Method for determining the phase measured by a time-of-flight measuring device, associated device and associated remote photoplethysmography system

Abstract

A method for determining the phase measured by a time-of-flight measuring device (20), comprising the emission of radiation by an illumination device (28) and the measurement of the backscattered radiation (R) on a target (3) by a photodetector (22) whose exposure time window phase shift with respect to the emitted radiation (E) successively takes k possible values, comprises the application of sequential iterations. These iterations include: reading and recording an electrical charge value associated with the backscattered radiation to the photodetector, accumulated during an exposure time window having a phase shift with respect to the phase of the emitted radiation; and determining the phase of the backscattered radiation from the charge value recorded in the current iteration and the charge values ​​recorded during the k-1 preceding iterations. Figure for the abstract: Fig. 1

Description

Title of the invention: Method for determining the phase measured by a time-of-flight measuring device, associated device and associated remote photoplethysmography system Technical field of the invention

[0001] The present invention relates generally to time-of-flight measuring devices.

[0002] It relates more particularly to a method for determining the phase measured by a time-of-flight measuring device, an associated device, and an associated remote photoplethysmography system.

[0003] The invention finds a particularly advantageous application in motor vehicles, and the monitoring of physiological parameters, in particular in the estimation of the heart rate, respiratory rate, heart rate variability, etc. of an occupant of a vehicle equipped with a time-of-flight measuring device. State of the art

[0004] Monitoring certain physiological parameters such as heart rate, heart rate variability, or respiratory rate by sensors makes it possible to obtain information on a user's state of health. Solutions based on optical sensors make it possible to take readings of these physiological parameters remotely.

[0005] A possible, but in no way limiting, field of application of these optical sensors dedicated to the monitoring of physiological parameters is the field of motor vehicles. Indeed, thanks to the installation of such sensors in the passenger compartment of a car it is possible to monitor the physiological parameters of its occupants, such as for example the driver. Cardiac and / or respiratory signs indicating stress, fatigue, or a state of intoxication of the latter can thus be detected remotely by the optical sensors, so as to limit the risks of road accidents.

[0006] For example, remote photoplethysmography (rPPG) consists of evaluating physiological parameters such as heart rate, its variability or the respiratory rate of a user, here, the driver of a motor car, using optical sensors monitoring, for example, the variation in blood volume or the volume of the rib cage.

[0007] The methods described in the state of the art may be based on the use of an RGB imaging camera, for “Red Green Blue” according to the name Anglo-Saxon, which allows RGB images of the reflectance of a person's skin to be obtained. These RGB images reflect the colorimetric variations induced by variations in blood volume, for example, and can be analyzed by rPPG reconstruction algorithms. An example of this type of algorithm is disclosed in document FR3114736, in order to extract the relevant information, here heart rate, for example.

[0008] rPPG measurements based on the use of RGB imaging cameras can be supplemented, or replaced by the addition of an imaging modality using optical time-of-flight measurement, commonly referred to as "TOF", for "time-of-flight", according to the Anglo-Saxon term. The addition of this additional data makes it possible to increase the reliability of physiological parameter readings, by limiting the influence of parameters such as ambient light as well as the influence coming from the user, such as their transverse or lateral movements, or even their skin color.

[0009] These time-of-flight measurement methods use active illumination and the detection of a backscattered signal to provide three-dimensional information related to depth, and two-dimensional information, related to the intensity or amplitude of the backscattered signal.

[0010] Examples of such rPPG measuring devices, as well as the algorithms for estimating the associated physiological parameters are described in the state of the art.

[0011] For example, in the document RONG et al., “Respiration and cardiac activity sensing using 3D cameras”. In: 54th Asilomar Conference on Signais, Systems and computers. 2020, the three-dimensional information provided by the time-of-flight measurement makes it possible to deduce the respiratory rate, while the association of the two-dimensional RGB and infrared information, acquired in this example, on two separate cameras makes it possible to trace the heart rate.

[0012] In a second example, the document NAHLER et al., “Exploring the usage of time of flight cameras for contact and remote Photoplethysmography”., In: 21st Euromicro Conference on Digital System design. 2018, uses only the two-dimensional infrared information acquired by a time-of-flight camera to deduce the heart rate, and the depth information is neglected.

[0013] In these two examples, the time-of-flight cameras used have an image frequency of 30 fps to 45 fps (fps, for "frames per second" in English, translated as images per second in French), knowing that the heart rate is usually between 60 and 200 beats per minute (also noted bpm), or between 1 Hertz and 3.33 Hertz.

[0014] However, it may be advantageous to carry out measurements at a frequency larger image size, in order to improve the reliability of remote photoplethysmography algorithms. Larger sampling allows in particular to be sensitive to more rapid variations in physiological parameters, while getting rid of the influence of measurement noise induced by the environment. Presentation of the invention

[0015] In order to overcome the aforementioned drawback of the state of the art, the present invention proposes updating at a so-called update frequency higher than the image frequency usually defined on time-of-flight measuring devices the quantities of interest, such as the phase, measured by a time-of-flight measuring device.

[0016] More particularly, according to the invention, there is proposed a method for determining the phase measured by a time-of-flight measuring device comprising the emission by an illumination device of amplitude-modulated radiation with a certain phase, and the measurement of the radiation backscattered on a target by a photodetector whose phase shift of the exposure time window with respect to the phase of the emitted radiation takes k possible values ​​as defined in the introduction, in which provision is made for the application of sequential iterations, where the phase shift of the exposure time window successively takes the k possible and distinct values ​​two by two. Each of the iterations includes: - a step of reading and recording an electric charge value associated with the radiation backscattered towards the photodetector, and accumulated during an exposure time window having started with a phase shift relative to the phase of the emitted radiation, - a step of determining the phase of the backscattered radiation from the charge value recorded in the current iteration, and the charge values ​​recorded during the k-1 previous iterations.

[0017] Thus, thanks to the invention, the value of the phase of the backscattered radiation can be updated each time a new accumulated charge value is recorded on the photodetector. In other words, a new charge value is determined at a frequency of up to k times the image frequency initially defined on the photodetector. One of the possibilities offered by this higher frequency sampling of the value of the phase of a backscattered radiation will be, for example, to improve the reliability of remote photoplethysmography algorithms.

[0018] Other advantageous and non-limiting characteristics of the method according to the invention, taken individually or in all technically possible combinations, are the following: - the k possible phase shift values ​​of the exposure time window are used successively, in a monotonous order, that is to say, in an ascending or descending order, - each iteration is of equal duration.

[0019] These two additional characteristics, for the first, make it possible to minimize the propagation of error that could be induced by too long a time between the phase shifts two by two, for the second, to ensure a continuous and regularly spaced flow in time of the phase values ​​determined according to the method.

