of image 2 illuminated with light from the first wavelength range. Here, the first and second image elements 8 and 9 can be seen.

[0041] The image recording 20 is then analyzed with the aid of the evaluation device 9. In this case, description data can be provided in the evaluation device 10, which indicate properties of the first and / or second image element 8, 9 such as edge length, size, shape, area and / or surface contour. The description data can be used in the inspection process to divide the captured image recording 20 into sub-areas. Alternatively, the division of the image recording 20 into sub-areas can take place without reference to or use of the description data. In one embodiment, according to Fig. 3 For example, it is provided to divide the image recording 20 into a first partial area 22 with the first image element 8 and a second partial area 23 with the second image element 9 (here - as expected - not visible), wherein the assigned partial area completely covers the respective image element.

[0042] For the digital image data representing the image recording 20, a texture analysis is carried out by means of the evaluation device 10, in which a respective texture is determined for the first and second partial areas 22, 23 formed for this purpose, wherein the texture of the partial area indicates a distribution of gray value intensities within the partial area, for example with a gray value intensity for each image point or each image pixel in the partial area. For the first and second partial areas 22, 23, a respective texture measure is also determined, which indicates the extent to which the partial area is covered by the texture or is free of such. The texture measure can, for example, be a value number, a value range and / or a threshold value. More complex texture measures are also possible, for example a matrix characterizing the texture of the assigned partial area or a tuple of numbers.

[0043] The texture analysis thus determines, on the one hand, the texture possibly present in the assigned partial area and provides a texture measure which at least makes it possible to determine whether the assigned partial area is essentially free of a texture or has one. For the second partial area 23 of the image recording 20 with the (as expected) missing, i.e., invisible, second image element 9 - as in Fig. 3 shown - it is found that the assigned texture measure indicates a substantially existing freedom of texture, it is concluded that the second image element 9 in image 2 is correctly present, i.e. the correct color production or formation.

[0044] Optionally, further second image elements can be provided in image 2, which are designed in a manner comparable to the second image element 9, at least with regard to the use of the same color, and can thus be analyzed in a similar manner in order to check the correct formation of the second image elements.

[0045] Of course, in order to check the authenticity of the security document 4, further security elements (not shown) can be checked on the security document 4, as is known per se in various embodiments.

[0046] With reference to the Fig. 4 to 6 Further aspects of texture analysis when checking image 2 of security element 3 are explained below. The texture analysis is exemplified using texture filters according to Laws (KI Laws: Textured Image Segmentation, Department of Electrical Engineering, Image Processing Institute, University of Southern California, January 1980 )executed.

[0047] According to Laws, a texture transformation of the image is performed using convolution with filter kernels to determine the texture. The filter kernels are, for example, 5x5 matrices consisting of vectors l, e, s, w and r to highlight certain texture properties ("Level", "Edge", "Spot", "Wave", "Ripple"): l = 1 4 6 4 1 e = − 1 − 2 0 2 1 s = − 1 0 2 0 − 1 w = − 1 2 0 − 2 1 r = 1 − 4 6 − 4 1

[0048] The texture filter mask " sl " then looks like this: s ⊗ l = − 1 − 4 − 6 − 4 − 1 0 0 0 0 0 2 8 12 8 2 0 0 0 0 0 − 1 − 4 − 6 − 4 − 1

[0049] Fig. 4a shows an image recording 40 of a facial image of a security element, in which - in contrast to, for example, Fig. 3- the first image element 8, which captures most but not all of the upper half of the face, is missing. Unlike the second image element 9, which captures the lower half of the face, the first image element 8 is not visible during the illumination for image capture 40. The edges of the first and second image elements 8, 9 abut in an area 41.

[0050] Within the framework of texture analysis, different filter masks according to Laws can now be used to determine the texture. Fig. 4b and 4c This is for the filter masks " ls " as well as " sl" shown. It can be seen that for images 42, 43 filtered as part of the texture analysis, the area with the first image element 8, which covers most but not all of the upper half of the face, is essentially texture-free, whereas the second image element 9 with the lower half of the face and the base of the neck clearly has a texture. In the example shown, the first image element 8 is formed with a tile shape (rectangle) when the image recording 40 is subdivided for the texture analysis, such that the tile covers most of the upper half of the face, but not completely. A partial area 45 of the upper half of the face lies outside and is visible in the image recording 40, and was therefore not produced with the color that is not visible under the lighting for the image recording 40. An area of ​​the filtered images 42, 43 assigned to the partial area 45 then also has a recognizable texture.

