Time-of-flight imager
View Patent ↗An improved solution for generating depth maps using time-of-flight measurements is described, more specifically a time-of-flight imager and a time-of-flight imaging method with an improved accuracy. A depth correction profile is applied to the measured depth maps, which takes into account propagation delays within an array of pixels of a sensor of the time-of-flight imager.
1. A time-of-flight imager configured to measure a depth map of an object comprising:
a light source for illuminating the object, and
a time-of-flight sensor with an array of pixels for detecting light reflected by the object to obtain a measured depth map comprising an array of depth values corresponding to different spatial coordinates of the object,
wherein the time-of-flight imager is configured to apply a depth correction profile to the depth map measured from the reflected light detected with the sensor with the array of pixels, to compensate for signal propagation delays in the time-of-flight sensor with the array of pixels, said depth correction profile varying spatially in dependence on coordinates within the measured depth map to compensate for signal propagation delays in the sensor.
2. The time-of-flight imager according to claim 1 , wherein the depth correction profile is a superposition of two n th order polynomials.
3. The time-of-flight imager according to claim 2 , wherein a first of the two n th order polynomials has a line number of the sensor as an input value and a second of the two n th order polynomials has a pixel number of a line of the sensor as an input value.
4. The time-of-flight imager according to claim 1 , wherein the depth correction profile is calculated on the fly or retrieved from a look-up table.
5. A method for measuring a depth map of an object with a time-of-flight imager, the method comprising:
illuminating the object with light emitted by a light source;
detecting light reflected by the object with a time-of-flight sensor having an array of pixels to obtain a measured depth map comprising an array of depth values corresponding to different spatial coordinates of the object; and
applying a depth correction profile to the depth map measured from the reflected light detected with the sensor with the array of pixels, to compensate for signal propagation delays in the time-of-flight sensor with the array of pixels, said depth correction profile varying spatially in dependence on coordinates within the measured depth map to compensate for signal propagation delays in the sensor.
6. The method according to claim 5 , wherein the depth correction profile is a superposition of two n th order polynomials.
7. The method according to claim 6 , wherein a first of the two n th order polynomials has a line number of the sensor as an input value and a second of the two n th order polynomials has a pixel number of a line of the sensor as an input value.
8. The method according to claim 5 , wherein the depth correction profile is calculated on the fly or retrieved from a look-up table.
9. A method for determining a depth correction profile for a time-of-flight imager, the method comprising:
illuminating a flat surface, which is parallel to an array of pixels of a time-of-flight sensor of the time-of-flight imager and located at a known distance from the time-of-flight imager, with light emitted by a light source;
detecting light reflected by the flat surface with the time-of-flight sensor to obtain a measured depth map comprising an array of depth values corresponding to different spatial coordinates of the flat surface; and
determining the depth correction profile from the depth map measured from the light reflected by the flat surface and detected with the sensor with the array of pixels, to compensate for signal propagation delays in the time-of-flight sensor with the array of pixels, said depth correction profile varying spatially in dependence on coordinates within a measured depth m.
10. The method according to claim 9 , wherein the depth correction profile is a superposition of two n th order polynomials.
11. The method according to claim 10 , further comprising:
performing line averaging on lines of the sensor;
fitting an n th order polynomial to the averaged line values to determine a first of two polynomials;
subtracting the averaged line values from the measured depth map;
performing column averaging on columns of the sensor; and
fitting an n th order polynomial to the averaged column values to determine a second of the two polynomials.
12. The method according to claim 9 , further comprising the step of storing the depth correction profile in a look-up table.