CH713061B1 - System and method for contactless biometric authentication. - Google Patents

Abstract

The invention relates to a biometric authentication system for capturing biometric identification data from a body part of a person and for comparing the captured biometric identification data with registered biometric identification data already captured from the body part for authenticating the person, comprising: a lighting unit (110) for illuminating the body part; an image capturing unit (114) for capturing an image of the illuminated body part; a processing unit (118) for extracting the biometric identification data from the image captured by the image capture unit (114); and an authentication unit (120) for comparing the extracted biometric data with registered identification data for authenticating the person; wherein the lighting unit (110) is provided with at least one pattern mask (112) in order to project a mask pattern onto the body part so that the captured image includes the projected mask pattern.

Description

Technical field of the invention

The present invention relates to the field of biometric authentication and in particular to a system and method for contactless biometric authentication, the authentication method also checking the liveliness of the person to be authenticated.

General state of the art

Authentication methods have found widespread use and application in the field of biometric authentication. The need to authenticate people has been around for a long time. The events that create such a need include financial transactions, entering foreign countries, voting, taking an exam, starting a business, access to buildings, etc. Biometric authentication methods use the specific and unique characteristics of a particular body part of an individual . Fortunately, humans are made up of complex organ systems that have a high degree of uniqueness. The most common organic feature used for authentication purposes is fingerprints. Other methods use the iris of the human eye, blood vessel patterns under the skin of the hand, or the face. These methods are usually based on image processing and recognition. Still other methods involve the use of voice.

One of the most frequently used biometric authentication is based on the fingerprint, but there is a risk of circumvention through the use of false / artificial fingerprints. Another problem with fingerprint authentication is that the user has to have physical contact with the surface, which can result in hygiene issues if the same authentication device is used by several people or if dirt or grease deposited on the surface can lead to a malfunction of the authentication device.

Other authentication methods, also contactless, use the handprint, palm vein or finger vein images. Examples are known from US20080107309, EP2244224, US9355236, EP1612718, US9223955, US6813010, US7359531, EP1612717 or EP1387309.

There is a need for contactless biometric authentication, the authentication method also checking the liveliness of the person to be authenticated. Some systems use 3D cameras or multiple cameras or even modulated or structured projector light sources for these operations. Because these systems and devices are typically very complex, there remains a need to perform these operations in a more economical and robust manner with minimal system design.

Summary of the invention

It is an object of the invention to provide a system, a device and a method to carry out a contactless biometric authentication, which has the possibility of checking the liveliness, in a cost-efficient and robust manner.

This is achieved by a biometric authentication system according to claim 1, a biometric authentication method according to claim 8 and a biometric authentication device according to claim 13.

The biometric authentication system for acquiring biometric identification data of a body part of a person and for comparing the acquired biometric identification data with registered biometric identification data already acquired from this body part for authenticating the person, comprising: a lighting unit for illuminating the body part; an image capturing unit for capturing an image of the illuminated body part; a processing unit for extracting the biometric identification data from the image captured by the image capture unit; and an authentication unit for comparing the extracted biometric data with registered identification data for authenticating the person. The lighting unit of the system is provided with at least one pattern mask in order to project a mask pattern onto the body part, so that the mask pattern is part of the captured image.

The projected mask pattern, which preferably consists of identically shaped polygons of the same area, can be used to find the optimal distance between body part and lighting or detection unit in a completely contactless manner by the average size of the individual pattern shapes projected onto the body part the size of the individual pattern shapes obtained during a previous calibration step is compared. The same mask pattern that is projected onto the body part and is therefore part of the captured image can also be used to define an area of interest of the body part and to extract the biometric characteristics by determining the individual pattern shapes, the structures of interest of the body part ( e.g. handprint and palm veins). In addition, the same mask patterns can be used to conduct a liveliness test based on the unevenness of the body part. Unevenness of the body part leads to a distortion of the mask pattern. This distortion can hardly be reconstructed from a false body part and can therefore be used as an indication of the liveliness of a person to be authenticated.

