WO2019092219A1 - Sensor system for checking hand vein patterns - Google Patents

Abstract

The invention relates to a sensor system for checking a vein pattern. The sensor system comprises a first light source, which is designed to emit, over the full surface, electromagnetic waves having wavelengths in the near infrared range during operation, which are absorbed by hemoglobin. The sensor system also comprises a second light source, which is designed to emit, over the full surface, electromagnetic waves having wavelengths in the range of visible light during operation. The sensor system also comprises a camera, which has a first camera chip, which is designed to capture reflection signals of electromagnetic waves having wavelengths in the near infrared range and to convert the same into a corresponding infrared image, and a second camera chip, which is designed to capture reflection signals of electromagnetic waves having wavelengths in the range of visible light and to convert the same into a corresponding light image. Finally, the sensor system comprises a first processor unit, which is connected to the first and second camera chips and which is designed to compare the light image and the infrared image, to determine a degree of correspondence between the light image and the infrared image, and to classify the vein pattern as authentic if the degree of correspondence is less than a predefined degree of correspondence and to classify the vein pattern as inauthentic if the degree of correspondence is greater than or equal to the predefined degree of correspondence.

Description

 Sensor system for testing hand vein patterns

The invention relates to a sensor system for testing a hand vein pattern and to a method for testing a hand vein pattern. Sensors and systems for recording hand vein patterns are already known from the prior art, the signals of which are used, for example, for access controls or identifications. Such a sensor system is described for example in DE 10 2013 208 654.

The object of the present invention is to provide an improved sensor system and an improved method for recording and testing manual vein patterns with which counterfeiting can be detected.

This object is achieved according to a first aspect of the invention by a sensor system which comprises a first light source, a second light source, a camera with a first and a second camera chip and a first processor unit. The first light source is designed to emit in its entirety electromagnetic waves with wavelengths in the near infrared range during operation, which are absorbed by hemoglobin. During operation, the second light source is designed to emit electromagnetic waves with at least one further wavelength, in particular in the region of visible light. The camera comprises a first and a second camera chip, wherein the first camera chip is designed to receive reflection signals of electromagnetic waves with wavelengths in the near infrared range and to be converted into a to convert speaking infrared image and wherein the second camera chip is adapted to receive reflection signals of electromagnetic waves of at least one further wavelength, and convert it into a corresponding light image. Infrared image here means electrical signals of an infrared sensor which represent an image taken under infrared illumination, while light image means electrical signals of an image sensor which represent an image recorded under illumination with the at least one further wavelength, which lies in particular in the visible light range. The first processor unit is connected to the first and second camera chips and configured to compare the light image with the infrared image, to determine a degree of coincidence between the light image and the infrared image, and to classify the vein pattern as true if the degree of matching is less than a predetermined degree of matching and Classify venous patterns as non-genuine if the degree of matching is greater than or equal to the predetermined degree of agreement. The venous pattern is preferably a palm vein pattern. According to a second aspect, the invention relates to a method for testing a vein pattern. The method comprises the steps:

 full exposure of a hand or part of the hand to electromagnetic waves with near infrared wavelengths absorbed by hemoglobin,

Recording reflection signals of electromagnetic waves having wavelengths in the near infrared range and converting them into a corresponding infrared image,

 full-surface irradiation of the hand or of a part of the hand with electromagnetic waves having at least one further wavelength, in particular in the region of visible light,

Recording reflection signals of electromagnetic waves with the at least one further wavelength, and converting to a corresponding light image,

 Comparing the light image and the infrared image with each other,

 Determining a degree of agreement between the photograph and the infrared image and

Classifying the palm vein pattern as genuine if the degree of matching is less than a predetermined degree of agreement and not true if the degree of agreement is greater than the predetermined degree of agreement.

