1460886298-0a8f4461-649e-4570-8308-d75afb425d6a

1. A computer-implemented method for discriminating between a real face and a two-dimensional reproduction of the face, in a biometric detection process, the method comprising:
(a) receiving at least two digital recordings of the face or its reproduction in time sequence, one after the other, the recordings taken by a camera;
(b) dividing each of the recordings into a number of image components wherein each image component comprises a number of pixels;
(c) determining the displacement of the individual image components from the first recording to the second recording by correlation and generating a displacement vector field therefrom;
(d) analyzing the displacement vector field for determining whether the recording has been made from a real face or from its reproduction; and
(e) generating a signal to be output on an output device, the signal indicating whether the recording has been made from a real face or from its reproduction.
2. The method of claim 1, further comprising the step of: for each of the image components, generating a displacement vector on the basis of the position of the same image component in a first recording and in a second recording.
3. The method of claims 1, further comprising the step of: between steps (a) and (b), localizing the face in each recording and, for subsequent processing, limiting the recording to a defined face clipping.
4. The method according to claim 1, further comprising, in step (b), dividing each recording into a regular grid of N\xd7M fields, forming image components, wherein the number of fields may be in the range between 64 and 576.
5. The method according to claim 4, wherein N=M, in particular N=M=8 or N=M=16.
6. The method of claim 4 wherein N=M=8 or 16.
7. The method according to claim 1, wherein, in step (c), correlation is performed on the basis of the image intensities of gray values of the image components.
8. The method according to claim 1, wherein, in step (c), correlation is realized by a sequential similarity detection, SSD, algorithm.
9. The method according to claim 8, further comprising the steps of:
calculating the difference of image intensities between an image component in the first recording and the image component in the second recording for a number of positions of the image component in the second recording;
for each position in a row of positions, comparing the difference of image intensities at the current position to the difference of image intensities at the previous position;
if the difference of image intensities at the current position is smaller than the difference of image intensities at the previous position, taking the current difference as minimum difference; and
if the difference of image intensities at the current position is larger than the difference of image intensities at the previous position, taking the previous position as the final minimum difference and terminating the comparison;
wherein the position of the image component in the second recording corresponds to the position at which the final minimum difference is determined
10. The method according to claim 1, wherein, in step (c), the correlation is limited to portions of the recording which are determined based on the knowledge of the general image structure or the evaluation of previous recordings.
11. The method according to claim 10, wherein the portion is not larger than the area which is defined by displacing the image component by one image component edge length, in any one direction.
12. The method according to claim 1, wherein, before executing the correlation in step (c), the image components are examined having regard to whether the pixels within one respective image component have a variance which is larger than a first threshold, wherein only such image components are processed, in step (c), the variance of which exceeds the threshold.
13. The method according to claim 1, wherein, after performing the correlation in step (c), the image components and their displaced components are compared and, if the difference in intensities of an image component and its displaced image components is larger than a second threshold, the displacement vector of the respective image component is discarded.
14. The method according to claim 1, wherein, for analyzing the vector field, at least one plane or another analytically-described surface is calculated, which approximates the x and y components of the vector field, and wherein the distance of the vector component of the vector field to the calculated plane is determined
15. The method according to claim 14, wherein the plane is calculated based on an approximation process, in particular according to the method of least square fit.
16. The method according to claim 14, wherein the mean distance of the vector components to the calculated plane is determined.
17. The method according to claim 14, wherein the distance of the vector components to the calculated plane is compared to a third threshold and wherein it is determined that the recording is one of a real face if the distance exceeds the third threshold and that the recording is one of a reproduction, if it does not exceed the third threshold.
18. The method according to claim 1, wherein the image components are determined by a predetermined grid and do not necessarily include characteristic features of the face.
19. The method according to claim 1, wherein not more than twenty, or not more than ten, or not more than five digital recordings of the face or its reproduction are being made.
20. The method according to claim 1, wherein exactly two digital recordings of the face or its reproduction are being made.
21. The method according to claim 20, wherein, in step (a), at least two different face positions are defined, when the recordings are being made.
22. A computer-implemented method for identifying a person in a distributed IT infrastructure, such as a cloud computing environment, wherein at least two digital recordings of the face of the person are received at a remote computing station and wherein the method is performed at one computer or in a distributed computer network remote from the person, wherein the recordings of the face of the person are also used for identifying the person, the method comprising the steps of:
(a) receiving the at least two digital recordings of the face or its reproduction in time sequence, one after the other, the recordings taken by a camera;
(b) dividing each of the recordings into a number of image components wherein each image component comprises a number of pixels;
(c) determining the displacement of the individual image components from the first recording to the second recording by correlation and generating a displacement vector field therefrom;
(d) analyzing the displacement vector field for determining whether the recording has been made from a real face or from its reproduction; and
(e) generating a signal to be output on an output device, the signal indicating whether the recording has been made from a real face or from its reproduction.
23. The method of claim 22 wherein the output device is located at proximate to the person.