[0020] Furthermore, it is possible in certain iterations of the method, or even in each iteration, to add a step of using the phase of the backscattered radiation to calculate a distance between the target and the time-of-flight measuring device.

[0021] According to the method, this distance thus determined between the target and the time-of-flight measuring device can be used as an input parameter in a remote photoplethysmography algorithm.

[0022] Another step can also be included in certain iterations, or even in each iteration of the method, which consists of determining an amplitude of the backscattered radiation, and / or a confidence in the measurement, this corresponding to twice the amplitude of the backscattered radiation, from the charge values ​​recorded during the k-1 previous iterations.

[0023] Thus, it would also be possible in another step included in some or all iterations of the method to use the phase and amplitude of the backscattered radiation or the confidence in the measurement to determine the reflectance of the target.

[0024] This reflectance value can then, depending on the method, be used as an input parameter in a remote photoplethysmography algorithm.

[0025] According to this method, it is also possible to include in some, or even all, of the iterations a motion blur detection step, this motion blur being induced by a movement of the target and / or time-of-flight measuring device, using the load value recorded in the current iteration, and load values ​​recorded during the k-1 previous iterations, and results in the cancellation of the following steps if an anomaly is detected.

[0026] The invention also proposes a time-of-flight measuring device, comprising an illumination device, a photodetector whose phase shift of the exposure time window relative to the phase of the emitted radiation takes k possible values, a storage unit and an electronic control unit, configured to: - Control the emission by the illumination device of continuous amplitude-modulated radiation, - Sequentially perform the following iterations using respec- tively the k possible values ​​for the phase shift: • Accumulate charges associated with the radiation backscattered by a target on the photodetector during an exposure time window that began with a phase shift relative to the phase of the emitted radiation, • Record the value of charges accumulated on the photodetector in the storage unit, • Calculate using the calculation unit and the memory unit, the phase of the backscattered radiation.

[0027] This device according to the invention can also be configured to, in certain iterations, or even in each one, calculate a distance between the target and the time-of-flight measuring device, using the phase of the backscattered radiation determined in the current iteration.

[0028] Similarly, the time-of-flight measuring device may be configured to, in at least certain iterations, calculate an amplitude of the backscattered radiation, and / or a confidence in the measurement, the latter corresponding to twice the amplitude of the backscattered radiation.

[0029] Thus, the described time-of-flight measuring device is also capable, in some or all iterations, of using the phase and amplitude of the backscattered radiation or the confidence in the measurement to calculate the reflectance of the target.

[0030] Finally, the time-of-flight measuring device can perform motion blur detection, in at least some of the iterations, the current iteration being interrupted to continue at the following iteration, with a new phase shift value of the exposure time window of the photodetector, if the detection is positive.

[0031] The invention also provides a remote photoplethysmography system including a time-of-flight measuring device as described above.

[0032] This remote photoplethysmography system is particularly suitable for evaluating the heart rate, and / or variations in the heart rate, and / or the respiratory rate using the phase of the backscattered radiation.

[0033] Of course, the various features, variants and embodiments of the invention may be combined with each other in various combinations to the extent that they are not incompatible or mutually exclusive. Detailed description of the invention

[0034] The description which follows with reference to the appended drawings, given as non-limiting examples, will make it clear what the invention consists of and how it can be implemented.

[0035] In the attached drawings:

[0036] [Fig. 1] is a schematic representation of the time-of-flight measuring device according to the invention and included in a remote photoplethysmography system;

[0037] [Fig.2] is a temporal representation of the radiation emitted and received by the time-of-flight measuring device of [Fig.l], as well as the signals controlling the acquisition of the radiation received by the photodetector of the device;

[0038] [Fig.3] is a flowchart presenting in iterative form, a method for determining the phase of the radiation received by the time-of-flight measuring device of [Fig.l];

[0039] [Fig.4] is a flowchart showing in detail the first iterations of the method described in [Fig.3], as well as its initialization.

[0040] A time-of-flight measuring device 20 is shown schematically in [Fig.l]. It comprises an illumination device 28, a photodetector 22, a processing unit 25, comprising a calculation unit 24 and a storage unit 26.

[0041] According to a non-limiting embodiment of the invention, this time-of-flight measuring device 20 is part of a remote photoplethysmography system 200, intended to monitor physiological parameters of a user.

[0042] These physiological parameters here designate the heart rate, the variability of the heart rate or the respiratory rate. By abuse of language, the frequency can be confused with the rhythm, thus, one can also speak of heart rate, of the variability of the heart rate, or even of the respiratory rate.

[0043] According to a non-limiting embodiment of the invention, the processing unit 25 corresponds for example to the electronic control unit of the remote photoplethysmography system 200, which also determines the aforementioned physiological parameters.

[0044] According to a second non-limiting embodiment, the processing unit 25 corresponds to an element distinct from a sensor (including for example the photodetector 22 and the illumination device 28) and / or from the aforementioned electronic control unit; such a distinct element may be an image processor.

[0045] According to a third non-limiting embodiment, that described here in [Fig.l], it is considered that the processing unit 25 forms an assembly with the illumination device 28 and the photodetector 22.

[0046] In an embodiment illustrated in [Fig.l], it is in particular a remote photoplethysmography system 200 implemented in the passenger compartment of a motor vehicle 1, to monitor the physiological parameters of the occupants of the vehicle, in particular the driver. The detection of an anomaly in the physiological parameters, such as the heart rate, its variability, or in the respiratory rate of the driver can trigger an alert if necessary.

[0047] The principle of time-of-flight measurement consists of emitting radiation using the illumination device 28 in the direction of a target 3. This radiation is designated by the term emitted radiation E in the remainder of the description.

[0048] Here, according to the embodiment described, the target 3 corresponds to the driver of the motor vehicle, in particular the region of the face, the neck as well as the bust.

[0049] The illumination device 28 thus emits radiation continuously over time.

[0050] This emitted radiation E is amplitude modulated, preferably periodically. Such periodic amplitude modulation is shown in [Fig.l].

[0051] This modulation of the continuous radiation takes, for example, the form of slots, with an amplitude alternating periodically between a high level and a low level. Other continuous and periodic radiation modulation patterns are also conceivable by those skilled in the art.

[0052] This involves alternating between a zero intensity level, where the illumination device 28 is off, and a high intensity level of arbitrary intensity, corresponding to the state where the illumination device 28 is on.

[0053] The emitted radiation E, continuous and periodic, also has a phase, here chosen to be zero, and serving as a reference for the origin. The phases defined below are defined as a function of a phase difference, called phase shift, relative to the phase of the emitted radiation.