The invention is based on the following findings: The near infrared light is able to penetrate into the deeper layers of the skin and be proportionately reflected from there. Hemoglobin in the veins absorbs most of the infrared light so that little infrared light is reflected from there. An image capture with infrared light is the Consequently, vein patterns result in a light-dark contrast. The infrared image shows the vein pattern. Since only a part of the infrared light penetrates into the skin, infrared light is also reflected by the hand surface. If a vein pattern is painted onto the surface of the hand, eg with a dark color, then the dark color also absorbs a part of the infrared light when imaging with infrared light. The painted vein pattern is thus also displayed in the light-dark contrast. Known prior art sensor systems and methods for venous pattern recognition can not determine with certainty whether it is a genuine or painted vein pattern. The invention is based on the finding that counterfeits such as painted venous patterns can be reliably detected via a photograph of a photograph and a comparison between the infrared image and the photograph.

If a second image is taken when illuminated with a non-infrared light source, in particular for visible light, in particular in the shorter visible wavelength range, then this light is not able to penetrate into the deeper layers of the skin. The light is for the most part reflected on the skin surface. A vein pattern painted with a dark color would absorb the light, and in the image this vein pattern would be visible in the light-dark contrast.

Thus, if veins are detectable both in the image taken with the near infrared light, the infrared image, and in the image recorded with the visible light, the light image, then the vein pattern is a fake.

The invention also includes the knowledge that a comparison of the image recordings in the particularly visible range and in the near infrared range also differences or the lack of such differences can be determined, which go back to surface structures and material properties. Thus, the differences in infrared image and image on a real hand surface are different than on counterfeit surfaces such as: paper (printed vein patterns), plastic coatings (e.g., rubber gloves), plastic hands, and other imitations. In principle, the reflection properties of the materials differ in different electromagnetic wavelength ranges. These differences can be recognized and evaluated in the image signals.

Thus, by means of the invention, it is possible to check whether a venous pattern is a true venous pattern or a forgery by merely classifying a venous pattern as genuine, in other words, if the comparison of infrared image and light image gives a degree of agreement that is less than a predetermined degree of agreement.

Thus, with the invention, an improved detection of venous patterns in terms of safety and the rejection of counterfeiting, and thus increased security for access controls based on recognition of venous patterns, are possible. The invention further includes the insight that hemoglobin only absorbs electromagnetic waves having wavelengths in the range of near infrared light, but therefore invisible light and a true venous pattern can only be detected by near infrared light from a corresponding camera but not with visible light Light. In the context of the present invention, near-infrared is understood as meaning light having wavelengths in the range from 780 to 1000 nm

Exemplary embodiments of the sensor system according to the invention and of the method according to the invention are described below. The additional features of the embodiments may be combined with each other to form further embodiments unless expressly described in the description as alternatives to each other.

The degree of matching is statistically determined in one embodiment by the population of the light image and infrared image data in the system. The degree of correspondence can be represented by a multi-dimensional vector giving a similarity measure. Alternatively, the degree of agreement can also be determined via an evaluation of the infrared image and the light image, in which deviations in predefined areas are determined and added up to a degree of agreement according to a point system.

In one embodiment, the predetermined degree of agreement includes a lower predetermined degree of agreement and an upper predetermined degree of agreement that is greater than the lower predetermined degree of agreement. The classification of the palm vein pattern is done as true if the degree of matching is less than the lower predetermined degree of agreement and not true if the degree of agreement is greater than the upper predetermined degree of agreement. With this embodiment, an improved check is possible as to whether a venous pattern is a true venous pattern or a forgery by classifying a venous pattern as genuine only if the comparison of the infrared image and the light image gives a degree of agreement sufficient to one of the two classifications suggesting. In a further development of this embodiment, issuing a warning signal or outputting a request for re-presentation to a user takes place when the degree of agreement lies between the lower predetermined degree of agreement and the upper predetermined degree of agreement. Preferably, the first processor unit for comparing the infrared image and the light image for performing a difference image method or a comparison of the information content of the infrared image and the light image or a comparison on the basis of selected points is formed. In one embodiment, the first processor unit is configured to determine the degree of agreement via a threshold value analysis.

