TIME-OF-FLIGHT SENSOR AND METHOD FOR ADJUSTING THE EXPOSURE TIME OF SUCH A SENSOR

Abstract

According to one aspect, an indirect time-of-flight measurement sensor is proposed, comprising: - a photosensitive pixel array (PXM) configured to acquire a succession of images of a scene during a given exposure time, - a control unit (CU) configured to control the acquisition of a succession of images by the pixel array (PXM) and to define an exposure time for this acquisition based on the saturation level of the pixels of said array (PXM), the distances (z) between the sensor and elements of the scene, and a standard deviation (σz) of the distances between the sensor and the elements of the scene. Figure for the abstract: Figure 1

Description

Title of the invention: TIME-OF-FLIGHT SENSOR AND METHOD FOR ADJUSTING AN EXPOSURE TIME OF SUCH A SENSOR

[0001] Embodiments and implementations relate to time-of-flight sensors, and more particularly to indirect time-of-flight measurement sensors (also known by the acronym “iTOF” for “indirect time of flight”).

[0002] A time-of-flight sensor is a sensor for measuring distances between a sensor and elements of a scene. The time-of-flight sensor comprises an array of photosensitive pixels. The time-of-flight sensor is configured to calculate a distance for each of its pixels.

[0003] There are direct time-of-flight measurement sensors and indirect time-of-flight measurement sensors.

[0004] A direct time-of-flight measurement sensor is configured to calculate a distance between this sensor and an element of a scene from a measurement of the time between the emission of the signal by the sensor and the reception of the signal reflected by said element.

[0005] An indirect time-of-flight measurement sensor, also referred to as an “iTOF sensor” hereinafter, is configured to calculate a distance between this sensor and an element of the scene from a measurement of a phase shift between a signal emitted by the sensor and a signal reflected by said element and detected by the sensor.

[0006] Each photosensitive pixel of the matrix delivers data relating to the signal detected by this pixel. The pixel matrix thus makes it possible to obtain a data matrix delivered by the pixel matrix. This data matrix then forms an image of the scene.

[0007] More particularly, the iTOF sensor is configured to acquire a plurality of successive images of the scene to determine the phase shift between the emitted signals and the reflected signals.

[0008] The iTOF sensor can be subjected to a wide variety of scene conditions (ambient light level, distance and reflectivity of objects). Therefore, the exposure time of the iTOF sensor should be adapted according to the scene conditions.

[0009] Conventionally, the exposure time of the iTOF sensor is adapted according to a saturation rate of the pixel matrix, i.e. a pixel filling rate.

[0010] In particular, the exposure time is adjusted until the saturation rate is greater than or equal to a given saturation rate threshold.

[0011] Such control of the exposure time has the disadvantage of significantly increasing the exposure time when the brightness of the scene is low.

[0012] However, the longer the iTOF sensor is exposed, the greater its energy consumption.

[0013] There is therefore a need to propose a solution making it possible to optimize the exposure time of the acquisitions of an iTOF sensor and to reduce the energy consumption of such a sensor.

[0014] According to one aspect, there is provided an indirect time-of-flight measurement sensor comprising: - a matrix of photosensitive pixels configured to acquire a succession of images of a scene during a given exposure time, - a control unit configured to control an acquisition of a succession of images by the pixel matrix and to define an exposure time for this acquisition according to a saturation rate of the pixels of said matrix, distances between the sensor and elements of the scene and a standard deviation of the distances between the sensor and the elements of the scene.

[0015] Such an indirect time-of-flight measurement sensor is thus configured to adapt the exposure time (also referred to as the “integration time”) for an acquisition of a succession of images not only as a function of a saturation rate of the sensor but also as a function of the distance between the elements of the scene and the sensor as well as the standard deviation of the distances between the elements of the scene and the sensor. In particular, the distance between the elements of the scene and the sensor as well as the standard deviation of the distances between the elements of the scene and the sensor make it possible to define an acquisition precision relative to the distance of the elements of the scene relative to the sensor. The sensor is therefore configured to adapt the exposure time as a function of a saturation rate of the sensor and an acquisition precision.Such an exposure time setting allows to optimize a compromise between sensor energy consumption and sensor signal-to-noise ratio. Indeed, in some cases, especially when the sensor is located in a dark room, if the exposure time is set only according to the pixel saturation rate, then the exposure time can be very long because only the light emitted by the sensor can increase the pixel saturation rate. This results in significant sensor energy consumption and sensor accuracy higher than desired. By taking into account the acquisition accuracy to define the exposure time, it is possible to reduce the exposure time to achieve a desired accuracy and saturation of the pixels. pixels. The sensor therefore avoids an unwanted increase in exposure time which would cause the sensor to be too precise in relation to the distance of the elements in the scene from the sensor.

