Lidar imaging system including fmcw heterodyne detection with reference signal phase correction device

Abstract

The invention relates to a FMCW type LIDAR imaging system comprising a light source (10), an optical projection device (20), an optical transmission device (30), an optical imaging device (40), and a matrix light detector (50). It further comprises a phase correction device (60) comprising a spatial phase modulator (61) for applying a corrected spatial phase distribution to a reference signal, and a calculation unit (62) for determining said corrected spatial phase distribution by taking into account a spatial distribution representative of a spatial intensity distribution of a backscattered object signal, so that said reference signal has a corrected spatial intensity distribution in the receiving plane, which optimizes a spatial distribution of a parameter of interest representative of a heterodyne signal.

Description

Technical Field

[0001] The field of this invention is the field of LIDAR imaging systems with frequency modulated continuous wave (FMCW) heterodyne detection. Background Technology

[0002] LiDAR imaging systems with FMCW-type heterodyne detection achieve the acquisition of distance maps of illuminated scenes. For this purpose, Aflatouni et al.’s paper entitled Nanophotonic coherent imager, Otp. Express 23(4), 5117-5125 (2015) describes an imaging system of this type that is suitable for instantaneous illumination of the entire scene and determination of distance maps of the illuminated scene.

[0003] Such imaging systems are based on the principle of heterodyne detection, which utilizes the heterodyne signal formed by the interaction between a reference signal and a signal backscattered from the scene. These two optical signals are coherent and originate from the same optical signal transmitted from the optical source, known as the principal signal.

[0004] More specifically, such imaging systems include:

[0005] The optical source of the coherent, continuous, and frequency-modulated signal, which is called the main signal;

[0006] о An optical device used to project the main signal, called the object signal, onto a scene to be illuminated in real time;

[0007] о An optical device used to transmit the main signal, called the reference signal, to a matrix optical detector;

[0008] An optical imaging device suitable for receiving a portion of the object signal backscattered by the illuminated scene and transmitting it to form an image of the illuminated scene on a matrix photodetector;

[0009] A matrix photodetector suitable for receiving backscattered object signals and reference signals that are coherent with each other to form a heterodyne signal.

[0010] The primary signal exhibits an instantaneous frequency change with a band B having, for example, a starting frequency f0 and a period T (called linear frequency modulation). A linear frequency modulation signal is a sine wave whose instantaneous frequency changes linearly with time in this case.

[0011] The matrix photodetector receives the backscattered object signal, which is a decayed and delayed response of the object signal with a delay τ. The delay is caused by the frequency difference f between two signals in the interval [τ; T] (where T >> τ), i.e., between the reference signal and the backscattered object signal. b To express.

[0012] The frequency f, called the difference frequency b This difference in frequency f is equal to the difference between the frequencies of the object signal (and therefore the reference signal) and the backscattered object signal. Therefore, the frequency difference f can be used as a basis for... b Determine the distance D between the illuminated scene and the matrix light detector.

[0013] LIDAR imaging systems also include, for example, those known in documents US 2015 / 177380 A1 and US 2020 / 256995 A1.

[0014] However, the spatial intensity distribution of the backscattered object signal may exhibit laser graininess or speckle grains, thus forming a speckle pattern. Therefore, it is desirable to improve the performance of the imaging system. Summary of the Invention

[0015] The object of this invention is to overcome at least some of the shortcomings of the prior art, and more specifically, to propose a LiDAR imaging system with FMCW-type heterodyne detection, whose performance is improved, even when the backscattered object signal has a speckle pattern. This performance capability enables the heterodyne signal detected by the detected pixel to have better quality in terms of, for example, the intensity of the heterodyne signal (DC and AC components), the intensity of the AC component of the heterodyne signal only, and even its signal-to-noise ratio.

[0016] For this purpose, the subject of the present invention is an FMCW type LIDAR imaging system comprising:

[0017] - A coherent light source suitable for transmitting the main signal of continuous frequency modulation;

[0018] - An optical device used to project the main signal, called the object signal, onto the scene to be illuminated in real time;

[0019] - An optical device used to transmit a portion of the main signal, called a reference signal, to an optical imaging device;

[0020] - An optical imaging device that is suitable for receiving the speckle pattern portion of the object signal backscattered by the scene, referred to as the backscattered object signal.

[0021] - A matrix photodetector comprising a matrix of detection pixels, each of which is designed to receive, in a receiving plane conjugate with the scene via an optical imaging system, a backscattered object signal and a reference signal forming a heterodyne signal having a frequency called the difference frequency, representing the distance between the matrix photodetector and the illuminated scene.

[0022] According to the present invention, the imaging system includes a phase correction device, which comprises:

[0023] - A spatial phase modulator, which is positioned upstream of the optical imaging device on the path of the reference signal, is adapted to apply a spatial phase distribution, known as a modified spatial phase distribution, to the reference signal;

[0024] - A computing unit, connected to a matrix optical detector and a spatial phase modulator, and adapted to:

[0025] -Based on the first spatial distribution of the spatial intensity distribution of the representative backscattered object signal in the receiving plane detected by the matrix photodetector,

[0026] - Determine the optimized, corrected spatial phase distribution to be applied to the reference signal by the spatial phase modulator.