[0054] In the example described here, the radiation E emitted by the illumination device 28 is located in the infrared range, in particular that of the near infrared extending from 700 nanometers to 2000 nanometers. More specifically, the emitted radiation E can have a wavelength included in narrower spectral windows, of around ten nanometers, around 850 nanometers or 940 nanometers for example.

[0055] The use of radiation emitted E in the infrared makes it possible on the one hand to be invisible to the eye, if the target 3 considered is the driver of a vehicle, thus eliminating any risk of dazzling the latter, especially during the night, but also to overcome variations in intensity linked to the environment, which is a frequent problem when the radiation is emitted in the visible range.

[0056] In particular, these variations in intensity may be linked to the ambient light in the passenger compartment of the vehicle depending on the sunlight, or even by shadows projected in the passenger compartment when a tree passes. It is also preferable, in the context of a remote photoplethysmography system, to favor wavelengths absorbed by the subcutaneous blood vessels. This is the case, for example, of the wavelengths cited previously in this description.

[0057] This infrared illumination device 28 can for example be an infrared laser suitable for being modulated at a high frequency, particularly between 50 and 500 megahertz. It can be a laser diode or a vertical cavity surface emitting laser diode (commonly called VCSEL, after the abbreviation of the Anglo-Saxon term).

[0058] The illumination device 28 may also correspond to a light-emitting diode (abbreviated as DEL in French, or known by the English abbreviation of LED, for “light-emitting diode”).

[0059] The modulation frequency of the illumination device 28 is chosen as a function of the desired range for the time-of-flight measuring device 20, as well as the precision of the desired measurement.

[0060] Here, in the embodiment presented where the time-of-flight measuring device 20 is included in a remote photoplethysmography system 200, placed in the passenger compartment of a car, the illumination device 28 is a VCSEL emitting at 940 nanometers, the intensity of which is modulated at a frequency of 100 megahertz. Indeed, the desired range, corresponding to the distance between the target 3, here the driver of the vehicle, and the time-of-flight measuring device 20, is of the order of 0.5 meters to 1.5 meters.

[0061] In the present embodiment, the use of a wavelength of 940 nanometers makes it possible in particular to increase the depth of penetration of the emitted radiation, and to limit the influence of the skin color of the conductor considered as the target 3.

[0062] The emitted infrared radiation is then backscattered by the target 3. In other words, the emitted radiation E is reflected diffusely by the target 3 towards the original direction, i.e. towards the illumination device 28.

[0063] Here, it is the driver of a vehicle, and the emitted radiation E is notably diffused at the level of the face, neck and bust region.

[0064] This radiation reflected by the target 3 and received by the time-of-flight measuring device 20 corresponds to backscattered radiation R. The backscattered radiation R is then collected on a photodetector 22. This makes it possible to transform the absorbed light into a measurable quantity, here, an electric current in the form of accumulated electric charges. In the remainder of this description, reference will be made to accumulated charges, or to accumulated electric charges, interchangeably.

[0065] This photodetector 22 is preferably similar to a camera provided with a matrix composed of a plurality of pixels, each of the pixels being capable of accumulating electrical charges, so as to be able to image a scene including the target 3. It may be, for example, a CCD camera, for “charge-coupled device”, or a CMOS camera, for “Complementary metal-oxide-semiconductor”. Commonly, it is a CMOS camera which is integrated in time-of-flight measuring devices 20.

[0066] The photodetector 22 is oriented so as to image the target 3 at least partially in its field of view.

[0067] The photodetector 22 is sensitive to the spectral range of the emitted radiation.

[0068] Here, it is the near infrared range, as previously described. The Photodetector 22 is possibly sensitive to a wide spectral range of wavelengths, from visible (from 400 nanometers) to near infrared (up to 2000 nanometers). The spectral range to which photodetector 22 is sensitive depends on its composition; here the example of a photodetector 22 made of silicon (symbol Si) is presented.

[0069] In the example described here, the photodetector 22 is sensitive in a narrow spectral window of a few tens of nanometers to a hundred nanometers and centered around 940 nanometers, corresponding to the wavelength of the emitted radiation E.

[0070] In order to limit the detection on the photodetector 22 of parasitic radiation, not emitted by the illumination device 28, a bandpass filter centered around the wavelength of the emitted radiation can also be added in front of the photodetector 22. This thus makes it possible to reject the parasitic radiation if its wavelength is not included in the passband.

[0071] This bandpass filter has for example a width between 20 and 40 nanometers, when the illumination device 28 corresponds to a VCSEL, or between 100 and 150 nanometers when the illumination device 28 corresponds to a light-emitting diode.

[0072] The photodetector 22 is controlled using control signals serving as an electronic shutter, and generated for example by the calculation unit 24.

[0073] Depending on the distance between the target 3 and the photodetector 22 of the time-of-flight measuring device 20, the backscattered radiation R has a phase shift relative to the emitted radiation E. This phase shift is illustrated in Figure 1, and indicated by the quantity therein.

[0074] Considering the phase of the emitted radiation E as a reference for the origin, the backscattered radiation therefore has a phase also noted in the remainder of the description, which is defined as the phase difference (or phase shift) relative to this origin.

[0075] Knowledge of the phase of the backscattered radiation R then makes it possible to determine a distance between the time-of-flight measuring device 20 and the target 3.

[0076] Here, the detection of the backscattered radiation R is carried out on each pixel of the matrix of the CMOS camera, used as photodetector 22, thus making it possible to make a three-dimensional reconstruction of the target 3 and the scene which surrounds it pixel by pixel. For this, the necessary imaging optics are added.

[0077] The invention, which comprises a method for determining the phase (P) of the backscattered radiation R, is set out in the remainder of the description by relating the photodetector 22 to a single pixel of the CMOS camera, but the measurement method detailed below can obviously be applied to the entire pixel matrix of the CMOS camera, in order to reconstruct an image reproducing the scene including the target 3.

[0078] In addition to the phase of the backscattered radiation, associated with distance information, the photodetector 22 is also sensitive to an amplitude of the backscattered radiation R. More particularly, in the case of time-of-flight measurement, it is a confidence C in the measurement which is determined. Here, the confidence C in the measurement corresponds to twice the amplitude of the signal. Thus, these two quantities are considered equivalent, to within a factor of two.

[0079] It is also possible, by knowing a distance between the target 3 and the time-of-flight measuring device 20 or the phase V, and the amplitude or the confidence C, to deduce a relative value of the reflectance p of the target. The reflectance p is a characteristic of the target 3, and reflects the proportion of the light which is re-emitted by the target 3 compared to the light initially received by this target 3.

[0080] It is possible to take into account the distance between the target 3 and the time-of-flight measuring device 20 in order to correct the reflectance p for variations induced by a displacement in the direction of the optical axis, i.e. a longitudinal displacement of the target 3 relative to the axis of the time-of-flight measuring device 20. This displacement corresponds to what is called motion blur.

