Time-of-flight camera

Abstract

Time-of-flight camera (20) for a time-of-flight camera system (1), with a light transit time sensor (22) with several light transit time pixels (23) for determining a phase shift of an emitted and received light (Sp2), where distance values ​​(d) are determined based on the detected phase shifts (Δφ), wherein the time-of-flight camera (20) has a memory in which at least parameters of a point spread function (PSF) are stored, wherein the point spread function (PSF) takes into account a scattered light behavior and a signal crosstalk of the time-of-flight camera (20) and the time-of-flight sensor (22), with an evaluation unit which is designed in such a way that a captured image (I(x)) is reduced in resolution, and that a correction (ΔI(x)) with this reduced resolution is determined based on the stored point spread function (PSF), where the correction (ΔI(x)) is then upscaled to the original resolution of the acquired image (I(x)), and the acquired image (I(x)) is corrected with the upscaled correction (ΔI(x)), whereby the determination of the phase shifts (Δφ) or distance values ​​(d) is carried out on the basis of a corrected image (I0(x)).

Description

[0001] The invention relates to a time-of-flight camera and a method for detecting a point spread function for correcting the detected signals of a time-of-flight sensor.

[0002] Time-of-flight cameras and time-of-flight camera systems refer in particular to all time-of-flight or 3D-TOF camera systems that obtain time-of-flight information from the phase shift of emitted and received radiation. PMD cameras with photonic mixer detectors (PMDs) are particularly suitable as time-of-flight or 3D-TOF cameras, such as those described in DE 197 04 496 C2 and available from ifm electronic GmbH or pmdtechnologies ag as the O3D frame grabber or CamCube. The PMD camera allows for a flexible arrangement of the light source and detector, which can be arranged either in one housing or separately.

[0003] Furthermore, DE 11 2008 003 342 T5 discloses a method for recording 3D images of a scene based on the time-of-flight principle. The method comprises illuminating a scene with intensity-modulated light, imaging the scene on a pixel array using an optical system, and detecting the reflected light in each pixel. For each pixel, a distance value is determined based on the phase of the detected light. To determine the distance values, a phase-sensitive deconvolution of the scene imaged on the pixel array is provided to compensate for phase errors caused by light spreading in the optical system. This enables the detection of more accurate 3D images and precise distance determination under difficult lighting conditions.Phase-sensitive deconvolution involves forming and deconvolving data fields corresponding to the amplitude and phase values ​​of the pixels and allows the correction of phase and amplitude errors caused by homogeneous light scattering and can be adapted to the contamination level of the optical system.

[0004] In the study NAVAS-MOYA, FA, et al.: Measurement of the optical transfer function using a white-dot pattern presented on a liquid-crystal display. Journal of the European Optical Society: Rapid Publications. 23 April 2013. Vol. 8. DOI 10.2971 / jeos.2013.13029, a method for measuring the optical transfer function (OTF) using a white-dot pattern presented on a liquid-crystal display (LCD) is presented. The OTF and its modulation transfer function (MTF) are established measures of the quality of optical systems. Traditional methods for MTF determination, such as the sine wave, bar target, edge gradient, series expansion, and random pattern methods, have limitations, particularly with regard to providing spatial information. The proposed method uses a white-dot pattern on an LCD, with white pixels acting as point sources and point spread functions (PSFs) being used to calculate the system's OTF.The method allows the acquisition of spatial information in the image field and covers all frequencies and directions in a single image without the need for an additional light source. For validation, an optical system with a monochrome camera and a liquid crystal tunable filter (LCTF) was characterized at different wavelengths. The method achieves an accuracy with errors below 3% and offers advantages over conventional OTF methods, such as the avoidance of derivative-dependent noise and the provision of frequency information in all directions. The experimental setup includes an LCD with a 1680 x 1050 pixel pattern, a CCD camera, and an LCTF. The results demonstrate that the proposed method not only provides the average OTF of the system but also enables detailed spatial distributions and distortion analyses.WO 2014 / 145722 A2 concerns collaborative photography, in which image data from multiple cameras is used to generate a 3D model of a scene. This image data is sent to a cloud server, which processes the data and creates a model that allows for the synthesis of arbitrary views of the scene. The technology uses various camera systems, including time-of-flight cameras, stereoscopic cameras, and plenoptic cameras, to capture depth information and create an accurate model.