[0027] - A spatial intensity distribution called a modified spatial intensity distribution is used to optimize the spatial distribution of parameters representing the heterodyne signal in the receiving plane, which is selected from the intensity of the heterodyne signal, the intensity of the AC component of the heterodyne signal, or the signal-to-noise ratio.

[0028] Some preferred, but not limiting, aspects of this imaging system are as follows.

[0029] The detection pixel can have a lateral dimension of less than 2×λ×NO, where λ is the wavelength of the backscattered object signal and NO is the number of apertures of the optical imaging device.

[0030] Optical projection and transmission devices may include free-space optical elements.

[0031] A spatial phase modulator can be a liquid crystal phase modulator.

[0032] Optical projection devices may include guiding optical elements implemented in a functionalized substrate, the functionalized substrate including a diffraction grating matrix adapted to receive a primary signal via a waveguide and project an object signal outside the functionalized substrate.

[0033] The phase correction device may include guiding optical elements implemented in the functionalized substrate, and the spatial phase modulator includes a diffraction grating matrix adapted to receive a reference signal via a waveguide and project it onto the exterior of the functionalized substrate after applying a corrected spatial phase distribution to it.

[0034] Optical transmission devices may include free-space optical elements adapted to transmit a reference signal projected by a spatial phase modulator to an optical imaging device.

[0035] The detection pixel matrix is ​​implemented in or on the functionalized substrate.

[0036] Optical transmission and imaging devices can be adapted to transmit reference signals and backscattered object signals along the same optical axis using a matrix photodetector.

[0037] Optical imaging devices may include an aperture stop and at least one free-space optical element, thereby defining a field of view for a matrix light detector and a central region laterally defined by light rays propagating to the edges of the backscattered object signal in the detection pixels, referred to as end pixels, located at the edges of the detection matrix.

[0038] Optical transmission and imaging devices can be adapted to form an image of a reference signal in an intermediate plane orthogonal to the optical axis of the optical imaging device, thereby forming an equivalent light source for the reference signal. This equivalent light source can be contained in the central region of light rays at the edges of the backscattered object signal without vignetting. The equivalent light source can have an emission angle at each point relative to the reference signal that is at least equal to the field of view of the optical imaging device.

[0039] The equivalent light source may have a lateral dimension that is at least equal to the lateral dimension of the central region of the light source at the edge of the backscattered object signal without vignetting.

[0040] The present invention also relates to a method for determining a distance map of a scene using an imaging system according to any one of the above features, wherein the parameter of interest is the intensity of a heterodyne signal, the method comprising the following steps:

[0041] a / Projecting object signals into the scene using optical projection devices to illuminate the scene in real time;

[0042] b / The first spatial intensity distribution of the incident optical signal representing the backscattered object signal is detected by a matrix photodetector;

[0043] c / The spatial phase distribution to be corrected and applied to the reference signal by the spatial phase modulator is determined by the calculation unit;

[0044] d / The corrected spatial phase distribution is applied to the reference signal through a spatial phase modulator;

[0045] e / The spatial intensity distribution of the heterodyne signal is detected by a matrix photodetector;

[0046] f / Through the computation unit, the spatial distribution of the parameters of interest is determined based on the spatial intensity distribution of the detected heterodyne signal;

[0047] If applicable, repeat steps c / to f / by changing the modified spatial phase distribution until the determination criterion for the spatial distribution as a function of the parameter of interest reaches a predetermined threshold.

[0048] g / When the determination standard reaches a predetermined threshold, the distance map is determined.

[0049] The determination method may include:

[0050] During step b / , the transmission of the reference signal is paused so that the incident optical signal is a backscattered object signal;

[0051] During step c / , a modified spatial phase distribution is determined based on a predetermined transformation function that expresses the spatial intensity distribution of the reference signal in the receiving plane as a function of the spatial phase distribution applied by the spatial phase modulator, such that the spatial intensity distribution of the reference signal in the receiving plane is approximately equal to the spatial intensity distribution of the detected backscattered object signal.

[0052] During step d / , the transmission of the reference signal is no longer paused.

[0053] The determination method may include:

[0054] During step b / , the detected optical signal is a heterodyne signal;

[0055] Repeat steps c / to f / , and change the spatial phase distribution according to the predetermined optimization algorithm and the optimization criteria in the previous iteration until the optimization criteria reach the predetermined threshold, thereby optimizing the spatial distribution of the parameters of interest. Attached Figure Description

[0056] Other aspects, objects, advantages, and features of the invention will become clearer from the following detailed description of preferred embodiments of the invention, provided as non-limiting examples with reference to the accompanying drawings, in which:

[0057] Figure 1A and 1B In free space optical construction ( Figure 1A ) and in guiding optical structure ( Figure 1B Here are two schematic partial views of an imaging system that implements a distance map of the entire illuminated scene;

[0058] Figure 2A and 2B yes Figure 1A Or, as shown in 1B, some of the detection pixels of the imaging system are presented as a cross-section. Figure 2A ) and as a top view ( Figure 2B The view shows the light field of the reference signal and the light field of the backscattered object signal with a speckle pattern.