[0081] However, the measurement of the phase of the backscattered radiation R requires the application of the method detailed in the remainder of this description.

[0082] Indeed, in order to measure the phase V of the backscattered radiation R, while avoiding possible ambiguities about its value, the time-of-flight measuring device 20 sequentially performs a series of several sub-measurements. These are then used in order to be able to deduce a phase value V, as well as a confidence value C.

[0083] For greater clarity in the remainder of this presentation, the number of sub-measurements used for determining a phase value P and a confidence value C will be designated by the reference k.

[0084] In other words, k sub-measurements made on the photodetector 22 are used to determine a phase value P and a confidence value C for the backscattered radiation.

[0085] As illustrated in [Fig.2], each of the k sub-measurements is made using a different control signal from the photodetector 22. These control signals differ by a phase shift relative to the phase of the emitted radiation. There are thus k different phase shift values ​​which are used on the photodetector 22, and these k values ​​are distinct two by two.

[0086] This control signal of the photodetector 22 has a periodic modulation, which is notably of the same period as the emitted radiation.

[0087] In this embodiment, the control signal is modulated with a frequency of 100 megahertz, identical to the amplitude modulation frequency of the emitted radiation.

[0088] This periodic modulation of the control signal of the photodetector 22 can have a form identical to that of the emitted radiation, with an alternation between a high level and a low level. The high level makes it possible to define an exposure time window. During this exposure time window, the photodetector 22 is exposed to the backscattered radiation R. It is thus capable of accumulating electrical charges by converting the light information received.

[0089] At the end of the exposure time window, the photodetector is no longer exposed to the backscattered radiation R, and stops accumulating charges.

[0090] It is also possible to consider the accumulated electrical charges integrated over several exposure time windows for a given phase shift value. For simplicity of description, the accumulated charges are considered over a single exposure time window.

[0091] In practice, the k sub-measurements are made sequentially, by applying a control signal to the photodetector 22 provided with a new phase shift value at the end of each of the acquisition time windows, as illustrated in [Fig.2].

[0092] Thus, in the remainder of the description, the acquisition time windows associated with a phase shift will be considered one by one, rather than a periodic control signal as a whole.

[0093] In summary, the photodetector 22 accumulates electrical charges during an exposure time corresponding to the width of the exposure time window. The accumulation of these electrical charges begins with a possible latency time, corresponding to the phase shift of the exposure time window relative to the phase of the emitted radiation E.

[0094] The electrical charges accumulated during a given exposure time window result from a conversion of the photons of the backscattered radiation R arriving at the photodetector 22 during the aforementioned window into electrical charges.

[0095] The different phase shift values ​​of the exposure time window of the photodetector 22 can be controlled by the processing unit 25 for example, or another separate element.

[0096] In the presented embodiment, four sub-measures are used to de- end a phase value V. So, k = 4 in this example.

[0097] Throughout the description, the symbol “°” preceded by a numerical value refers to an angle value indicated in degrees.

[0098] Four examples of phase shift of the exposure time window relative to the phase of the emitted radiation can be 0°, 90°, 180° and 270°.

[0099] In another possible embodiment, only three phase shift values ​​are used, and are 0°, 120°, 240°.

[0100] In another possible embodiment, six phase shift values ​​are used, 0°, 60°, 120°, 180°, 240°, and 300°.

[0101] These two other possible embodiments will be detailed later.

[0102] Taking the example embodiment in which four phase shift values ​​of the acquisition time windows are used, the electrical charges accumulated for a phase shift of 0°, 90°, 180°, 270° will be called respectively Qb Q2, Q3, Q 4*

[0103] The possible phase shift values ​​are thus used sequentially on the photodetector 22, preferably in a monotonic order, that is to say in an increasing order or in a decreasing order. This makes it possible to ensure regular temporal sampling between the measurements, thanks to the maintenance of the same temporal period between the updating of the pairs of values ​​Qi and Q3, and Q2 and Q4, between the series of sub-measurements, as explained later in the description.

[0104] Indeed, advantageously, the subtraction between the pairs of values ​​of opposite phase shift of 180 degrees, namely Qb Q3 and Q2, Q4 eliminates the parasitic influence of ambient light. Thus, the measurement is not disturbed by a fluctuation of the parasitic light even at a frequency of 1 hertz, or 60 beats per minute, which corresponds to the normal heart rate of a human.

[0105] Here, the ascending order is applied. Thus, during one embodiment of the method of the invention, the value of the phase shift of the exposure time window of the photodetector is first set to 0°, then to 90°, then 180° and 270°, before being reset to 0°, then to 90°, and so on, cyclically during the following sub-measurements. This order is illustrated in [Fig.2].

[0106] After each exposure time window, during which the electrical charges are accumulated, these are read on the photodetector 22. This action is associated with a time, called reading time.

[0107] An additional waiting time is nevertheless present, and corresponds to the duration necessary to avoid possible overheating of the photodetector 22, as well as to ensure time synchronization between the photodetector 22 and the elements with which it communicates.

[0108] Preferably in the invention, this waiting time is distributed from identically between the k sub-measures, so that the sub-measures all have the same duration.

[0109] The duration of a sub-measurement thus includes the exposure time, the reading time, and the waiting time.

[0110] The image frequency (also called in English language “frame per second”) of the photodetector 22 of a time-of-flight measuring device 20 is usually defined as the inverse of the sum of the durations of the k sub-measurements used to generate a phase and confidence value C.

[0111] A method for determining the phase of the backscattered radiation R is described here, making it possible to update the value of the phase 9 at a frequency, called the update frequency, which is higher than the image frequency defined on the photodetector 22.

[0112] Indeed, the value of the phase is determined, and therefore updated each time a new sub-measurement is made, providing a new value of accumulated electrical charge.

[0113] The determination of phase V is made using this new accumulated charge value, obtained during the sub-measurement considered and the accumulated charge values ​​determined during the k-1 previous sub-measurements.

[0114] In other words, each accumulated charge value obtained during a sub-measurement will be used k times to determine k sequential values ​​of the phase P of the backscattered radiation. The value of the phase V of the backscattered radiation R is thus modified incrementally, as soon as a new accumulated charge value is recorded.

[0115] The invention thus makes it possible to have an update frequency k times faster than the image frequency defined on the photodetector 22.

[0116] The method set out in the invention is also valid for determining the confidence C of the backscattered radiation R at the same update frequency as for the phase and will therefore be detailed in parallel.

[0117] This method is described in the flowchart of [Fig.3]. The method can be described recursively, by describing an initialization, followed by the sequential application of iterations.