Further embodiments of the invention are set out in the dependent claims.

In some embodiments, the pattern mask can have a regular pattern of polygons, preferably triangles, rectangles, squares, pentagons or hexagons, that is, the polygons have identical shapes and areas.

In some embodiments, the lighting unit can comprise at least one light source of visible light and / or at least one light source near infrared light (NIR), wherein the light source preferably comprises a plurality of light emitting diodes (LEDs). The light source or LEDs can be arranged in a circle around the detection unit. Different light sources can be arranged alternately. The lighting unit preferably comprises light sources for visible and for NIR light, for example in order to detect handprints / wrinkles or palm veins. In this way, the authentication analysis can be based on polygons which include handprints and palm veins.

In some embodiments, the biometric authentication system can further comprise an optical guide unit which is able to determine the optimal distance between the body part and the detection unit on the basis of the mask pattern projected onto the body part. The optical guide unit can display the optimal distance, for example by means of a green light. If the distance to the lighting / detection unit is too short or too great, a red light can light up. In addition, as the optimum distance is approached, a gradual change from red to yellow to green can be provided. In this way, the user receives information about whether he is moving his hand in the right direction.

In some embodiments, the detection unit can comprise a microlens array for obtaining depth information of the body part. The microlens arrangement is an optional function, since the mask pattern projected onto the body part already provides certain depth information due to the distortion of the pattern.

The invention further relates to a method for capturing biometric identification data of a body part of a person and for comparing the captured biometric identification data with registered identification data already captured by this body part for authenticating the person, comprising the steps of: illuminating the body part using a lighting unit that is provided with at least one pattern mask for projecting a mask pattern onto the body part; Capturing an image of the illuminated body part including the projected mask pattern using an image capturing unit; Extracting the biometric identification data from the image captured by the capture unit using a processing unit; Comparing the extracted biometric data with registered identification data for authenticating the person using an authentication unit.

In some embodiments, the pattern mask can have a regular pattern of polygons, preferably triangles, squares, squares, pentagons or hexagons, and is used to determine the optimal distance between the body part and the detection unit.

In some embodiments, the mask pattern is used to guide the body part in order to determine an optimal distance between the body part and the detection unit. This can be done by comparing the average size / area of the polygons of the projected mask pattern with the size / area of the polygons obtained during a calibration step.

In some embodiments, the lighting unit can illuminate the body part with visible light and NIR light in order to obtain combined images that show both structures of the body part under visible light (e.g. handprint) or under NIR light (e.g. palm veins) are recognizable, as well as include the mask pattern. The combined image is then used to extract the biometric identification data.

In some embodiments, the mask pattern can be used to extract the biometric identification data. For example, the mask pattern divides the image into several small areas / areas, which can be analyzed with regard to the presence of biometric characteristics (e.g. handprint and / or palm veins).

The invention also relates to a biometric authentication device for capturing images for performing the method described above. The device comprises a lighting unit for illuminating the body part; an image capturing unit for capturing an image of the illuminated body part; wherein the lighting unit is provided with at least one pattern mask in order to project a mask pattern onto the body part, so that the captured image includes the projected mask pattern.

In some embodiments the device further comprises a processing unit or means for connecting the device to a processing unit. The processing unit extracts the biometric characteristic data from the image captured by the capture unit.

In some embodiments, the device further comprises an authentication unit or means for connecting the device to an authentication unit. The authentication unit compares the extracted biometric data with registered identification data to authenticate the person.

According to an extended embodiment, the authentication can be started or signaled by up-and-down movement or forward / backward movement or lateral movement of the palm or any independent pan, nod or roll movement. Thereafter, the area of interest is defined as having the presence of palm wrinkles and palm veins, which are then detected in visible and NIR light, as illustrated in accordance with an embodiment of the invention.

According to another extended embodiment, a 3D model of the palm of the hand is generated by the stereoscopic effect when the hand is moved up and down or in any of the panning, pitching or rolling movements, whereby a series of images with different LED- Patterns at different focal lengths are created to create a 3D image of the palm and veins. This can also be used for the liveliness test.