[0051] Fig. 4dshows a figure 44 of maximum gray value intensities from figures 42, 43 in the Fig. 4b and 4c For each image pixel, the maximum of the gray value intensities from the images in the Fig. 4b and 4c Here, too, the freedom of texture is evident in the area with the first image element 8.

[0052] A beam 46 in Fig. 4a serves only to formally anonymize the displayed facial image.

[0053] Fig. 5a and 5b show representations for a captured image 50, 51 (left illustration) and figures 52, 53, which show the result of the texture determination (right illustration). When creating figures 52, 53, filter masks according to Laws were again applied to the images 50, 51. This shows that the filtered figures 52, 53 on the right side in the Fig. 5a and 5b clearly highlight the texture-free areas and thus make them identifiable, even if, as in Fig. 5a, in the image recording 50 the facial image is not arranged centrally in the image recording 50. Also for the image recording 51 in Fig. 5b , in which the facial image "shines through", the textured and texture-free areas also emerge.

[0054] Beams 54, 55 in Fig. 5a and 5b serve only to formally anonymize the displayed facial image.

[0055] To check whether a partial area of ​​the image is essentially free of texture, an application-specific texture measure is used. For example, a histogram of the images filtered according to the texture analysis can be used for this purpose. Figs. 6 and 7 each show - based on previously acquired images - specific images 60, 70 with four partial areas 60, ..., 63; 70, ... ,73 formed for texture analysis. The images 60, 70 were determined using texture filters according to Law.

[0056] For each of the partial areas 60, ..., 63; 70, ... ,73, a texture dimension determination is carried out depending on the number N the image pixels with a gray value intensity above a certain threshold (here for example: 127) a texture measure P determined: σ is a freely selectable scaling parameter, which can make it possible in certain embodiments that the result P better lies in the interval [0;1]. Since N the number of pixels in the corresponding subarea with a gray value intensity (after filtering) above a certain threshold, σ in the order of magnitude of N. In the example σ at σ = 0.04 · A , where A specifies the size of the subarea in pixels.

[0057] For the subareas 60, ..., 63 the following results: P 60 = 0.989, P 61 = 0.248, P 62 = 0.000, P 63 = 0.987.

[0058] The partial areas 60, 63 exhibit a high number of pixels with grayscale intensity. The image elements contained therein are therefore not visible under the lighting used for the underlying image acquisition (texture freedom). The image elements contained in the partial areas 60, 63 are each not visible, for example, printed with a color that is not visible under the lighting used for the image acquisition.

[0059] For Fig. 7 the following results: P 70 = 0.989, P 71 = 0.984, P 72 = 0.000, P 73 = 0.987.

[0060] The high values ​​for the partial areas 70, 71, and 73 indicate a lack of texture. If, for the facial image captured here and examined using texture analysis, a visible image element in the partial area 71 would also be expected in the image capture (for example, comparable to Fig. 6), then the texture analysis determines that the image (in this example: a facial image) of the security element was not produced correctly. This could be due, for example, to the image element expected to be visible in partial area 71 not being printed with the correct color. The image element could also simply not have been printed.

[0061] The features disclosed in the above description, the claims and the drawings may be important for the realization of the various embodiments both individually and in any combination. List of reference symbols

[0062] 1Device 2Image 3Security element 4Security document 5Document body 6Camera device 7Illumination device 8First image element 9Second image element 10Evaluation device 20, 21Image capture 22, 23First and second partial areas in image capture 40Image capture 41Edge of image elements 42, 43, 44Image filtered for texture analysis 50, 51Image captures 52, 53Image filtered for texture analysis 60, ... ,63Partial areas for texture analysis 70, ... ,73Partial areas for texture analysis