[0118] Unlike the sub-measurements, the iterations include additional calculation steps, while the sub-measurements only include the exposure of the photodetector 22 to the backscattered radiation R, the reading of the accumulated charge, and a waiting time.

[0119] During initialization, k sub-measurements are carried out, one after the other, using the k phase shift values ​​of the exposure time window, the values ​​of which are distinct two by two. Each sub-measurement consists of exposing the photodetector 22 to the backscattered radiation R during an exposure time window that began with a phase shift relative to the emitted radiation.

[0120] In the embodiment considered, k is equal to 4, and the four phase shift values ​​used are the following: 0°, 90°, 180° and 270°, as mentioned previously in the description.

[0121] As mentioned previously, it is also preferable here that these four possible phase shift values ​​are used in ascending order.

[0122] Thus, in the embodiment presently described, the first sub-measurement is made with a phase shift of 0° of the exposure time window of the photodetector 22, the second with a phase shift of 90°, the third with a phase shift of 180°, and the fourth and last sub-measurement of the initialization with a phase shift of 270°.

[0123] During the exposure time window, the photodetector 22 accumulates electrical charges. The value of these is noted Qi0, with i = {1, ...k]. The second sub-index indicates that these charges were accumulated during initialization.

[0124] At the end of the exposure time window, the charges accumulated on the photodetector 22 are read, and recorded in the storage unit 26 as they go.

[0125] At the end of the k sub-measurements made during initialization, a first phase value of the backscattered radiation R can be determined, as well as a first confidence value Co.

[0126] The determination is made by the calculation unit 24, using the k accumulated charge values ​​recorded in the storage unit 26 during initialization.

[0127] A non-limiting example of possible formulas for determining the phase and confidence values ​​is the following, in the embodiment where four phase shift values ​​have been used on the photodetector 22:

[0128] ( ,-Q, ) , which corresponds to the phase of the backscattered radiation. (p n = arc tan 7——“ 0 ( <et .0-e ; ,0)

[0129] E / TT", which corresponds to the confidence in the Q) = y ( ô ! 0 - 0 ) + \ ^2,0 " ^4,o ) measurement of backscattered radiation R, i.e. twice the amplitude of the backscattered radiation.

[0130] Obtaining the phase and confidence values ​​marks the end of initialization.

[0131] In the remainder of the method described in the invention, sequential iterations are applied. Each of the iterations is described in a unique form, and comprises at least two steps.

[0132] During these sequential iterations, the k phase shift values ​​are used respectively, and cyclically. In other words, a phase shift value of the window exposure time is used per iteration.

[0133] The first step of an iteration is a step of reading an accumulated charge value Q;, i={1,...,k] on the photodetector 22, acquired during an exposure time window whose phase shift has taken the ith possible value. This accumulated charge value Q;, is recorded in the storage unit 26.

[0134] The second step is to determine the value of the phase and the confidence of the backscattered radiation R in order to update them, given that a new accumulated charge value Q; has been recorded. The phase and the confidence of the backscattered radiation R are calculated by the calculation unit 24 using the newly recorded charge value Q; and the accumulated charge values ​​recorded during the previous k-1 iterations, and which are stored in the storage unit 26.

[0135] Thus, at each iteration, the storage unit 26 has in memory an accumulated charge value for each of the k phase shift values ​​of the exposure time window, i.e. the set of values ​​{Qi, ..., Qi,. Qk}. These values ​​are updated as the iterations progress, always in the same order.

[0136] In the described embodiment in which four phase shift values ​​are used, during a given iteration, the determination of the phase and the confidence of the radiation can be done with the following formulas, in a non-limiting manner:

[0137] , and L x2 / x2, where the values ​​of Qb Q2, c = vl ô r + \Qi-Qj Q3, Q4 are stored in the storage unit, and were determined during the previous k-1 = 3 iterations, and during the current iteration. The amplitude of the backscattered radiation R is then deduced as being equal to twice the confidence C in the measurement.

[0138] Thus, a new phase value V and confidence value C for the backscattered radiation R is thus determined at the end of each iteration, as soon as a new accumulated charge value is recorded.

[0139] This method makes it possible to provide a continuous stream of phase and confidence data associated with backscattered radiation, each time a new sub-measurement is carried out.

[0140] It is thus preferable, although not necessary, that the sub-measurements be of the same duration, in order to obtain a regular flow of data.

[0141] In particular, the choice of using the k different phase shift values, here four in number, following an order relationship, here increasing, makes it possible to symmetrically update the two terms of the differences q _ qj and q _ qj, so that the durations between the updating of Ql and are equal to the duration between the updating of and and so on between the different iterations se- quantitative.

[0142] The initialization and the first four iterations carried out in an embodiment where four phase shift values ​​are used are explained in the remainder of the description.

[0143] This initialization and the first four iterations are illustrated in the flowchart of [Fig.4].

[0144] The initialization for an embodiment where four phase shift values ​​are used has already been described previously, and illustrated in [Fig. 3], by the brace "INIT.". Here, the four phase shift values ​​can also be arranged in ascending order. An accumulated charge value is determined and stored in the storage unit 26 for each of the four possible phase shift values, here, 0°, 90°, 180° and 270°, associated with the accumulated charge values ​​(Ko, (Ko, (Ko, Q4,o respectively.

[0145] At the end of the fourth sub-measurement with a phase shift value of the exposure time window of the photodetector 22 of 270°, a first phase and confidence value is calculated for the backscattered radiation R, for example using the formulas stated previously.

[0146] The phase shift of the exposure time window of the photodetector 22 is then reset to 0°, and then the first iteration of the method begins. This begins with a first step, which consists of accumulating electrical charges on the photodetector 22 during the 0° phase shift exposure time window. At the end of this step, the accumulated electrical charge value is read and recorded in the storage unit 26. This accumulated electrical charge value will be noted Qb

[0147] A second step then begins, where the phase P and confidence C values ​​are updated, given that a new accumulated electric charge value has just been recorded. The three accumulated charge values ​​recorded in the storage unit 26 during the three previous sub-measurements, namely Q4jo, (Ko, and (Ko are used for the calculation of the new set of phase shift and amplitude values ​​by the calculation unit 24, for example using the following formulas:

[0148] (Q, ,-Q,) and / 7?2 / TT-. ^3.0 /

[0149] In other words, the set of phase shift and amplitude values ​​is updated using the new recorded value Qb and the k-1, here three in number, other accumulated charge values ​​which are kept identical.