[0005] The object of the invention is to further improve the compensation of phase errors.

[0006] The object is advantageously achieved by the time-of-flight camera system according to the invention according to the preamble of the independent claim.

[0007] A time-of-flight camera is particularly advantageous for a time-of-flight camera system, with a time-of-flight sensor with several time-of-flight pixels for determining a phase shift of a transmitted and received light, where distance values ​​are determined based on the recorded phase shifts, wherein the time-of-flight camera has a memory in which at least parameters of a point spread function are stored, where the point spread function takes into account stray light behavior and signal crosstalk of the time-of-flight camera and the time-of-flight sensor, with an evaluation unit which is designed in such a way that the captured image (I(x)) is reduced in resolution, and that a correction (ΔI(x)) with this reduced resolution is determined based on the stored point spread function (PSF), where the correction (ΔI(x)) is then upscaled to the original resolution of the acquired image (I(x)), and the acquired image (I(x)) is corrected with the upscaled correction (ΔI(x)), The phase shifts (Δφ) or distance values ​​(d) are determined based on the corrected image (I0(x)). This approach has the advantage that distance values ​​can be corrected during operation using a pre-stored point spread function. Furthermore, the computational effort for the corrections can be significantly reduced.

[0008] Preferably, the point spread function is complex-valued.

[0009] Furthermore, it is planned to reduce the resolution by averaging the amplitudes of neighboring pixels and to upscale by duplicating the amplitudes.

[0010] Preferably, the point spread function is stored in memory as a matrix or lookup table and / or as a Fourier transform.

[0011] It is particularly useful if the time-of-flight camera is designed in such a way that the point spread function stored in the memory is determined as follows by arranging a point light source and a time-of-flight camera in such a way that that a time-of-flight sensor of the time-of-flight camera detects the point light source, wherein a distance between the point light source and that of the time-of-flight camera and / or a beam profile of the point light source are selected such that on the time-of-flight sensor, less than 5 time-of-flight pixels in a pixel row or column or a maximum of 16x16 pixels are illuminated, wherein the point spread function is determined based on at least a subset of the time-of-flight pixels of the time-of-flight sensor.

[0012] This approach has the advantage that the light source can be easily constructed within certain limits to determine a point spread function.

[0013] In a further embodiment, it is provided to operate the point light source unmodulated.

[0014] The modulation gates of the time-of-flight sensor's time-of-flight pixels are controlled in such a way that charge carriers in the time-of-flight pixel are primarily accumulated at one integration node. This approach ensures that the generated photoelectrons are preferentially collected at one integration node.

[0015] In a further embodiment, it is provided to control the point light source and the time-of-flight sensor in phase synchronization with a modulation signal, and to determine sensor difference signals for at least three different phase positions.

[0016] It is particularly useful to perform at least two recording frames with different integration times of the time-of-flight sensor and / or different light intensities of the point light source to determine the point spread function.

[0017] The invention is explained in more detail below using exemplary embodiments with reference to the drawings.

[0018] They show: Fig. 1 schematically shows a time-of-flight camera system, Fig. 2 a modulated integration of generated charge carriers, Fig. 3 a setup for determining a point spread function, Fig. 4 a cross-section of images to determine a point spread function Fig. 5 a capture of a reference scene, Fig. 6 a detection of an object in front of the reference scene, Fig. 7 measured distance values ​​after Fig. 6 in relation to the actual distances, Fig. 8 a detection of two reference surfaces with different distances, Fig. 9 measured distance values ​​after Fig. 8 in relation to the actual distances, Fig. 10 shows a possible schematic sequence of stray light correction according to the invention.