According to another expanded embodiment, an area of interest is determined using the following steps. The first step is to determine the wrist position of the hand to be authenticated. Once the wrist position has been determined, the areas are shown as rays. In another scenario, the next step after determining the wrist position is determining the thumb position and radiating the areas in order to connect the thumb to the other fingers and to connect all the joints to one another.

According to another extended embodiment, the palm is in a first position in which the side of the palm is visible to the sensor, i.e. the back of the hand and palm side are at right angles to the sensor. The palm of the hand is rotated in such a way that it is arranged parallel and directed towards the surface of the sensor. When performing the twisting motion, the outermost visible points become the area of interest as the palm of the hand is flexed. This can also be used to conduct a liveliness test.

According to another expanded embodiment, for a liveliness test, the user is asked to close his palm into a fist and the extraction of the area of interest is marked along the closing finger grooves on the top to draw a square. This action is used as a liveliness test whenever there is a risk of incorrect palm authentication. Attempts to hack the system through the use of a fake / artificial palm may randomly ask a user to perform this gesture as a liveliness test.

According to another expanded embodiment, for a liveliness test, the user is asked to open and close their fingers and the extraction of the area of interest is marked along the active finger ridges on the top to draw a polygon. This action is used as a liveliness test whenever there is a risk of incorrect palm authentication. Attempts to hack the system through the use of a fake / artificial palm may randomly ask a user to perform this gesture as a liveliness test.

The present invention provides a non-contact biometric authentication system. In order to ensure a more hygienic method, the present invention does not require that the user place his hand on a contact surface, whereby the authentication method guarantees that the user does not have to touch the device anywhere during the authentication process. Furthermore, the invention discloses various methods for checking the liveliness of the authenticating person. This helps deter fraudulent practices in the use of hand or finger biometric authentication.

Brief explanation of the figures

The foregoing and other features of embodiments will become more apparent from the following detailed description of the embodiments in conjunction with the accompanying drawings. Elements in the figures do not necessarily appear to be drawn to scale for the sake of clarity and understanding of these various elements and embodiments of the invention. Therefore, in the interests of clarity and clarity, the drawings are in generalized form, wherein: FIG. 1 is a block diagram of a system for contactless biometric authentication. Fig. 2 shows a device comprising a lighting unit and a detection unit. Fig. 3a shows an image acquired using a NIR light source without a pattern mask. 3b shows an image with a pattern mask captured by using a NIR light source. Fig. 4 shows a calibration system in which the optimal position of the hand is determined. Figure 5a illustrates a manner in which a palm is positioned in front of the lighting unit, as well as the manner in which the light from the palm is scattered. Fig. 5b illustrates a function of an optical guide unit. Figure 6 illustrates different positions into which the palm can be moved. Figure 7 illustrates a method of determining an area of interest on a palm. Figure 8 illustrates a method in which the features of a palm are extracted in accordance with an embodiment of the invention. 9 shows various feature points in the area of the palm. Figure 10 illustrates a neural network of a plurality of neurons. 11 illustrates a mask pattern made up of hexagons (left) and a distorted mask pattern due to uneven body parts (right). Figure 12 illustrates one possible way of defining pattern blocks using the pattern mask.

Description of the invention

In the following detailed description, reference is made to the accompanying drawings which form a part hereof, and in which there are shown certain illustrative embodiments that may be practiced. These embodiments are described in sufficient detail to enable those skilled in the art to practice the invention, and it is understood that other embodiments can be used and that changes can be made in logic, mechanical, and other changes within the scope of the embodiments. Thus, the following detailed description is not to be considered as limiting the scope of the invention; instead, the invention is to be defined by the appended claims.

The present invention avoids the disadvantages of the prior art by providing a contactless biometric authentication system which is used to authenticate the person to be authenticated without using 3D cameras or a projector system and also to check their liveliness.