[0150] A second iteration begins, where the value of the phase shift of the exposure time window is set to 90°. This second iteration begins with a first step, where the electrical charges are accumulated on the photodetector 22 according to the exposure time window. At the end of this window, the accumulated charge value Q2 is

[0151]

[0152]

[0153]

[0154]

[0155]

[0156]

[0157]

[0158]

[0159]

[0160]

[0161]

[0162]

[0163] read and recorded in the storage unit 26. A second step then consists of updating the values ​​of the phase P and the confidence C, given that a new electric charge value has been recorded. The computing unit 24 can determine the phase shift and amplitude values ​​as follows: = e ' c =J(6r6 3j) ) 2 + (62-64.0) 2 ' ^3.0 ) Once again, the value set is updated using the new recorded accumulated electric charge value and the three values ​​recorded in the three previous iterations, in descending chronological order Qb Q40, and Q30. A third iteration is made following the second. The phase shift value of the exposure time window of the photodetector 22 is set to 180°. A first step consists of accumulating the electric charges on the photodetector 22, followed by their readings and their recordings in the storage unit 26. The accumulated electric charge value recorded during this first step of the third sub-measurement is noted Q3. The second step of the iteration is then to determine a new value for the phase and for the confidence C, using the accumulated charge value Q3 recorded during the current iteration, and the three accumulated charge values ​​recorded during the three previous sub-measurements, in descending chronological order, Q2, Qi and Q40. Thus, the phase shift and amplitude can be formulated as follows: ( o -Q, „ ), and / 3 / 75“. =arctan_^ c^q^) +(e2-e40) The fourth iteration is performed following the third. The phase shift of the exposure time window of the photodetector 22 is then equal to 270°. A first step consists of reading and recording the value of electrical charges accumulated on the photodetector 22. This is noted Q4. Then a second step is to update the phase value and the backscattered radiation confidence value, using the accumulated charge value from the current iteration, and the three accumulated charge values ​​recorded in the previous iterations, Q3, Q2, and Qb. For example, a possible formula for their values ​​might be: (ag,) , and / 7 ~2 / ~2~. p = c= v(6r63)+(e2-e4) For the next iteration, the phase shift of the exposure time window of the photodetector 22 is reset to 0°, and the associated accumulated charge value Qi is ac technicalized, as well as the phase value V and confidence C using the accumulated charge values ​​Q2, Q3, and Q4 recorded during the three previous iterations.

[0164] This succession of iterations will thus be repeated, sequentially using the four phase shift values, and updating the phase and confidence C of the backscattered radiation R using the accumulated charge values ​​recorded in the three previous iterations as well as the accumulated charge value recorded in the current iteration.

[0165]

[0166]

[0167]

[0168] The method described above also makes it possible to determine other quantities of interest, as mentioned at the start of the description. These quantities of interest, which include but are not limited to the phase and confidence C in the measurement and / or the amplitude of the backscattered radiation R, may also be generated at an update rate which may be higher than the frame rate set on the photodetector 22. In particular, the phase of the backscattered radiation R determined according to the method is associated with a distance P. This distance P corresponds to the distance separating the target 3 from the time-of-flight measuring device 20. The distance P is calculated as a function of the phase (P of the backscattered radiation R determined according to the method according to the following formula: P = , °ù C ■ jfM'J corresponds to the speed of light, P to the phase of the backscattered radiation R determined according to the method, and f to the modulation frequency of the emitted radiation E.

[0169] Here, as a reminder, the modulation frequency of the emitted radiation E is J — 100 megahertz.

[0170] The values ​​of c and j are constants, only the value of the phase P of the backscattered radiation R varies. Thus, according to the method described above, a new distance value P can be determined at each update of the phase V of the backscattered radiation R, at an update frequency which can be k times higher than the image frequency defined on the photodetector 22.

[0171] The value of the distance P thus determined according to the invention can be used as an input parameter within an algorithm.

[0172] In certain embodiments, the main quantity of interest is the distance P, and its determination only requires knowledge of the phase V of the backscattered radiation. The calculation of the confidence C in the measurement, or equivalently of the amplitude of the backscattered radiation R during the method, can then be optional in these embodiments. In practice, the determination of the confidence C in the measurement and of the associated amplitude is most often advantageous in the context of the measurement, since it is a quantity translating the amplitude of the emitted light. reflected by target 3, and which is not influenced by ambient light or even its possible fluctuations.

[0173] For example, if we consider the embodiment in which the time-of-flight measuring device 20 is part of a remote photoplethysmography system 200, the distance P determined according to the method described, and the three-dimensional scene thus reconstructed by considering a photodetector 22 containing a matrix of pixels, can be used as an input parameter of a heart rate estimation algorithm. The distance P can for example be used for the contour detection of the occupant of the motor vehicle. This contour can be used for object tracking within the framework of the aforementioned algorithm, while its variation can possibly be used to determine the respiratory rate.

[0174] This is also the case if we consider an embodiment where the time-of-flight measuring device 20 is integrated into a simple three-dimensional imager, where the photodetector 22 includes a pixel matrix. In this case, the method described in the invention can allow a three-dimensional reconstruction with a scene update frequency that can be k times faster than the image frequency initially defined on the photodetector 22.

[0175] According to other exemplary embodiments, the reflectance p is also considered as a quantity of interest.

[0176] The reflectance p is defined relatively using the following formulas: p CP2 ex C çp where the symbol “ means “proportional to”, P corresponds to the reflectance, C to the confidence in the measurement, P to the distance, (P to the phase. The amplitude of the backscattered radiation R being equal to the confidence C in the measurement to within a factor of two, the transposition of the reflectance P as a function of the amplitude is obvious to those skilled in the art.

[0177] Here, the reflectance p is defined relatively by taking into account the distance P, defined as the distance between the target 3 and the time-of-flight measuring device 20. This makes it possible to reduce the impact caused by a movement of the target 3 along the measurement axis.

[0178] The determination of the reflectance p thus requires both the determination of the associated confidence C or, equivalently, that of the amplitude of the backscattered radiation, and the determination of the phase (P) of the backscattered radiation, or even of the distance P associated with the phase P.

[0179] This reflectance information p of the target 3, although relative, is sufficient, in the embodiment where a remote photoplethysmography system 200 is considered, since the determination of the physiological parameters is done as a function of the variation of the reflectance p of the target.

[0180] The value of the reflectance p of the target 3 can be updated during each of the iterations described in the method which is the subject of the invention, at an update frequency which can be k times higher than the image frequency defined on the photodetector 22.

[0181] This reflectance value p of the target 3 thus obtained can for example be used as an input parameter for certain algorithms for estimating physiological parameters, such as heart rate within a remote photoplethysmography system 200.

[0182] Indeed, by taking the skin as target 3 here, the determination of its reflectance makes it possible to access the variations thereof induced by blood circulation. During a blood influx caused by heartbeats, the increase in blood volume results in a reduced reflectance p.

[0183] These variations in reflectance P thus make it possible to trace the heart rate, as well as the variability of the heart rate or the respiratory rate.