Furthermore, when the first light source is formed, it is advantageous to use electromagnetic waves having a wavelength between 820 nm and 900 nm, e.g. of 850 nm. At this wavelength, the absorption of electromagnetic waves by the hemoglobin is particularly strong, so that venous patterns are particularly easy to recognize. Furthermore, when the second light source is formed, it is preferable to form electromagnetic waves having a wavelength between 450 nm and 500 nm, e.g. of 470 nm.

In one embodiment of the sensor system, the first and the second light source in the form of a broadband light source are formed with a first and a second filter. In this case, the first filter is designed to allow only electromagnetic waves with wavelengths in the near infrared range to pass through and the second filter is designed to allow only electromagnetic waves to pass at least one further wavelength, in particular in the region of visible light. In this case, the first and the second filter can be formed either as part of the broadband light source or as part of the camera. With these embodiments, it is made possible, by changing the filter, to adapt changes in the illumination scenarios between defined wavelengths to illumination scenarios to particular circumstances in which the sensor system is used. In addition, the use of filters with broadband light source in place of light sources of narrowly defined radiation characteristics often cheaper and the structure of the sensor system can be made much more compact using only one light source.

The first and second camera chips are advantageously designed together as a common camera chip, ie both functions are on a common camera chip integrated. This allows the construction of a more compact system. In particular, the common camera chip may be a single camera chip sensitive to both visible and infrared light.

In an advantageous embodiment, the first and the second light source are designed to emit electromagnetic waves alternately in direct succession from image to image or partial images, so that immediately behind each other an infrared image (first light source) and a normal image (second light source) be recorded. The first and / or second or the common camera chip is in particular a semiconductor camera chip. Advantageously, the sensor system comprises an optical double band pass filter in front of the first and second or common camera chip, which is transparent only in the wavelengths of the first and second light source and all other wavelengths especially disturbing background light, such as solar radiation hides. The wavelengths of the first and second light sources can also include narrow wavelength ranges.

In one embodiment, the processor unit is additionally configured to cause an output of a warning signal when classifying the manual vein pattern as not real. This warning signal can be output, for example, in the form of an acoustic signal, which possibly indicates security personnel present on an unwanted access attempt. In a development, however, the warning signal can also be output in the form of a signal to electronic security systems which, for example, prevent further access to a downstream system to be protected for a predetermined period or initially block the sensor system for a predetermined period of time. In a preferred embodiment, a release unit connected to the sensor system is provided, which is designed to cause classifications of the vein pattern as genuine another classification process or to allow access to a downstream system. In this embodiment, the classification of the vein pattern is used as the real hurdle for either further checking of the true vein pattern or in systems where it is only a question of granting access to people as the sole access hurdle. Preferably, the first processor unit is additionally designed to carry out the further classification process, which includes an image preprocessing of the infrared image, a feature extraction and a classification algorithm which classifies the venous pattern as sufficiently matching or not sufficiently matching. Advantageously, the first processor unit is designed to calculate a distance from the reflection signals in a sleep mode of the sensor system, in which only the second light source emits electromagnetic waves at regular time intervals, and to start irradiation with the first light source when falling below a predetermined minimum distance. This allows an efficient operation of the sensor system, since the irradiation with the first light source and the subsequent generation and evaluation of the light images are only carried out when an object is in the irradiated area.

In a preferred embodiment, when the vein pattern is classified as true, the first processor unit is additionally adapted to compare the infrared image with at least one stored vein pattern or deposited feature vector corresponding to the vein pattern, and the true vein pattern matches as matching or not sufficiently matching a pattern matching degree to classify. In this case, the release unit is additionally designed to allow access to a downstream system in the case of a vein pattern classified as sufficiently coincident. With this embodiment, a double access control is possible by first having to classify a vein pattern as genuine and, in the subsequent step, as sufficiently matching in order to then possibly provide access to downstream systems.