Claims

1. A method for checking a security element for a security document, the method comprising: - providing a security element (3) with an image (2) which has a first image element (8) and a second image element (9), wherein the first image element (8) is visible when illuminated with light from a first wavelength range and when illuminated with light from a second wavelength range which is different from the first wavelength range, and wherein the second image element (9) is visible when illuminated with light from the first wavelength range, but not when illuminated with light from the second wavelength range;- capturing an image (20) of the image (2) with a camera device (6) while illuminating the image (2) with light from the second wavelength range, and - determining the second image element (9) by evaluating digital image data indicating the image (20), the following being provided: - dividing the image (20) into partial areas (22, 23), one of the partial areas (22, 23) being formed to comprise the second image element (9); - determining a respective texture for the partial areas (22, 23), the texture of the partial area indicating a distribution of grayscale intensities within the partial area of the image (20); - determining a respective texture measure for the partial areas (22, 23), which indicates the extent to which the partial area is covered by the texture or is free of such a texture;and - determining the presence of the second image element (9) in the image (2) of the security element (3) if it is found by evaluating the texture dimensions that one of the partial areas (22, 23) comprising the second image element (9) is substantially texture-free; 2. Method according to claim 1, characterized in that the image recording (20) is divided into partial areas (22, 23) in such a way that adjacent partial areas abut one another in the region of edges of the partial areas (22, 23).

3. Method according to claim 1 and 2, characterized in that the image recording (20) is divided into partial areas (22, 23) in such a way that one of the partial areas substantially completely overlaps with the second image element (9).

4. Method according to at least one of the preceding claims, characterized in that the security element (3) is provided with a printed image comprising the first and second image elements (8, 9).

5. Method according to at least one of the preceding claims, characterized marked that a threshold texture measure is provided and the presence of the second image element (9) in the image (2) of the security element (3) is determined if, upon evaluation of the texture measures, it is found that the texture measure for the one of the partial areas (22, 23) comprising the second image element (9) is less than or equal to the threshold texture measure.

6. Method according to at least one of the preceding claims, characterized marked that description data are provided which indicate properties of the second image element (9) such as edge length, size, shape, area and / or area contour, and the description data are evaluated in order to form one of the partial areas (22, 23) in such a way that it comprises the second image element (9) when dividing the image recording (20) into the partial areas (22, 23).

7. Method according to at least one of the preceding claims, characterized marked that - the security element (3) is provided with an image which has a further second image element; - the further second image element is visible when illuminated with light from the first wavelength range, but not when illuminated with light from the second wavelength range; - the image recording is divided into partial areas such that a further one of the partial areas comprises the further second image element; and - the presence of the further second image element in the image of the security element is determined if it is found by evaluating the texture dimensions that the further partial area which comprises the further second image element is substantially texture-free.

8. Method according to at least one of the preceding claims, characterized marked that the texture measure for the partial surfaces (22, 23) is determined using a texture filter according to Laws.

9. Method according to at least one of the preceding claims, characterized in that a correct production of the image (2) of the security element (3) is checked, wherein a correct production is determined if at least the presence of the second image element (9) in the image (2) of the security element (3) is determined.

10. A method for checking a security document, in which a security document (4) with a security element (3) is provided and the security element (3) is checked according to a method according to at least one of claims 1 to 8.

11. A device for checking a security element for a security document, comprising: - a camera device (6); - an illumination device (7) and - an evaluation device (10); wherein the device is configured to capture an image (20) of an image (2) of a security element (3) with a camera device (6), wherein - the image (2) has a first image element (8) and a second image element (9), wherein the first image element (8) is visible when illuminated with light from a first wavelength range and when illuminated with light from a second wavelength range which is different from the first wavelength range, and wherein the second image element (9) is visible when illuminated with light from the first wavelength range, but not when illuminated with light from the second wavelength range;and - the image recording (20) can be carried out when the image (2) is illuminated by means of the illumination device (7) with light from the second wavelength range; and - the second image element (9) is to be determined by evaluating digital image data indicating the image recording (20), the following being provided: - dividing the image recording (20) into partial areas (22, 23), one of the partial areas (22, 23) being formed to comprise the second image element (9); - determining a respective texture for the partial areas (22, 23), the texture of the partial area indicating a distribution of grayscale intensities within the partial area of the image recording (20); - determining a respective texture measure for the partial areas (22, 23), which indicates the extent to which the partial area is covered by the texture or is free of such a texture;and - determining the presence of the second image element (9) in the image (2) of the security element (3) if it is found by evaluating the texture dimensions that one of the partial areas (22, 23) comprising the second image element (9) is substantially texture-free;

Images