[0184] According to another possible embodiment, the confidence C of the backscattered radiation R generated according to the method of [Fig.4] is directly used in an algorithm for estimating physiological parameters, in particular for carrying out object tracking.

[0185] The embodiments described above are by way of example, and are in no way intended to be limiting.

[0186] Thus, the different quantities of interest, such as the phase and confidence C, the distance P and the reflectance p of the target 3, generated according to the method described in the invention can be used in various applications, in association with each other, or on the contrary, in a dissociated manner.

[0187] Thanks to the method, the quantities are generated at an update frequency which can be k times higher than the image frequency defined on the photodetector 22. These quantities can then be put into input parameters of algorithms, here, it is for example an algorithm for estimating physiological parameters within a remote photoplethysmography system 200, in which the photodetector 22 can take the form of a camera comprising a matrix of pixels.

[0188] In this exemplary embodiment, the exploitation of the generated quantities of interest can be done according to several variants. Thus, the determination of the physiological parameters, such as the heart rate, the variability of the heart rate, as well as the respiratory rate can be done using various combinations or not of the different quantities determined. For example, it is possible to use the distance P, the reflectance p, the confidence C in the measurement, alone, or even the association of the reflectance p and / or the distance P and / or the confidence in the measurement C.

[0189] In addition, in order to reduce calculation times, it is also possible to include object tracking within the algorithm, in order to only take into account the pixels of the photo- detector 22 which are relevant.

[0190] According to different variants, object tracking can be done in different ways, such as, for example, using confidence C, or using distance P, or a combination of the two.

[0191]

[0192] The formulas used for determining the phase of the backscattered radiation R and the confidence C thereof are given as examples. Other formulas can also be used in the method described in the invention, and their use is left to the discretion of those skilled in the art. For example, the phase *P of the backscattered radiation R can also be determined in the following way, considering the quantities X and Y defined from the accumulated charge values, such that X = Q1-Q3 and Y = Q2-Q4; then:

[0193]

[0194] 3 + X otherwise In this example, the distance P is then determined by: P = , where c corresponds 8. f 'moc! to the speed of light, to the phase of the backscattered radiation R and fmod to the modulation frequency of the emitted radiation.

[0195] It is also possible to add an additional step, for example within each iteration of the method. This additional step is also illustrated in Figures 3 and 4, by the box entitled “F” and consists of performing a motion blur detection, linked to the movement of the target 3 relative to the photodetector 22, here assumed to be fixed.

[0196] This motion blur detection can be done using the checks described on p.ll of the document HANSARD et al., “Time-of-flight cameras: Principles, Methods and applications”, Springer, 2012.

[0197] These checks consist of detecting anomalies in the electrical charges accumulated on the photodetector 22 during the different phase shift values, which can be, among other things, induced by motion blur.

[0198] Here, in the embodiment where four phase shift values, 0°, 90°, 180°, and 270° are used, the exposure time windows are in phase opposition, two by two, the 0° phase shift window with that of 180° and that of 90° with that of 270°. These pairs of exposure time windows in phase opposition make it possible to generate the pairs of accumulated charge values ​​Qi and Q3, and Q2 and Q4. In the absence of any measurement anomaly, in particular in the absence of motion blur, the total charge accumulated on two windows in phase opposition must be equal to a constant K.

[0199] Indeed, taking into account the sum of the values ​​of charges accumulated on two time windows in phase opposition is equivalent to considering that the photodetector 22 is constantly exposed.

[0200] Thus, in the absence, for example, of movement, which would modify this sum, the “Rule of plus” defined in the previous document is verified: Qi + Q3 = Q2+Q4 = K.

[0201] Equivalently, the “rule of least”, also defined in the previous document, can also be verified: IQ1-Q3I+IQ2-Q4I = K.

[0202] These two examples of verifications are not limiting, and other criteria for evaluating motion blur can be envisaged.

[0203] In practice, within the method described, for the embodiment considered, where four phase shifts are used on the photodetector 22, these verifications can take the form of three equivalent conditions:

[0204] I, I . | 6,+63 . | . | 62+64 | , where 162+64 I x "ciiil 5 | I ' | | threshold ^threshold is chosen arbitrarily according to experience, and whose value is close to zero. These conditions follow directly from the equalities defined in the rules of plus and minus defined above.

[0205] This step is done during each iteration using the accumulated load value recorded during the current iteration and the load values ​​recorded during the previous kl=3 iterations.

[0206] Thus, this verification can be done at each of the iterations, at the update frequency defined previously.

[0207] If one of the three conditions stated above is not verified, the measurement is considered corrupted. In practice, in the embodiment where the photodetector 22 comprises a matrix of pixels, the pixel considered is deactivated at least temporarily, and marked as corrupted. The quantities, such as the phase P, the confidence C, the distance P and the reflectance p are not determined.

[0208] This motion blur detection step is particularly interesting to avoid propagating an error within an algorithm.

[0209] In particular, if a remote photoplethysmography system 200 is considered, as well as the associated physiological parameter estimation algorithm, the rejection of the pixels of the photodetector 22 estimated to be fuzzy makes it possible to improve the reliability of the estimated physiological parameters, and to reduce the influence of the movements of the target 3.

[0210] Until now, the embodiment described for the invention used four phase shift values ​​of the exposure time window of the photodetector 22. However, it was mentioned at the beginning of the description the possibility of using three or six phase shift values ​​of the exposure time window, without there being impact on the principle of the process.

[0211] For example, considering a second possible embodiment where three phase shift values ​​of the exposure time window would be used, then the phase of the backscattered radiation R would be determined by the calculation unit 24 each time a new charge value is recorded on the storage unit, using the accumulated charge values ​​recorded during the two previous iterations. Then, the phase of the backscattered radiation R can be expressed in the form:

[0212] _ -1 / [T \, where Qi, ...,Q3 correspond to the values ​​of charges ac- ~îan \ v5 • ïëTO; ) accumulated for a phase shift of 0°, 120° and 240° respectively; Similarly, the confidence in the radiation measurement can be expressed as: [02B] c = i 2 + e Î+ e 2.6], where Q„ ....Q, correspond to the accumulated charge values ​​for a phase shift of 0°, 120° and 240° respectively

[0214] A third embodiment can also be described, using six phase shift values ​​of the exposure time window. In this case, the accumulated charge values ​​recorded during five previous iterations would be used to determine the phase of the backscattered radiation, which can then be expressed as:

[0215] _ .ri \ where Qi, ...,Q5 correspond to the values ​​of charges ac- accumulated for a phase shift of 0°, 60°, 120°, 180°, 240° and 300° respectively; Similarly, the confidence in the radiation measurement can be expressed as:

[0216] -Q +Q -o j2 + 3 (oQ )2 ' °ù Q1' • • 'Qs correspond to the charge values C~ 2.^3 ' accumulated for a phase shift of 0°, 60°, 120°, 180°, 240° and 300° respectively.