1 shows a block diagram of a system for contactless biometric authentication in accordance with an embodiment of the invention. The system comprises at least one lighting unit 110 comprising at least one light source of visible light and / or at least one light source of near infrared light (NIR) . The light source can comprise a plurality of light emitting diodes (LEDs). The LEDs can be arranged in a circle. Visible light LEDs and NIR light LEDs can be arranged alternately.

The lighting unit 110 or its individual light sources are masked with at least one pattern 112. The pattern is preferably defined as a regular group of polygons of the same area (see Fig. 11, left). The size of a polygon is preferably in the region of 1 mm or has an area of approximately 1 mm 2. Various types or shapes of polygons can be used, for example triangle, square, square, pentagon, or hexagon. Other shapes are also possible. Pattern masks on different LEDs can be arranged at different angles to one another. If, for example, a pattern mask is arranged on two LEDs furthest away in a circular arrangement, then these can be used for calculation purposes, since the distances between them are triangulation for calculating the optimal distance of the body part from the lighting or detection unit or for determining the Form area extension of the body part. The LED lights with different pattern masks are used to indicate the optimal distance at which the hand must be placed over the sensor of the registration unit. This arrangement also helps to define the area of interest by clearly recognizing the hand against the rest of the background.

The system further comprises an image acquisition unit 114 for acquiring the image of a body part, for example a hand, of the person to be authenticated. The image acquisition unit 114 has a camera and one or more lenses 116. The acquisition unit can have a microlens array in order to acquire depth information of the body part, which provides images with different focal lengths or deformation of the surface, and thus 3D information. The system also has a processing unit 118 for processing the images and for extracting various biometric characteristics, such as palm folds / imprints and / or palm veins, from the captured image. The system also has an authentication unit 120, the authentication unit 120 comparing the biometric identification data extracted from the captured image with previously extracted and registered identification data for authenticating the person. The registered data can be stored in a storage unit of the authentication unit or in its own storage unit.

The units described can all be part of a single device. The processing unit, the authentication unit or the storage unit can each also be separate from the device and connected to one another or to the device via communication means. 2 shows a device comprising a lighting unit 110 together with an image capturing unit 114. In this case, the light sources are arranged in a circular ring. The LEDs with NIR and those with a visible light source are arranged alternately.

In accordance with one embodiment, the invention further discloses a method for contactless biometric authentication. The authentication is performed by capturing images by an image capturing unit 114. The image capturing unit 114 may use a 2D camera (hereinafter referred to as a camera). The camera captures the masked image of the hand from the masked light source. If it is provided with a microlens array, the camera can also capture a depth image by building up the image created by the microlens array (FIG. 11). The microlens array is used to extract different focal lengths of the same image and to extract the depth image or surface deformity information of the body part therefrom.

3a shows an image captured by a NIR light source without a pattern mask, while FIG. 3b shows the image captured with a pattern mask.

When using a pattern mask, a mask pattern appears as a geometric pattern that can be predetermined on a flat surface, for example. A manual or automatic calibration test can be carried out for the camera without the presence of the body part in order to optimally image the area of interest by arranging a flat surface at the desired distance from the lighting unit and the detection unit. After manual or automatic calibration, a visual guide instrument or guide indicates the distance at which the hand should be placed based on the predetermined geometric pattern size of the hand. The guide unit is explained in more detail below.

4 shows a calibration system in which the optimal position of the palm is determined on the basis of the mask pattern projected onto the palm. Here, a contactless biometric authentication device 100 comprising at least the lighting unit 110 with a sample mask 112 and the detection unit is placed at a predetermined position in relation to a frame 511 of a calibration device 500. The frame 511 holds a pre-attached flat plate 512 a known distance with respect to the predetermined position of the authentication device. The image processing unit captures the mask pattern on a pre-attached flat plate 512 at a known distance. The image processing unit 118 detects the mask pattern and stores the pattern information (pattern area, pattern size, etc.) for future recognition of the optimal distance of a palm for authentication. When a palm is later placed in front of the authentication device, the image processing unit 118 detects the mask pattern on the palm, extracts the pattern information and compares it with the pattern information previously stored during the calibration.