[0019] In the following description of the preferred embodiments, like reference numerals designate like or comparable components.

[0020] Fig. 1 shows a measuring situation for an optical distance measurement with a time-of-flight camera, as known, for example, from DE 197 04 496 A1.

[0021] The time-of-flight camera system 1 comprises a transmitting unit or an illumination module 10 with an illumination 12 and an associated beam-forming optics 15 as well as a receiving unit or time-of-flight camera 20 with a receiving optics 25 and a time-of-flight sensor 22.

[0022] The time-of-flight sensor 22 has at least one time-of-flight pixel, preferably also a pixel array, and is designed in particular as a PMD sensor. The receiving optics 25 typically consist of several optical elements to improve the imaging properties. The beam-shaping optics 15 of the transmitting unit 10 can be designed, for example, as a reflector or lens optic. In a very simple embodiment, optical elements can be omitted on both the receiving and transmitting sides.

[0023] The measuring principle of this arrangement is essentially based on the fact that, based on the phase shift of the emitted and received light, the propagation time and thus the distance traveled by the received light can be determined. For this purpose, the light source 12 and the light propagation time sensor 22 are jointly subjected to a specific modulation signal M0 with a base phase position φ0 via a modulator 30. In the example shown, a phase shifter 35 is also provided between the modulator 30 and the light source 12, with which the base phase φ0 of the modulation signal M0 of the light source 12 can be shifted by defined phase positions φ var For typical phase measurements, phase positions of φ var = 0°, 90°, 180°, 270°.

[0024] According to the set modulation signal, the light source 12 sends an intensity-modulated signal S p1with the first phase position p1 or p1 = φ0 + φ var This signal S p1 or the electromagnetic radiation is reflected in the case shown by an object 40 and hits due to the distance traveled 2d, or the light travel time t L , phase-shifted Δφ(t L ) with a second phase position p2 = φ0 + φ var + Δφ(t L ) as received signal S p2 to the time-of-flight sensor 22. In the time-of-flight sensor 22, the modulation signal M0 is combined with the received signal S p2 mixed, whereby the phase shift or the object distance d is determined from the resulting signal.

[0025] Infrared light-emitting diodes or surface emitters (VCSELs) are preferably suitable as the illumination source or light source 12. Of course, other radiation sources in other frequency ranges are also conceivable, in particular, light sources in the visible frequency range.

[0026] The basic principle of phase measurement is shown schematically in Fig. 2. The upper curve shows the time course of the modulation signal M0 with which the illumination 12 and the time-of-flight sensor 22 are controlled. The light reflected from the object 40 arrives as the received signal S p2 according to its light travel time t L phase shifted Δφ(t L) to the time-of-flight sensor 22. The time-of-flight sensor 22 collects the photonically generated charges q over several modulation periods in the phase position of the modulation signal M0 in a first integration node Ga and in a phase position shifted by 180° in a second integration node Gb. For this directing of the charges to the integration nodes, the pixels 23 of the time-of-flight sensor 22 have at least two modulation gates Gam, Gbm, which direct the charges to the first or second integration node Ga, Gb depending on the applied modulation signals. From the difference between the charges qa, qb collected in the first and second integration nodes Ga, Gb, taking into account all phase positions φ var the phase shift Δφ(t L ) and thus determine a distance d of the object.

[0027] Fig. Figure 3 schematically shows a setup for determining a point spread function PSF. Here, the light source 112 and the time-of-flight sensor 22 can be operated unmodulated or with at least one predetermined modulation frequency. When using unmodulated light, it is advantageous if the time-of-flight sensor 22 or the pixels 23 are also operated unmodulated. It is helpful if a constant voltage is applied to the modulation gates Gam, Gbm of the pixels 23 such that the photogenerated charges are primarily collected in only one integration node Ga, Gb.