In a preferred embodiment, the sensor system comprises an additional topography sensor for detecting three-dimensional topographies and a second processor unit connected to the first processor unit and the topography sensor and configured to operate from the infrared image of the camera by illumination with the first light source and the three-dimensional topography data of the topography sensor to generate a normalized venous pattern of a hand or a feature vector corresponding to the venous pattern. In the processor unit, the topography data and the vein pattern are linked in such a way that a standardized vein pattern as well as a normalized hand geometry image, ie a normalized topography of the hand, can be calculated. It is therefore a location normalization in space. Both hand geometry and venous ramification, ie venous patterns, are both spatial structures whose 2D images on a camera chip are dependent on their spatial position (twisting, tilting, curvature, etc.). A detection system based on this embodiment of the sensor system is therefore tolerant to the hand position and does not require any manual hand supports in order to ensure a "correct" positioning of the hand.

In a preferred development, the release unit is then designed to start the topographical sensor in the presence of a true venous pattern. This provides an efficient system by only capturing topographical data and starting a second safety check when it is certain that the venous pattern is a true venous pattern rather than a forgery, and second, the topography sensor does not need to be kept in continuous operation, as long as no venous pattern was recognized as genuine.

Advantageously, the first processor unit is additionally designed to compare the normalized venous pattern generated by the second processor unit or the generated feature vector with at least one stored venous pattern or at least one stored feature vector, and the generated normalized venous pattern or the generated feature vector as sufficiently coincident or not -sufficiently consistent to classify. Furthermore, it is advantageous if this release unit is additionally designed to allow access to a downstream system in the case of a vein pattern or feature vector classified as sufficiently coincident. This makes it possible to test hand vein patterns in a particularly secure manner since, on the one hand, the release is checked in two steps, namely first in checking the authenticity of the vein pattern and then in a check of sufficient agreement with stored vein patterns, and at the same time the device comes without Specifications for the introduction of the hand, so is tolerant of the hand position.

In a further embodiment, the second processor unit is additionally designed to determine, based on the topography data and the infrared images, whether the venous pattern is a venous pattern of a carpal, a palm of the hand or a palm of the hand and to output this information to the first processor unit. With such an embodiment, it is possible for the system to achieve even greater independence from the position of the hand during the check, by allowing the system to determine on its own whether the true venous pattern should be compared with stored venous patterns from the carpal, palm, or palm der Handaußenfläche originate. On the one hand, a further safety mechanism can be built in by requiring access to the downstream system to have two vein patterns classified as sufficiently safe, for example one of the palm of the hand and one of the palm of the hand. On the other hand, however, systems that only require a sufficiently consistent venous pattern can be made even more tolerant of the hand position, if there are venous patterns of the carpal, palmar or palm of the hand, so that, for example, an examination can also be performed if the palms or only the wrist is facing the sensor and not the palm of the hand. In an alternative embodiment, the topography sensor is a time-of-flight sensor. This determines for each pixel, the distance of an object from the sensor by the transit time of an infrared light pulse is detected and calculated by the speed of light, the distance. Furthermore, it is advantageous if the first processor unit is a host processor, in particular an RFID (Radio Frequency Identification) host processor, and the comparison takes place with a vein pattern or feature vector stored in an RFID slave processor. In this case, classification is only possible if a connection channel has been established between the RFID host processor and an RFID slave processor. This allows identification without the storage of data in a database system and thus greater security, since both the valid vein pattern and the RFID slave processor must be present for identification on the sensor system.

The alternative embodiment of the RFID host processor can be replaced by another type of connection processor that realizes at least one active radio link capability, preferably according to the known standards WLAN, Bluetooth, ZigBee or NFC. In this case, the slave processor is also provided with at least the same active radio communication capability as the host processor. The comparison also takes place, as described with respect to RFID, with a vein pattern or feature vector stored in the slave processor.