After the calibration, the optical guide unit shows the height at which the palm of the hand must be placed on the basis of the change in size of the mask pattern, which is compared with the previously stored calibration mask pattern. If the palm is too close (FIG. 5b, middle) or too far from (FIG. 5b, left) the detection unit, a red-colored light appears, preferably projected onto the palm. When the palm is placed at the optimal distance (the actual mask pattern size corresponds to the calibration mask pattern size), a green colored light appears (Fig. 5b, right), preferably projected onto the palm. The user notices the color that, when there is a gap between the fingers, is reflected around the fingers, or around the whole palm. Or, if there is no gap between the fingers, the user will notice the color of the light from the light reflected from the palm back to the sensor area, which the user is looking at in side or profile view. In addition, the colored light can gradually change from red to yellow and finally to green, whereby it indicates whether the palm is being moved in the correct direction away from or towards the detection unit. This way the correct position can be easily found. For example, when the color changes from yellow to red, the user realizes that they are moving the palm in the wrong direction.

Fig. 5 shows on the left the way in which a palm is positioned in front of the authentication device 100, and on the right side, also the way in which the guide light 611 of the optical guide unit from the Palm is sprinkled. This figure shows the guide light 613 scattered from the palm of the hand. The reflected light is seen around the fingers by the user. The user has to change the position of the hand until a green light is reflected on the palm of the hand. The green light indicates that the distance between the palm of the hand and the authentication device is optimal. As soon as the optimal position has been determined, the lighting unit 110 is activated for the authentication. The calibration based on the mask pattern enables the authentication device to function without the need for assistance in placing the hand in the optimal position. The device therefore works completely contactless.

A user must log into the authentication system before use. To register, the user determines the optimal distance from the device at which he holds his palm so that the green light appears. As soon as the green light appears, the user who wants to log in moves their hand in different directions, for example left, right, forwards and backwards. The palm is moved in the plane transverse to the different directions in order to grasp different parts of the palm. So next time every part of the palm can be used for authentication. During this process, a threshold can be set for the minimum surface area for authentication. The threshold value thus indicates the minimum area of the palm that is required for a particular authentication to be accepted. Figure 6 illustrates the various positions into which the palm is moved in accordance with an embodiment of the invention.

Once the various parts of the palm are scanned, a single image of the palm is built up. The image includes pattern information and depth information due to the unevenness of the palm (explained in more detail below), which are captured using a microlens array. Further, the image includes palm vein information captured under NIR light and palm print information captured under visible light. For example, the palm of Figure 6 shows only handprints. The palm of Figure 7 shows handprints and palm veins.

7 illustrates the determination of a region of interest (ROI) for authentication in accordance with an embodiment of the invention. The palm vein endpoints are determined for the entire palm. The identified palm vein ends are traced to the extreme points to form the area of interest. In Fig. 7 (middle) the outermost vein ends, for example 201a, 201b, 201c, 201d, 201e, 201f and 201g, are marked. The palm of the hand has many more vein ends, but only the outermost vein ends are considered to determine the area of interest. Once these ends are identified and registered with their coordinates, the extreme endpoints of veins 201a, 201b, 201c, 201d, 201e, 201f and 201g are traced to form the area of interest 202. Thus, the area of interest for obtaining the biometric identification data can be determined by the boundaries of the palm veins. The mask pattern can also help define the area of interest by clearly recognizing the hand against the rest of the background.

The size of the mask pattern on the palm varies depending on the deformity of the palm. The palm of the hand does not have a flat surface, so that the size of the mask pattern in different areas of the palm of the hand differs slightly depending on its distance from the lighting unit. The size of the mask pattern in the captured image can therefore be used as depth information. For example, the mask pattern size in the center of the palm is slightly larger than in the area around the center of the palm, which means that the center is slightly further away from the lighting unit than the areas surrounding it. Thus, the unevenness of the palm results in a distortion of the mask pattern (see Fig. 11, right). This unevenness of the palm and the information obtained thereby can hardly be reproduced with an artificial palm and can therefore be used as an indication of the liveliness of a person to be authenticated.