[0028] To determine the PSF, it is advantageous if the light source 112 essentially illuminates only a single pixel 23, preferably less than 3x3 and in particular less than 5x5 pixels 23, of the time-of-flight sensor 22. To provide such a light spot, an aperture 150 with a sufficiently small aperture 152 is provided in front of the light source 112. The original light signal I0 emerging from the aperture 150 is influenced by a multitude of factors on its way to the sensor up to the captured image signal I(x), for example, by properties of the optical system or the optics 25 or reflections between the sensor 22 and the optics 25. Intrinsic properties of the sensor 22 itself also play a role, such as signal crosstalk or electron diffusion between the pixels 23.As a result, the image signal I(x) captured at the sensor can be viewed as a convolution between the incoming light I0 and a point spread function PSF, which essentially encompasses all properties of the entire system. Due to the singular illumination of one or a few pixels, the captured image signal I(x) essentially corresponds to the point spread function PSF. To determine the point spread function, all pixels are preferably evaluated. However, it is also conceivable to evaluate only a partial area around the singularly illuminated pixel.

[0029] The quality of the point spread function (PSF) can be improved if necessary by determining multiple point spread functions based on several singly illuminated pixels 23. For example, it is useful to also illuminate pixels 23 outside the optical axis in order to determine additional point spread functions at these positions. Based on the determined point spread functions, a point spread function can then be determined that will be used for subsequent corrections.

[0030] Since the aforementioned electron diffusion typically occurs at a diffusion speed that is significantly lower than the propagation of light, the electrons reach neighboring pixels with a time delay, so that the influence of electron diffusion is also noticeable as a phase shift. The point spread function PSF thus also contains complex-valued components. For more precise determination of these quantities, it is therefore advantageous to operate the light source 112 in different phase positions.

[0031] Since a point spread function typically has a high dynamic range over several orders of magnitude, it is also advantageous to operate the point light source 112 with different intensities and / or the sensor 22 with different integration times to detect the PSF.

[0032] To compensate for dark currents, it is helpful to record image signals I(x) both when the light source 112 is switched on and off.

[0033] From the sum of all measurements, a model of a point spread function can then be generated that is applicable to all pixels 23.

[0034] Such a model can be generated according to the following considerations: Since the measured PSF is noisy and may, for example, contain artifacts that are very specific to the pixel position on the sensor, a "clean" PSF can be obtained, for example, by fitting the measured PSF to a suitable model. Suitable models include, for example, PSF(x→)=A(x→)+B(‖x→‖pB) where for example A(x→)=A0 exp(−s(‖x→‖p)) can be chosen.

[0035] Here, x→ the distance vector from the central pixel x→0 the PSF in pixels and ‖x→‖p=(x1p+x2p)1 / p the p-norm of x→. For example, p = 2 would result in an exactly radially symmetric PSF. Since the PSF is not necessarily radially symmetric, but can be diamond-shaped, for example, p ≠ 2 can yield better results. By choosing the appropriate p-norm, anisotropies of the PSF can be taken into account.

[0036] Since most of the light falls on the central pixel of the PSF, it is helpful to add a locally narrow function B(r) to the model that reflects this component. This could be, for example, a Dirac delta or a Gaussian function that describes, for example, lens blur.

[0037] For efficiency reasons, it is advantageous to describe the PSF, for example, in the form of a spline curve. To describe phase shifts with this PSF, the spline can, for example, have a complex-valued component in addition to the real component. This also makes the PSF complex-valued. Suitable fitting parameters then include the values ​​at the spline nodes, the norm parameters p and p B , as well as parameters that specify the form of B(r). Instead of storing the entire PSF, it is advantageous to store only the necessary parameters so that the PSF can be generated from these parameters during software initialization.

[0038] During operation of the time-of-flight camera, it is then possible to adjust the distance values ​​for stray light influences using the stored parameters and the PSF generated from them.

[0039] Using the described setup, a first image is preferably I˜k(x→) with short exposure time t k Specifically, the exposure time must be chosen so that none of the pixels are saturated. If modulated light is used, no pixel in the resulting raw images may be saturated.