But also the comparison with venous patterns or feature vectors stored in a database is advantageous. Not every user has to be equipped with an additional identification element, but he can make the classification on the basis of the already existing pattern. In a development of the invention, the sensor system is mounted on a keyboard. Here, a venous pattern is detected in the back of the hand. This allows easy identification even during an input process. In this case, the first processor unit of the sensor system may be designed and connected to the keyboard, that whenever a key is struck, a verification process of the vein pattern can be triggered simultaneously, so the lighting can be started with the first light source. The input of this character or a group of characters is only recognized as valid if a genotype classified as genuinely is detected. If, in addition, it is to be ensured that only certain users receive access, we will only recognize the input as valid if, in addition, the vein pattern has been classified as sufficiently matching.

The advantage of such an arrangement is that a permanent authentication of the user can be ensured in a communication process with a computer, for example, and thus it can be permanently ensured that only authorized users can access the computer or the system connected to it.

Equally advantageously, however, the sensor system can also be integrated in a keyboard and use a venous pattern of the palm of the hand for identification. The keyboard is in this case designed so that the palm of the hand can always hover over the user-facing part of the keyboard or rest on this, while the fingers can reach the keys. So it is chosen a deliberately narrow arrangement of keys. Number keys or other function keys are created either to the right or to the left of the letter keys. In the user-facing part of the keyboard, an inventive sensor system is integrated in an embodiment for the left and right hand respectively. Depending on the security requirement, a release of downstream systems can be activated on detection of a true and sufficiently matched vein pattern by only one hand or both hands. The sensor system is preferably designed such that it can detect the vein structure with a very short distance to the palm or carpal, it is capable of the palm with an extension of up to 120 mm at a distance from the sensor surface of 5 mm to 20 mm capture.

In one exemplary embodiment, the sensor system comprises a plurality of first light sources which are embedded in the computer keyboard in such a way that each first light source illuminates a full area of a section of the palm of the hand or of the carpal. In a further embodiment, the sensor system comprises a plurality of second light sources for visible light, which are also embedded in the keyboard so that every other light source fully illuminated a section of the palm or the carpal. In addition, this embodiment comprises a plurality of first and second camera chips for receiving and converting the respective sections of the palm vein pattern. In this case, the arrangement of the camera chips and the light sources is selected so that the combination of all the light images or infrared images of the camera chips produces a largely complete image of the palm or the carpus. This complete image can then be supplied to a processing process for determining the venous pattern template as a whole. The arrangement of several camera chips results from the need for a very short image distance. When using only a first and second camera chips, it may sometimes happen that the palm of the hand can not be fully optically recorded and thus a complete image of the vein pattern can not be made available for processing. In one embodiment of the method, a warning signal is output when classifying the vein pattern as not real. This allows the initiation of further security measures due to the detection of an unwanted access attempt.

Furthermore, it is advantageous if, after classification of a vein pattern, real further processes are released. It is particularly preferred that the further processes comprise the following steps:

 Recording topographical data of the hand, and

 Generate a normalized venous pattern of the hand or a feature vector corresponding to the venous pattern from the infrared images and the topography data. Generating can be done by calculating the corresponding patterns or feature vectors.

Preferably, a comparison of the generated normalized vein pattern or the generated feature vector with at least one stored vein pattern or feature vector and a classification of the generated vein pattern or the generated feature vector as sufficient matching or not sufficiently in accordance with the stored vein pattern or feature vector. In a further development of the method, after a classification of a vein pattern or a feature vector, additional processes are released as sufficiently coincident. Thus, a multi-level access control process is feasible. Further exemplary embodiments of the method according to the invention and of the device according to the invention are explained below with reference to the figures. Show it:

1 shows a flowchart for an embodiment of the method according to the invention; FIG. 2 shows a schematic representation of an embodiment of a sensor system according to the invention for testing a hand vein pattern; FIG.