[0016] In an advantageous embodiment, the control unit is configured to define the exposure time for each acquisition from the saturation rate of the pixels of said matrix and a precision of the sensor corresponding to a ratio between the distances between the sensor and elements of the scene and the standard deviation of the distances between the sensor and the elements of the scene.

[0017] In an advantageous embodiment, the control unit is configured to compare the saturation rate of the pixels to a saturation rate threshold, and to: - if the saturation rate of the pixels is greater than said saturation rate threshold, reduce the exposure time, - otherwise compare said precision to a precision threshold, and O if the precision is greater than said precision threshold, reduce the exposure time, O otherwise, increase the exposure time.

[0018] The saturation rate is thus the first statistic checked to define the exposure time. This makes it possible to avoid saturating the pixels of the matrix.

[0019] Preferably, the control unit is configured to adjust the exposure time of the sensor by increasing or reducing the value of the exposure time by a percentage between a minimum step and a maximum step. This minimum step and this maximum step are defined parameters known by the control unit.

[0020] Advantageously, the precision is evaluated from the mathematical formula 21, (7? where z is a distance between a scene element from the sensor and oz is the standard deviation of the distances between the sensor and the scene elements.

[0021] The mathematical formula 21 has the advantage of varying linearly with respect to the exposure time.

[0022] In an advantageous embodiment, if the saturation rate threshold is incompatible with the precision threshold, then the control unit is configured to update the precision threshold with the saturation rate. This makes it possible to avoid a phenomenon of flickering of the brightness of the acquired images.

[0023] According to another aspect, a method is proposed for defining an exposure time for an acquisition of a succession of images of a scene by photosensitive pixels of a pixel matrix of an indirect time-of-flight measurement sensor during a given exposure time, the method comprising a definition of the exposure time for said acquisition according to a saturation rate of the pixels of said matrix, distances between the sensor and elements of the scene and a standard deviation of the distances between the sensor and the elements of the scene.

[0024] Preferably, the method comprises a definition of the exposure time for said acquisition based on the saturation rate of the pixels of said matrix and a precision of the sensor corresponding to a ratio between the distances between the sensor and elements of the scene and the standard deviation of the distances between the sensor and the elements of the scene.

[0025] Advantageously, the method comprises a comparison of the saturation rate of the pixels with a saturation rate threshold, and - if the pixel saturation rate is higher than the said saturation rate threshold, a reduction in the exposure time, - otherwise a comparison of said precision with a precision threshold (Tprec), and O if the precision is greater than said precision threshold, a reduction in the exposure time, Or otherwise, an increase in exposure time.

[0026] In an advantageous embodiment, the method comprises adjusting the exposure time of the sensor by increasing or reducing the value of the exposure time by a percentage between a minimum step and a maximum step.

[0027] Preferably, the accuracy is evaluated from the mathematical formula 21, in <7? where z is a distance between a scene element relative to the sensor and oz is the standard deviation of the distances between the sensor and the scene elements.

[0028] In an advantageous embodiment, if the saturation rate threshold is incompatible with the precision threshold, then the method comprises updating the precision threshold with the saturation rate.

[0029] Other advantages and characteristics of the invention will appear on examining the detailed description of embodiments, which are in no way limiting, and the appended drawings in which:

[0030] [Fig.l]

[0031] [Fig.2]

[0032] [Fig.3]

[0033] [Fig.4] illustrate embodiments and implementations of the invention.

[0034] [Fig.l] illustrates an indirect time-of-flight measurement sensor (iTOF sensor). The iTOF sensor has a PXM pixel array. The iTOF sensor also has a CPU control unit configured to control the PXM pixel array.

[0035] The iTOF sensor is configured to emit a light signal into a scene in which the iTOF sensor is placed. This light signal can then be reflected by the different elements of the scene towards the PXM pixel matrix.

[0036] The PXM pixel matrix is ​​configured to carry out acquisitions of successions of images of the scene by accumulating the light signals that it detects.

[0037] The successions of acquired images allow the control unit UC to determine a phase shift between the signal emitted by the sensor and the signals reflected by the elements of the scene and detected by the pixel matrix PXM. The control unit UC is then configured to calculate a distance between this sensor and the elements of the scene from the determined phase shift.

[0038] The acquisition of the images is carried out during an adaptable exposure time so as to compensate for the different brightnesses of the acquired images.