[0217] The present invention is in no way limited to the embodiments described and shown, but those skilled in the art will be able to provide any variant in accordance with the invention.

[0218] Typically, the choice of the photodetector 22 is left to the discretion of the person skilled in the art. It may preferably be a photodetector 22 comprising a pixel matrix, in order to reconstruct an extended scene. In this case, the method of the invention is applied to each of the pixels.

[0219] Alternative solutions may also be envisaged for the photodetector 22, such as for example photodetectors provided with several wells. These photodetectors are better known under the Anglo-Saxon term “multi-TAP sensor”. Such photodetectors may be provided with two (“dual”) or four (“quad”) wells, forming regions of distinct pixels on the pixel matrix of the photodetector. Each of these wells operates independently and in parallel, and can thus accumulate, read and record. The use of the different phase shift values ​​of the exposure time window could thus be parallelized on the different wells during the same iteration, while continuing to update the phase of the backscattered radiation, and the other potential quantities of interest at each acquisition.

[0220] Similarly, the method described can be implemented by different elements of the remote photoplethysmography system 200, according to alternative embodiments. In the preceding description, the determination of the quantities of interest, in particular the phase of the backscattered radiation, its confidence, as well as the distance P associated with the distance between the target 3 and the time-of-flight measuring device 20, or the reflectance p of the target 3 is done using a storage unit 26 and a calculation unit 24, within a processing unit 25. In practice, this processing unit 25 can be physically located in several elements within a remote photoplethysmography system 200.

[0221] In a first possible embodiment, the processing unit 25 can be included in a sensor comprising an illumination device 28, a photodetector 22. The assembly would then form a time-of-flight measuring device as described previously.

[0222] An alternative solution would be to include the processing unit 25 in a separate image processor (ISP), which would receive as input the electrical charge values ​​from a sensor comprising an illumination device 28 and a photodetector 22, and would output the aforementioned quantities of interest. These quantities would then be input to a remote photoplethysmography algorithm in the associated system. The assembly comprising the image processor and the sensor would then form a time-of-flight measurement device, according to the preceding description.

[0223] A second alternative could be to group the processing unit 25 with the electronic control unit of the remote photoplethysmography system 200, which would take the form of a system on chip dedicated to the determination of the physiological parameters. This system on chip would then be configured to carry out the method described, by receiving as input the charge values ​​accumulated on the photodetector 22, and the assembly formed by the sensor (including the photodetector 22 and the illumination device 28) and the electronic control unit would be designated by the term time-of-flight measuring device 20.

Claims

Claims

1. Method for determining the phase measured by a time-of-flight measuring device (20) comprising the emission by an illumination device (28) of amplitude-modulated radiation with a certain phase, and the measurement of the backscattered radiation (R) on a target (3) by a photodetector (22) whose phase shift of the exposure time window relative to the phase of the emitted radiation (E) takes k possible values and characterized in that: - the method comprises the application of sequential iterations, where the phase shift of the exposure time window successively takes the k possible and distinct values two by two, each of the iterations including: - a step of reading and recording an electric charge value associated with the backscattered radiation (R) towards the photodetector (22), and accumulated during an exposure time window having started with a phase shift relative to the phase of the emitted radiation,- a step of determining a phase of the backscattered radiation (R) from the charge value recorded in the current iteration, and the charge values recorded during the k-1 previous iterations.,

2. A method according to claim 1, wherein the k possible phase shift values of the exposure time window are used successively, in monotonic order.

3. Method according to one of claims 1 to 2, in which each iteration is of equal duration.

4. Method according to one of claims 1 to 3, including in certain iterations, a step of using the phase of the backscattered radiation (R) to calculate a distance between the target (3) and the time-of-flight measuring device (20).

5. A method according to claim 4, wherein the determined distance between the target (3) and the time-of-flight measuring device (20) is used as an input parameter in a remote photoplethysmography algorithm.

6. Method according to one of claims 1 to 5, including in certain iterations, a step of determining an amplitude of the backscattered radiation (R) from the electric charge value recorded in the current iteration, and the charge values recorded during the k-1 previous iterations.

7. A method according to claim 6, wherein in some iterations, the phase and amplitude of the backscattered radiation (R) determined in the current iteration are used to determine the reflectance of the target (3).

8. A method according to claim 7, wherein the reflectance of the determined target (3) is used as an input parameter in a remote photoplethysmography algorithm.

9. Method according to one of claims 1 to 8, in which, in certain iterations, a motion blur detection step is carried out, using the load value recorded in the current iteration, and load values recorded during the k-1 previous iterations, and causes the cancellation of the following steps if an anomaly is detected.

10. Time-of-flight measuring device (20) comprising an illumination device (28), a photodetector (22) whose phase shift of the exposure time window relative to the phase of the emitted radiation (E) takes k possible values, a storage unit (26) and a calculation unit (24), configured to: - control the emission by the illumination device (28) continuously of amplitude-modulated radiation, - sequentially carry out the following iterations using respectively the k possible values for the phase shift: - accumulate charges associated with the backscattered radiation (R) by a target (3) on the photodetector (22) during an exposure time window having started with a phase shift relative to the phase of the emitted radiation, - record the value of charges accumulated on the photodetector (22) in the storage unit (26),- calculating using the calculation unit (24) and the memory unit (26) the phase of the backscattered radiation (R) from the value of accumulated charges recorded in the current iteration and the k-1 previous iterations.,

11. A time-of-flight measuring device (20) according to claim 10, wherein said time-of-flight measuring device (20) is configured to, in certain iterations, calculate a distance between the target (3) and said time-of-flight measuring device (20) using the phase of the backscattered radiation (R) determined in the current iteration.

12. Time-of-flight measuring device, according to one of claims 10 and 11, configured to, in certain iterations, calculate an amplitude of the backscattered radiation (R).

13. A time-of-flight measuring device (20) according to claim 12, configured to, in certain iterations, use the phase and amplitude of the backscattered radiation R to calculate the reflectance of the target (3).

14. Time-of-flight measuring device (20) according to one of claims 10 to 13, in which a motion blur detection is carried out in certain iterations, said current iteration being interrupted, to continue at the following iteration with a new phase shift value of the exposure time window of the photodetector (22) if the detection is positive.

15. A remote photoplethysmography system (200) including a time-of-flight measuring device (20) according to one of claims 10 to 14.

16. A remote photoplethysmography system (200) according to claim 15, wherein said remote photoplethysmography system (200) is adapted to evaluate the heart rate, or variations in the heart rate, or the respiratory rate as a function of the phase of the backscattered radiation (R).

Images