Fig. 3 is a schematic representation of another embodiment of a sensor system according to the invention.

1 shows a flowchart of a method for testing palm vein patterns. In step S1, in this case, a hand or an object introduced into the beam path is irradiated with a first light source which emits full-surface electromagnetic waves with wavelengths in the near infrared range, which are absorbed by hemoglobin. In step S2, reflection signals of electromagnetic waves with wavelengths in the near infrared range are recorded and converted into a corresponding infrared image. The recording and conversion is done by means of a first camera project. In step S3, the hand or object is irradiated over the entire surface with a second light source which emits electromagnetic waves having wavelengths in the range of visible light, and reflection signals of electromagnetic waves with wavelengths in the visible light range are recorded in step S4 and converted into a corresponding light image.

In a preferred embodiment, step S3 takes place immediately after step S1, so that there is an immediate sequence of image-to-image or partial images and the hand or object is in the same position in both irradiations. In step S5, the infrared image and the light image are compared, and a degree of coincidence between the light image and the infrared image is detected. The comparison can be made, for example, by differential image methods or a comparison of the information content of the infrared image and the light image or a comparison on the basis of selected points. In step S6, a vein pattern is classified as true if the degree of matching is less than a predetermined degree of agreement and not true if the degree of agreement is greater than or equal to the predetermined degree of agreement. The degree of correspondence in this embodiment will be represented by a multi-dimensional vector giving a similarity measure. About the classifi- It ensures that the object being irradiated is a true hand rather than a forgery in which a vein pattern is only imprinted or otherwise applied. Optionally, after classification of the vein pattern, a further process, for example further access controls in which the true venous pattern is compared with stored venous patterns, and after that a final release or the allocation of an access authorization, can be connected in step S7.

If the light image and the infrared image coincide to a great extent, the vein pattern is therefore not classified as genuine, so a warning signal can be output in step S8. This warning signal can be output, for example, as an acoustic warning signal to security personnel in the vicinity of the sensor system or an access control system in which the sensor system is integrated, so that the unauthorized access attempt is reported. Alternatively, however, the warning signal can also be configured in such a way that it is output to further security systems, which then block access devices electronically, for example for certain periods, or also block access to the sensor system for a predetermined period of time.

As steps of further access control, for example, the following steps not shown here are suitable: recording of topography data of the hand after the venous pattern has been classified as genuine and calculating or generating a normalized venous pattern of the hand or a feature vector corresponding to the venous pattern from the infrared image and the topography data and a subsequent comparison of the generated normalized vein pattern or the generated feature vector with at least one stored vein pattern or feature vector and classification of the generated vein pattern or the generated feature vector as sufficiently coincident or non-sufficiently matching depending on the degree of actual match. After an identification of the vein pattern or the feature vector as sufficiently coincident then further additional processes can be released, such as access to an electronic system. FIG. 2 shows an embodiment of a sensor system according to the invention for testing palm vein patterns. The sensor system comprises a first light source L1, a second light source L2, a camera with a first camera chip C1 and a second camera chip C2 and a processor unit P1. The first light source Lides sensor system 100 is formed in the operation of the entire surface electromagnetic waves Emitting wavelengths in the near infrared range that are absorbed by hemoglobin. The second light source L2 is designed to emit electromagnetic waves of wavelengths in the range of visible light over its entire area during operation. The camera has a first camera chip C1 and a second camera chip C2. In this case, the first camera chip C1 is designed to receive reflection signals of electromagnetic waves with wavelengths in the near infrared range and to convert them into a corresponding infrared image. The second camera chip C2 is designed to receive reflection signals of electromagnetic waves with wavelengths in the visible light range and to convert them into a corresponding light image. The processor unit P1 is connected to the first and second camera chips and configured to compare the light image and infrared image, to determine a degree of coincidence between the light image and the infrared image, and to classify the vein pattern as true if the degree of matching is less than a predetermined degree of matching and Classify venous patterns as non-genuinous if the degree of congruence is greater than or equal to the predetermined degree of agreement.