[0039] The control unit UC is also configured to adapt the exposure time of the pixels of said PXM matrix for each acquisition of a succession of images.

[0040] The exposure time is adapted according to the exposure time adjustment method described below in relation to [Fig.2].

[0041] [Fig.2] illustrates a method of adjusting the exposure time of the pixels of a PXM pixel array of an iTOF sensor such as that illustrated in [Fig.l].

[0042] The adjustment method comprises an acquisition step 20 in which a succession of images is acquired by the pixel matrix PXM during a given exposure time Tint. The pixels then deliver the acquired data to the control unit.

[0043] The adjustment method further comprises a step 21 of determining STAT statistics to be taken into account for adjusting the exposure time Tint . As illustrated in [Fig.3], the statistics to be taken into account are a saturation rate S AT of the pixels (corresponding to the average AVG(SAT) of the saturation rate SAT of each pixel), the distances Z between the elements of the scene and the sensor and the standard deviation of these distances oz. Thus, the determination of said statistics comprises a determination of a saturation rate SAT of the pixels. The determination of said statistics also comprises a calculation of distances Z of the elements of the scene with respect to the sensor from the data delivered by the pixels. The determination of said statistics further comprises a calculation of a standard deviation oz of the distances of the elements of the scene with respect to the sensor. The standard deviation is calculated from a confidence index CI determined by the control unit UC.

[0044] In particular, the distances between the elements of the scene and the sensor and the standard deviation of the distances are used to determine an accuracy of the measurements with respect to the distances. In particular, the accuracy is determined by the formula , where z corresponds to the distance of an element of the scene with respect to the sensor, and oz corresponds to the standard deviation of the distances of the elements of the scene with respect to the sensor. An average of the accuracy AVG( ) corresponding to the accuracy of said succession of images acquired. However, for reasons of simplification, it is preferable to use the formula to evaluate the accuracy of the sensor. Indeed, this formula has the advantage of varying linearly with respect to the exposure time.

[0045] The adjustment method then comprises a step 22 of determining an exposure time Tint to be applied. In this step 22, the control unit UC determines the exposure time Tint to be applied for a next image acquisition based on the statistics determined in step 21.

[0046] Finally, the adjustment method comprises a step 23 of updating the exposure time Tint. In this step 23, the control unit UC updates the exposure time to be applied for the next acquisition by the exposure time Tint determined in step 22.

[0047] Steps 20 to 23 are repeated for each image acquisition.

[0048] [Fig.4] illustrates step 22 of determining the exposure time Tint to be applied for a next acquisition. This step 22 of determining the exposure time firstly comprises a comparison 30 of the saturation rate calculated in step 21 with respect to a given saturation rate threshold Tsat. A measured saturation rate higher than said saturation rate threshold Tsat means that the pixels of said PXM matrix are overexposed. A measured saturation rate lower than said saturation rate threshold Tsat means that the pixels of said PXM matrix are underexposed.

[0049] If the calculated saturation rate is greater than said given saturation rate threshold Tsat, then the exposure time Tint to be applied is reduced compared to the exposure time Tint previously applied.

[0050] If the calculated saturation rate is lower than said given saturation rate threshold, then step 22 comprises a comparison 31 of the saturation rate threshold Tsat and the precision threshold Tprec making it possible to determine whether the given saturation rate threshold Tsat is compatible with a given precision threshold Tprec.

[0051] The fact of initially comparing the saturation rate with said saturation rate threshold Tsat makes it possible to avoid saturating the pixels of the matrix.

[0052] If the given saturation rate threshold Tsat is compatible with the given precision threshold Tprec, then step 22 comprises a comparison 32 between a calculated precision and said given precision threshold Tprec. A precision greater than said precision threshold Tprec means that the data acquired from the PXM pixel matrix are too precise. A precision lower than said precision threshold Tprec means that the data acquired from the PXM pixel matrix are not sufficiently precise.

[0053] If the calculated precision is greater than the given precision threshold Tprec, then the exposure time Tint to be applied is reduced compared to the previously applied exposure time Tint.

[0054] If the calculated precision is lower than the given precision threshold Tprec, then the exposure time Tint to be applied is increased compared to the previously applied exposure time Tint.

[0055] Furthermore, if the given saturation rate threshold Tsat is incompatible with the given precision threshold Tprec, then step 22 comprises an update 33 of the precision threshold Tprec with the saturation rate Tsat. In practice, the precision threshold Tprec is updated to a value proportional to Tsat in order to avoid a phenomenon of flickering of the brightness of the images. The proportionality coefficient being an adjustable parameter of the algorithm between zero and one. Then, step 22 comprises said comparison 32 between the calculated precision and said given precision threshold Tprec, before adjusting the exposure time Tint as described previously.