Figure 8 illustrates a method in which the biometric characteristics of the palm of the hand are extracted in accordance with an embodiment of the invention. After defining the area of interest, the presence of the handprints and palm veins that are detected under visible light or NIR light is determined in a sample block. The pattern block can refer to a polygon or a previously defined number of polygons of the pattern mask. One way of defining pattern blocks is described below.

The presence of the palm veins and handprints is determined as in pattern block 301 where a palm vein and a handprint overlap, i.e., cross. The system identifies such overlaps and stores the details, such as the overlap of the palm vein with the palm vein with its polar coordinates along with the size / area of the polygon or pattern block. The size of the polygon or pattern block, polar coordinates and / or depth information acquired using a microlens array, palm veins and handprint features are stored as a weighted mathematical formula named a neuron node. This neuron node recognizes the direction in which to jump for the next list of connected neurons for the pattern comparison. After capturing the palm image with all of this neuron information, the entire list of neuron nodes is saved as a mathematical formula.

Figure 9 illustrates various sample blocks in the previously defined area of interest in accordance with an embodiment of the invention. For further analysis, the system only considers sample blocks in the area of interest that have vein and imprint information. If only a handprint or only palm veins are available for a feature point, this block is ignored.

Figure 10 illustrates a neural network of multiple neuron nodes in accordance with an embodiment of the invention. Once the features have been extracted, they are presented as a network of neuron nodes such as neuron 1 (401), neuron 2 (402) , Neuron 3 (403), Neuron 4 (404) and Neuron 5 (405). Consider neuron 1 (401), which has a vein print and a hand print 410a, 410b, 410c. The handprint information of neuron 1 (401) is connected to neuron 2 (402) by handprint line 410a, and neuron 2 (402) is further connected by handprint line 410c to neuron 3 (403), handprint line 410a to neuron 4 (404), and handprint line 410b is connected to neuron 5 (405). Neuron 3 (403) is connected to neuron 2 (402) only, with neuron 4 (404) connected to neuron 2 (402). Neuron 5 (405) is connected to neuron 2 (402) and not neuron 1 (401) because the angle of the handprint line is pointing towards neuron 2 (402) this information is read as 124 which means that neuron 1 (401 ) can jump from neuron 2 (402) to neuron 4 (404), as well as 123, which means that neuron 1 (401) can jump from neuron 2 (402) to neuron 3 (403). This is repeated for the entire list of feature neurons in the area of interest of a palm and represented as a mathematical model. The same process is done and saved for the palm vein lines.

The extracted features can be stored as weighted neural nodes in a network. The extracted features also include the presence of the creases on the palm of the hand in relation to the pattern of the mask. In this scenario, the extracted features include the folds of the palm of the hand with respect to the patterns present in the mask. Each nodal point is defined as the presence of a pattern and how the palm folds and veins overlap in that pattern, the coordinates of the features, the depth information acquired using a microlens array, and the size of that predetermined geometric pattern. The network is expanded by gradually jumping along the vein or crease wherever a pattern has both a palm vein and a palm crease feature. This is repeated until all features are extracted and is represented as a mathematical weighted neural network.

In order to compare two palm images in the authentication process, the system executes a multi-stage neuromorphic algorithm in order to determine the degree of agreement with respect to the mathematical model stored in the database. The system thus uses a multi-phase object recognition. This multi-phase object recognition system uses extensive arrays of information-rich neurons to build a biologically plausible model of visual information processing. 12 illustrates one possible way of defining pattern blocks 704 by using the pattern mask of the lighting unit. The captured image of the palm includes palm veins 702, handprints 703 and the mask pattern 701 projected onto the palm. The mask pattern 701 shown consists of regular polygons a to s. In order to define a pattern block 704 for the further analysis, polygons are determined which contain palm veins 702 and handprints 703. In this case the polygon would be g, which has a crossing of a palm vein 702 and the palm print 703 (FIG. 12, left). The pattern block 704 is then defined by polygon g and its surrounding polygons b, c, f, h, I and m (Fig. 12, right).