By means of this sensor system, it is possible to identify an object H in the irradiated area as a hand with a true venous pattern or as a forgery and thus prevent access to downstream systems via a counterfeiting. In one embodiment, the processor unit P1 is further configured to cause an output of a warning signal when classifying a vein pattern as being non-genuine. In a further embodiment, a release unit, not shown here, is connected to the sensor system, which is designed to allow classification of the vein pattern as genuine access to a downstream system. As an alternative to the embodiment shown here, in which the first and the second camera chip are each formed as a separate chip, in an alternative embodiment, a single camera chip can combine the functions of the first and the second camera chips, so that the first and the second camera chip are identical.

3 shows a schematic representation of a further exemplary embodiment of a sensor system according to the invention. The basic structure of the sensor system in Fig. 3 is similar to the system shown in Fig. 2, therefore, the differences are described below in more detail mainly. In FIG. 3, the sensor system 200 additionally comprises a topography sensor TS for detecting three-dimensional topographies and a second processor unit P2 connected to the first processor unit P1 and the topography sensor TS. The second processor unit P2 is formed during operation from the infrared image of the camera, ie the camera chip C1, when illuminated with the first Light source L1 and the three-dimensional topography data of the topography sensor to generate a normalized venous pattern of a hand or a feature vector corresponding to the vein pattern. Similar to the sensor system 100 of FIG. 2, the processor unit P1 is advantageously additionally designed as a vein pattern generated or calculated by the second processor unit or as a generated or calculated feature vector with at least one stored vein pattern or with at least one stored feature vector and classify the generated normalized vein pattern or the generated feature vector as sufficiently coincident or non-sufficiently matching.

In one embodiment, the sensor system 200 additionally has a release unit which, on the one hand, is designed to start the topographical sensor when the light pattern and the infrared image are sufficiently close, ie when classifying a vein pattern as genuine, and the other is additionally designed, one subsequently corresponding as sufficient classified vein pattern or feature vector to allow access to a downstream system. The release unit is not shown in FIG. In one embodiment, in a rest mode of the sensor system in which only the second light source emits electromagnetic waves at regular intervals, the first processor unit P1 is further configured to calculate a distance from the reflection signals and to irradiate the first light source when it falls below a predetermined minimum distance start. The embodiment of the sensor system 200 is particularly advantageous for examining venous patterns when the area in which the hand to be detected is to be as free as possible, that is, no restrictions for the position of the hand should be specified. By standardizing the vein pattern using the topography sensor, the system is tolerant to the position of the hand. In particular, systems can be realized which are suitable both for venous patterns in carpal roots as well as palms and hand outer surfaces.