[0056] During step 22, the control unit UC thus adjusts the exposure time of the sensor by increasing or reducing the value of the exposure time Tint by a percentage between a minimum step and a maximum step. This minimum step and this maximum step are defined parameters of the algorithm and known by the control unit.

[0057] The iTOF sensor is therefore configured to adapt the exposure time for an acquisition of a succession of images not only as a function of a saturation rate of the sensor but also as a function of the distance between the elements of the scene and the sensor as well as the standard deviation of the distances between the elements of the scene and the sensor.

[0058] In particular, the distance between the elements of the scene and the iTOF sensor as well as the standard deviation of the distances between the elements of the scene and the sensor make it possible to define an acquisition precision relative to the distance of the elements of the scene relative to the sensor.

[0059] The iTOF sensor is therefore configured to adapt the exposure time as a function of a saturation rate of the sensor and an acquisition precision. Such an adjustment of the exposure time makes it possible to optimize a compromise between an energy consumption of the sensor and a signal-to-noise ratio of the sensor. By taking into account the acquisition precision to define the exposure time, it is possible to reduce the exposure time to obtain a desired precision and a desired saturation of the pixels. The sensor therefore makes it possible to avoid an unwanted increase in the exposure time which would result in too great a precision of the sensor in relation to the distance of the elements of the scene from the sensor.

Claims

Claims

1. Indirect time-of-flight measurement sensor comprising: - a matrix of photosensitive pixels (PXM) configured to acquire a succession of images of a scene during a given exposure time, - a control unit (UC) configured to control an acquisition of a succession of images by the matrix of pixels (PXM) and to define an exposure time for this acquisition from a saturation rate of the pixels of said matrix (PXM) and a precision of the sensor, the precision of the sensor being: • determined by a ratio between the distances (z) between the sensor and elements of the scene and the standard deviation (oz) of the distances between the sensor and the elements of the scene, or • evaluated from the mathematical formula Ê., in which z is a distance between an element of the scene relative to the sensor and oz is the standard deviation of the distances between the sensor and the elements of the scene.

2. Sensor according to claim 1, in which the control unit (UC) is configured to compare the saturation rate of the pixels with a saturation rate threshold (Tsat), and to: - if the saturation rate of the pixels is greater than said saturation rate threshold (Tsat), reduce the exposure time (Tint), - otherwise compare said precision with a precision threshold (Tprec), and O if the precision is greater than said precision threshold (Tprec), reduce the exposure time (Tint), O otherwise, increase the exposure time (Tint).

3. The sensor of claim 2, wherein the control unit (UC) is configured to adjust the exposure time of the sensor by increasing or reducing the exposure time value (Tint) by a percentage between a minimum step and a maximum step.

4. Sensor according to one of claims 2 or 3, wherein, if the saturation rate threshold (Tsat) is incompatible with the precision threshold (Tprec), then the control unit is configured to update the precision threshold (Tprec) with the saturation rate (Tsat).

5. Method for defining an exposure time for an acquisition of a succession of images of a scene by pixels of a matrix of photosensitive pixels (PXM) of an indirect time-of-flight measurement sensor during a given exposure time, the method comprising a definition of the exposure time (Tint) for said acquisition from the saturation rate of the pixels of said matrix (PXM) and a precision of the sensor, the precision of the sensor being: • determined by a ratio between the distances (z) between the sensor and elements of the scene and the standard deviation (oz) of the distances between the sensor and the elements of the scene, or else • evaluated from the mathematical formula îi, in which z is a distance between an element of the scene relative to the sensor and oz is the standard deviation of the distances between the sensor and the elements of the scene.

6. Method according to claim 5, comprising a comparison of the saturation rate of the pixels with a saturation rate threshold (Tsat), and - if the saturation rate of the pixels is greater than said saturation rate threshold (Tsat), a reduction in the exposure time (Tint), - otherwise a comparison of said precision with a precision threshold (T 1 and prec / 9 O if the precision is greater than said precision threshold (Tprec), a reduction in the exposure time (Tint), O otherwise, an increase in the exposure time (Tint).

7. A method according to claim 6, comprising adjusting the exposure time (Tint) of the sensor by increasing or reducing the value of the exposure time (Tint) by a percentage between a minimum step and a maximum step.

8. Method according to one of claims 6 or 7, in which, if the saturation rate threshold (Tsat) is incompatible with the precision threshold (Tprec), then the method comprises an update of the precision threshold (Tprec) with the saturation rate (Tsat).

Images