[0040] In addition, a second image I˜l(x→) with long exposure time t l Here, the exposure time should be selected so that the portion of the PSF caused by stray light and / or crosstalk is as completely visible as possible, i.e., not affected by noise. The exposure time here is typically 1,000–10,000 times longer than in the first image.

[0041] The images can be captured using either unmodulated light or the light source and sensor can be modulated in the usual way. In the latter case, the images I˜k(x→) and I˜l(x→) As usual, they are complex-valued, i.e. they contain phase information that reflects the time from the emission of the light to the reception of the generated electrons at the gates of the sensor.

[0042] For both images, it may be helpful to take a series of images instead of just one and then average them to further reduce noise.

[0043] To obtain consistent values ​​between the first and second image, for example, the brightnesses (or amplitudes) are normalized with the different integration times: Il(x→)=I˜k(x→) / tlIk(x→)=I˜k(x→) / tk

[0044] In the images obtained, the exact position of the illuminated central pixel is x→0 still unknown. To determine the position of the central pixel x→0 To determine the first image Ik(x→) For example, binarized using a thresholding method, which should result in the bright LED spot in a coherent area.

[0045] The center of the connected area is a good estimate for the central pixel or the central point x→0 on the sensor at which the light source is directed. This central point x→0 does not necessarily fall on the center of a pixel, ie the found position for the central point x→0 does not have to be an integer. Now the image Ik(x→) of the short exposure to the model of a sharp spot. Such a model is, for example, in equation (1) with B(‖x→‖p) Specifically, (PB,pB)=arg minPB,pB(∑x→|Ik(x→)−B(‖x→−x→0pB‖)|2) determined, where P B the parameters of the function B(r) and p Bare the parameters of the norm. For example, B(r) = B0 exp (-br 2 ), where P B = (B0, b) would.

[0046] There are numerous algorithms for the numerical minimization according to equation (4), such as the Nelder-Mead method.

[0047] In addition to P B and p B It may also provide better results to use the center in equation (4) x→0 of the light source into the optimization. Then the previously determined value from the binarized image would serve as a suitable starting value.

[0048] Now the second image Il(x→) with the long exposure time. Analogous to equation (4), the image is fitted to the model of the stray light signature, A(x→) in equation (1), fitted: (PA,pA)=arg minPA,pA(∑x→|Il(x→)−A(x→−x→0)|2)

[0049] If necessary, the central part of the PSF, which is described by B(r), can be ignored.

[0050] Analogous to the first fit, here P A the parameters of the model function A(r→). A useful function is, for example, the following: A(r→)=A0exp(−s(‖r→‖pA)+iΦ(r→)) where s(r) and represents a (real) spline curve. The function Φ(r→) Describes a phase delay of the incident light spot, which can be caused, for example, by phase crosstalk between pixels. Since this is not necessarily isotropic, it may be necessary to Φ(r→) as a two-dimensional function (e.g. a 2D spline or a 2D look-up table) instead of assuming a radially symmetric function as for s(r).

[0051] The fit parameters P A are in this case A0, p A, as well as the function values ​​of the splines at the nodes. If required, the position of the nodes can also be part of the fit parameters P A be.

[0052] With the obtained parameters P A and P B , as well as the PSF model, for example, according to equation (1), it is now possible to generate an artifact-free and noisy PSF. Instead of saving the entire PSF, it is advantageous to only save these or other suitable parameters from which the PSF can be generated during software initialization.

[0053] It is preferred to process the images taken with different exposure times separately: PSF(x→)=A1(x→)+A2(x→)+⋯

[0054] The partial models A1(x→),A2(x→),… can, for example, correspond to different dynamic ranges of the PSF and each separately on recordings I1(x→),I2(x→),… Based on these fit parameters, the PSF can then be summarized according to equation (7).

[0055] In the above, calibration was described using a point light source with an aperture as the light source or light source system. Of course, calibration is not limited to such a light source; any light source or light system capable of producing a suitable light point can be considered.