Claims

claims A sensor system for testing a venous pattern comprising: a first light source, which is designed to emit in its entirety electromagnetic waves having wavelengths in the near infrared range which are absorbed by hemoglobin during operation, a second light source, which is designed, during operation, to emit electromagnetic waves with at least one further wavelength over its entire surface, in particular in the region of visible light, a camera with a first camera chip, which is designed to receive reflection signals of electromagnetic waves with wavelengths in the near infrared range and to convert them into a corresponding infrared image and with a second camera chip which is formed reflection signals of electromagnetic waves having the at least one further wavelength, in particular in the region of visible light and convert it into a corresponding light image, - a first connected to the first and the second camera chip processor unit which is adapted to compare the light image and the infrared image, to determine a degree of agreement between the light image and infrared image and the vein pattern as true classify when the degree of matching is less than a predetermined degree of coincidence and classify the vein pattern as not genuine if the degree of coincidence is greater than or equal to the predetermined degree of agreement is. 2. Sensor system according to claim 1, characterized in that the first processor unit for comparing the infrared image with the light image for performing a difference image method or a comparison of the information content of the infrared image and the light image or a comparison based on selected points is formed. 3. Sensor system according to one of the preceding claims, characterized in that the first and second light source in the form of a broadband light source are formed with a first and a second filter, wherein the first filter is formed to pass only electromagnetic waves with wavelengths in the near infrared range leave and the second filter is designed to allow only electromagnetic waves with wavelengths in the visible light range to pass. 4. Sensor system according to one of the preceding claims, characterized in that the first and second camera chip are formed together as a common camera chip. 5. Sensor system according to one of the preceding claims, characterized in that the first processor unit is further configured to cause when classifying the Handvenenmusters as not real an output of a warning signal. 6. Sensor system according to one of the preceding claims, characterized in that there is provided a release unit connected to the sensor system, which is designed to cause classification of the vein pattern as genuine another classification process. 7. The sensor system of claim 6, wherein the first processor unit is further configured to perform the further classification process comprising image processing of the infrared image, feature extraction, and a classification algorithm that classifies the vein pattern as sufficiently coincident or not sufficiently coincident. 8. The sensor system according to claim 1, wherein the first processor unit is additionally configured, when classifying the vein pattern as true, to compare the infrared image with at least one stored vein pattern or stored feature vector corresponding to the vein pattern, and the real vein pattern is dependent on a pattern matching degree to classify as sufficiently consistent or insufficiently consistent and in which the enabling unit is additionally adapted to allow access to a downstream system in a vein pattern classified as sufficiently consistent. 9. Sensor system according to one of claims 1 to 7 with an additional topography sensor for detecting three-dimensional topographies and a second with the first processor unit and the topography sensor connected processor unit which is formed, in operation from the infrared image of the camera when illuminated with the first light source and the three-dimensional topography data of the topography sensor, a normalized venous pattern of a hand or the vein pattern corresponding Feature vector to generate in which the first processor unit is additionally designed to compare the normalized venous pattern generated by the second processor unit or the generated feature vector with at least one stored venous pattern or at least one stored feature vector and the generated normalized venous pattern or the generated feature vector as sufficiently classified or insufficiently consistent to classify and in which the release unit is additionally adapted to allow access to a downstream system in a vein pattern or feature vector classified as sufficiently consistent. 10. The sensor system according to claim 8, wherein the second processor unit is additionally configured to use the topography data and the infrared image to determine whether the venous pattern is a venous pattern of a carpal, a palm, or a palm exterior, and to pass this information to the human output first processor unit. 1 1. A method for testing a venous pattern comprising: full exposure of a hand or part of the hand to electromagnetic waves with near infrared wavelengths absorbed by hemoglobin, Recording reflection signals of electromagnetic waves having wavelengths in the near infrared range and converting them into a corresponding infrared image, full-surface irradiation of the hand or of a part of the hand with electromagnetic waves having at least one further wavelength, in particular in the region of visible light, Recording reflection signals of electromagnetic waves having the at least one further wavelength and converting into a corresponding light image, Comparing the light image and the infrared image, Determining a degree of agreement between the photograph and the infrared image and Classifying the palm vein pattern as genuine if the degree of agreement is less than a predetermined degree of agreement and not true if the degree of agreement is greater than or equal to the predetermined degree of agreement. 12. The method of claim 1 1, characterized in that are released after classifying the vein pattern as real further processes. 13. The method according to claim 12, characterized in that the further processes comprise the following steps: - Recording topographical data of the hand Generating a normalized venous pattern of the hand or a feature vector corresponding to the venous pattern from the infrared image and the topography data. 14. The method according to claim 13, characterized in that a comparison of the generated normalized vein pattern or of the generated feature vector with at least one stored vein pattern or feature vector is performed and the generated vein pattern or the generated feature vector is classified as sufficiently concurrent or not sufficiently matching. 15. The method according to claim 14, characterized in that after a classification of a vein pattern or a feature vector as sufficiently matching additional processes are released.