The foregoing description of the preferred embodiment of the present invention has been presented for purposes of illustration and description. It is not to be regarded as complete and does not limit the invention to the specifically disclosed form. Numerous changes and variations are possible in light of the above teachings. This detailed description is not intended to limit the scope of the present invention in any way.

It is understood that the present invention is not limited to the embodiments discussed above. A person skilled in the art will be able, with knowledge of the invention, to derive further variants which also belong to the subject matter of the present invention.

Claims (14)

A biometric authentication system for acquiring biometric identification data from a body part of a person and for comparing the acquired biometric identification data with registered biometric identification data already acquired from the body part for authenticating the person, comprising: a lighting unit (110) for illuminating the body part; an image capturing unit (114) for capturing an image of the illuminated body part; a processing unit (118) for extracting the biometric identification data from the image captured by the image capture unit (114); and an authentication unit (120) for comparing the extracted biometric data with registered identification data for authenticating the person; characterized in that the lighting unit (110) is provided with at least one pattern mask (112) in order to project a mask pattern (701) onto the body part so that the projected mask pattern is contained in the captured image. 2. Biometric authentication system according to claim 1, wherein the pattern mask has a regular pattern of polygons, preferably triangles, squares, squares, pentagons or hexagons. 3. Biometric authentication system according to one of the preceding claims, wherein the lighting unit comprises at least one light source of visible light and / or at least one light source of near-infrared light, and wherein the light source preferably comprises a plurality of light-emitting diodes. 4. Biometric authentication system according to claim 3, wherein the light sources are arranged in a circle around the detection unit. 5. Biometric authentication system according to claim 3 or 4, wherein different light sources are arranged alternately. 6. Biometric authentication system according to one of the preceding claims, wherein the biometric authentication system further comprises an optical guide unit which is able to determine the distance between the body part and the detection unit on the basis of the mask pattern projected onto the body part and to assess whether this is too small is large or optimal. 7. Biometric authentication system according to one of the preceding claims, wherein the detection unit comprises a microlens arrangement for obtaining depth information of the body part. 8. Biometric authentication method for capturing biometric identification data from a body part of a person and for comparing the captured biometric identification data with registered biometric identification data already captured from the body part for authenticating the person, comprising the steps of: illuminating the body part using a lighting unit (110) which is provided with at least one pattern mask (112) for projecting a mask pattern onto the body part; Capturing an image of the illuminated body part comprising the projected mask pattern using an image capturing unit (114); Extracting the biometric identification data from the image captured by the capture unit (114) using a processing unit (118); Comparing the extracted biometric data with registered identification data for authenticating the person using an authentication unit (120). 9. Biometric authentication method according to claim 8, wherein the pattern mask has a regular pattern of polygons, preferably triangles, squares, squares, pentagons or hexagons and is used to determine the distance between the body part and the detection unit and to assess whether it is too small , too big or optimal. 10. Biometric authentication method according to one of claims 8 to 9, wherein the mask pattern is used to guide the body part in order to determine a distance between body part and detection unit and to assess whether this is too small, too large or optimal. 11. Biometric authentication method according to claim 8 to 10, wherein the lighting unit illuminates the body part with visible light and NIR light in order to obtain combined images, which structures of the body part which can be detected under visible light or the NIR light, and the mask pattern exhibit. 12. The biometric authentication method according to claim 11, wherein the mask pattern is used to extract the biometric identification data. 13. Biometric authentication device for capturing images for performing the method according to one of claims 8 to 12, the device comprising a lighting unit (110) for illuminating the body part; an image capturing unit (114) for capturing an image of the illuminated body part; characterized in that the lighting unit (110) is provided with at least one pattern mask (112) in order to project a mask pattern (701) onto the body part so that the captured image includes the projected mask pattern. 14. The biometric authentication device according to claim 13, the device further comprising means for connecting the device to an external processing unit (118) for extracting the biometric identification data from the image captured by the image capturing unit (114) and means for connecting the device to an authentication unit (120) for comparing the extracted biometric data with registered identification data for authenticating the person.