[0056] Fig. 4 to 9 show further methods for determining a suitable point spread function PSF. In the method according to Fig. 4, a first 3D image I1(x) of a reference scene and a second 3D image I2(x) with an object 40 in the foreground of the reference scene are captured. As already discussed, a change in the distance values ​​known from the first 3D image I1(x) is to be expected due to systemic influences. To determine a point spread function suitable for the correction, parameters of a first model PSF are then varied until differences between the first and second images, in particular distance errors, are minimal or smaller than a tolerated limit. In this case, preferably only those image areas or a sub-area thereof in which the reference scene is visible in both images are taken into account.

[0057] The images can be used as in Fig. 5 and Fig. 6. In a first step, a first 3D image I1(x) of a reference scene is acquired ( Fig. 5). As a reference scene, a wall or a floor can be easily captured, for example, but in principle, any scene with any height profile can also be captured. In the second step, according to Fig. 6, an object is placed above the reference scene, for example, a hand or another object, and a second distance image I2(x) is acquired. Here, too, the properties of the object are essentially uncritical. Fig. As described in section 4, a correction PSF can then be generated based on the difference between the two images.

[0058] In Fig. Figure 7 shows a variant in which the reference scene and the object are arranged flat and plane-parallel to each other. Using such prior knowledge, the optimization of the PSF can be simplified if necessary.

[0059] Alternatively, it is possible, for example, to acquire only one image of a target at a sufficient distance from a flat reference scene or plane (e.g., wall, table, floor) instead of two images. To determine the PSF, the parameters are varied until the reference surface behind the target is as flat as possible, or the deviations of the corrected reference surface from a plane are smaller than a tolerated limit.

[0060] It is particularly advantageous if the dimensions and distances of the reference scene and / or the introduced target are known in advance.

[0061] Fig. 8 and Fig. 9 show another variant of the above-mentioned procedure. Fig. The object 40 shown in Figure 8 has a step defined in height. The height Δd = d T2 - d T1is preferably known in advance. As in the above example, the parameters of a PSF model are varied until the distance error is minimal or below a tolerated limit.

[0062] The raw images measured by the sensor D j (e.g. j = 0,1,2,3 corresponding to the phase positions 0°, 90°, 180°, 270°) are, mathematically speaking, a convolution of the unknown, non-scattered raw images Dj0 with the PSF: Dj(x)=∑ΔxDj0(x−Δx)⋅PSF(Δx) Interesting for further processing are the complex-valued images I(x):=(D0(x)−D2(x))+i(D1(x)−D3(x))

[0063] Since the convolution is a linear operation, the following applies analogously to I (x) and the non-scattered complex-valued image I0(x): I(x)=∑ΔxI0(x−Δx)⋅PSF(Δx) or I(x)=I0(x)∗PSF(x)

[0064] The deconvolution is performed in Fourier space. For this purpose, I(x) and the PSF are Fourier transformed (F[ ]): Î(k) = F[I(x)] and PSF^(k)=F[PSF(x)].

[0065] Equation 4 thus becomes: I^(k)=I^0(k)⋅PSF(k) and therefore I^0(k)=I^(k)PSF(k)^

[0066] This gives the image without stray light distortion I0(x)=F−1[I^0(k)]

[0067] If one is interested in the correction ΔI(x) := I0(x) - I(x), i.e. the difference between the acquired and corrected image, Equation (13) can be rearranged as follows: ΔI^(k):=I^0(k)−I^(k)=I^(k)(1PSF^(k)−1) where ΔÎ(k) = F[ΔI(x)] is, analogous to the above treatment, the Fourier transform of the correction ΔI (x).

[0068] This correction can be downscaled or reduced in resolution before the Fourier transformation, for example, for performance reasons, and then scaled back up to the original resolution after stray light correction. The resulting correction can then be added to the acquired image I(x) to obtain the corrected image I0 (x).