The claims below are in addition to those above.
All refrences to claim(s) which appear below refer to the numbering after this setence.

1. A method for determining an object property of an object with the aid of images that have been recorded with the aid of a camera, the method comprising the steps of:
defining at least one object in a first image recorded with the aid of the camera,
determining the brightness of the defined object,
recording a further image at least each time after a preset time interval with the aid of the camera,
defining the object in each of the further images,
determining each time, the brightness of the defined object of each image,
assigning an object property to the object dependent on a curve defined by the brightness, and
wherein an actual brightness of the object is changed with a first frequency which differs from the sampling rate defined by the time interval between two images.
2. The method according to claim 1, wherein with the aid of the brightness curve at least one of a characteristic brightness variation or characteristic brightness variations of the light emitted by the object is determined.
3. The method according to claim 2, further comprising the step of determining whether at least one of the curve of brightness variation or, of brightness variations corresponds to a stored brightness variation curve.
4. The method according to claim 1, further comprising the steps of generating a sampling rate for determining the brightness of the object, and within a time interval between two images additional images are recorded with the aid of the camera, which are at least not used as further images for determining the curve of brightness, the image recording frequency being a multiple of the sampling rate.
5. The method according to claim 1 , wherein the first frequency is twice a mains frequency of an alternating voltage of a power supply network.
6. The method according to claim 1, wherein the object is a light source, powered with an alternating voltage, in particular with an alternating voltage from a public power supply network, which voltage preferably has a frequency of 50 HZ, 60 Hz, 16\u2154 Hz or 16.7 Hz.
7. The method according to claim 1, wherein in accordance with the time interval, a sampling rate for detecting the brightness of the object is generated which is lower than the first frequency, the brightness curve determined with the aid of the sampling rate having an oscillating curve with periodic amplitude variations that have a third frequency, the third frequency being lower than the sampling rate.
8. The method according to claim 7, wherein the oscillating curve with the periodic amplitude variations is the resulting oscillation from the sampling rate and the first frequency, in particular the beat of the sampling rate and the first frequency.
9. The method according to claim 1 wherein the brightness of the object is determined with the aid of a gray scale value of the object image.
10. The method according to claim 1, wherein the camera is moved at a speed in space during the recording of the images.
11. The method according to claim 1, wherein dependent on the determined brightness curve it is determined that the object is a moving object or a stationary object.
12. The method according to claim 1, wherein at least one of the object and the brightness is determined with the aid of an image analysis method.
13. A device for determining an object property of an object with the aid of images that have been recorded with the aid of a camera, comprising:
a camera which at least at a preset time interval records images of at least one object,
means for defining at least one object in a first image recorded with the aid of the camera,
means for determining a brightness of the defined object,
means for defining the object in each of the further images recorded with the aid of the camera at the preset time interval,
means for each time, determining the brightness of the defined object of each image,
means for assigning an object property to the object dependent on the brightness curve, and
means for determining an actual brightness of the object with a first frequency which differs from the sampling rate defined by the time interval between two images.
14. The device according to claim 13, wherein all of said means include a common control unit which preferably includes a DSP.
15. The device according to claim 13, wherein the camera is a camera of a stereo camera system.