[0069] Due to the data reduction, the computational effort is naturally also reduced. Reference symbol 1 time-of-flight camera system 10 lighting module 12 Lighting 15 Beam shaping optics 20 receivers, time-of-flight camera 22 Light transit time sensor 30 Modulator 35 phase shifters, lighting phase shifters 40 objects φ, Δφ(t L ) runtime-related phase shift φ var Phase position φ0 base phase M0 modulation signal p1 first phase p2 second phase S p1 Transmission signal with first phase S p2 Received signal with second phase t L Light travel time Ga, Gb integration nodes d object distance q charge

Claims

[1] Time-of-flight camera (20) for a time-of-flight camera system (1), with a light transit time sensor (22) with several light transit time pixels (23) for determining a phase shift of an emitted and received light (Sp2), where distance values ​​(d) are determined based on the detected phase shifts (Δφ), wherein the time-of-flight camera (20) has a memory in which at least parameters of a point spread function (PSF) are stored, wherein the point spread function (PSF) takes into account a scattered light behavior and a signal crosstalk of the time-of-flight camera (20) and the time-of-flight sensor (22), with an evaluation unit which is designed in such a way that a captured image (I(x)) is reduced in resolution, and that a correction (ΔI(x)) with this reduced resolution is determined based on the stored point spread function (PSF), where the correction (ΔI(x)) is then upscaled to the original resolution of the acquired image (I(x)), and the acquired image (I(x)) is corrected with the upscaled correction (ΔI(x)), whereby the phase shifts (Δφ) or distance values ​​(d) are determined using a corrected image (I0(x)). [2] Time-of-flight camera (20) according to claim 1, wherein the point spread function (PSF) is complex-valued. [3] Time-of-flight camera (20) according to one of the preceding claims, in which the reduction of the resolution is carried out by averaging amplitudes of adjacent pixels and the upscaling is carried out by duplicating the amplitudes. [4] Time-of-flight camera (20) according to one of the preceding claims, in which the point spread function (PSF) is stored in the memory as a matrix or lookup table. [5] Time-of-flight camera (20) according to one of the preceding claims, in which the point spread function (PSF) is stored in the memory as a Fourier transform. [6] Time-of-flight camera (20) according to one of the preceding claims, in which the point spread function (PSF) is stored on an external device and the correction of the phase shifts (Δφ) or distance values ​​(d) is carried out on the external device. [7] Time-of-flight camera (20) according to one of the preceding claims, in which the point spread function (PSF) stored in the memory was determined as follows: by arranging a point light source (112) and a time-of-flight camera (20) such that the time-of-flight sensor (22) of the time-of-flight camera (20) detects the point light source (112), wherein a distance between the point light source (112) and that of the time-of-flight camera (20) and / or a beam profile of the point light source (112) are selected such that on the time-of-flight sensor (22) fewer than 5 time-of-flight pixels (23) are illuminated in a pixel row or column or a maximum of 16x16 pixels, wherein the point spread function (PSF) is determined at least based on a subset of the time-of-flight pixels (23) of the time-of-flight sensor (22). [8] Time-of-flight camera (20) according to claim 7, wherein the point light source (112) is operated unmodulated to determine the point spread function (PSF). [9] Time-of-flight camera (20) according to claim 8, wherein, in order to determine the point spread function (PSF), the modulation gates (Gam, Gbm) of the time-of-flight pixels (23) of the time-of-flight sensor (22) are controlled in such a way that charge carriers in the time-of-flight pixel (23) are accumulated primarily only at one integration node (Ga, Gb). [10] Time-of-flight camera (20) according to claim 9, wherein, in order to determine the point spread function (PSF), the point light source (112) and the time-of-flight sensor (22) are controlled in phase synchronism with a modulation signal, and sensor difference signals are determined for at least three different phase positions. [11] Time-of-flight camera (20) according to one of claims 7 to 10, wherein at least two recording frames with different integration times of the time-of-flight sensor (22) and / or different light intensities of the point light source (112) are carried out to determine the point spread function (PSF).

Images