[0059] Figure 3A and 3B The imaging system according to different embodiments is in free-space optical construction ( Figure 3A ) and in guiding optical construction ( Figure 3B A schematic partial view of the imaging system, wherein the imaging system includes a device for correcting the phase of a reference signal;

[0060] Figure 4AThe diagram is schematically shown partially as a sectional view and as a top view, based on... Figure 3B A portion of the imaging system with guiding optical structure. Figure 4B A partial schematic view shown as a top view Figure 4A The spatial phase modulator shown in the figure;

[0061] Figure 5A and 5B It is based on Figure 3A and 3B A cross-sectional view of some detection pixels of the imaging system in an embodiment ( Figure 5A ) and as a top view ( Figure 5B The view shows the optical field of a reference signal with a corrected spatial phase distribution and the optical field of a backscattered object signal with a speckle pattern.

[0062] Figure 6A and 6B This is a flowchart illustrating the steps of two methods for determining a distance map of a scene using an imaging system according to one embodiment;

[0063] Figures 7A to 7D This is a schematic partial view of the optical imaging device and the detection pixel matrix, showing:

[0064] о Figure 7A The diagram shows the end rays of the backscattered object signal and the end rays of the equivalent light source of the reference signal for two opposite detection pixels located on the edge of the detection matrix.

[0065] о Figure 7B The light rays at the edge of the non-vignetting region of the backscattered object signal are shown, highlighting the central region near the aperture stop.

[0066] о Figure 7C The diagram shows the end rays of the backscattered object signal and the end rays of the equivalent light source for the reference signal located upstream of the aperture stop, for a detection pixel located at the edge of the detection matrix.

[0067] о Figure 7D Showing and such Figure 7C The same light ray is shown in the example, in which the equivalent light source is located in the plane of the aperture stop. Detailed Implementation

[0068] In the accompanying drawings and throughout the remainder of this specification, the same reference numerals represent the same or similar elements. Furthermore, for clarity of the drawings, individual elements are not shown. Moreover, the various embodiments and variations are not mutually exclusive but can be combined together. Unless otherwise stated, the terms "approximately," "about," and "about" refer to a difference of up to 10%, preferably up to 5%. Furthermore, the term "range from..." and equivalent terms refer to the limit, unless otherwise stated.

[0069] This invention relates to a LiDAR imaging system with heterodyne detection of the frequency modulated continuous wave (FMCW) type. It can be fabricated using free-space optics and / or guiding optics.

[0070] This imaging system is called a "LIDAR" (Light Detection and Ranging) system because it is suitable for providing a map (or distance image) of the distance to a scene illuminated by coherent optical signals. Within the scope of this invention, the imaging system uses continuous optical signals to illuminate the entire scene in real time and simultaneously acquires images of the illuminated scene to determine distances. The imaging system according to the invention thus differs from LIDAR systems (Time-of-Flight (TOF) measurement type or even FMCW type) in which the scene is "scanned" by optical signals, i.e., spatially scanned.

[0071] Furthermore, the imaging system is called "heterodyne detection" because, in order to determine the distance to the illuminated scene, the frequency of the heterodyne signal, called the difference frequency, is determined by the interaction between the reference signal from the local resonator and the signal backscattered from the illuminated scene. These two optical signals are coherent with each other. Both the reference signal and the signal projected onto the scene actually originate from the same principal optical signal emitted by the optical source. Heterodyne detection is of the FMCW type because the principal signal is a continuous and frequency-modulated signal.

[0072] Figure 1A and 1B This is a schematic partial view of a LiDAR imaging system 1 with FMCW-type heterodyne detection, suitable for determining the distance image of a scene instantaneously illuminated by a continuous and frequency-modulated optical signal. The imaging system 1 in this case has a free-space configuration ( Figure 1A ) or guiding optical structure ( Figure 1B In this case, given the illustrative examples provided, it should be understood that combinations and variations of these constructions are possible.

[0073] The imaging system can be of the global shutter type, in which all detected pixels are read simultaneously. The imaging system can also provide an instantaneous distance image. As a variation, it can be of the rolling shutter type, thus involving the sequential reading of detected pixels, such as row-by-row, where all detected pixels in the same row are read simultaneously.

[0074] Generally, imaging system 1 includes:

[0075] о refers to the optical source 10 of the coherent, continuous, and frequency-modulated signals that are the main signals;

[0076] о An optical device 20 used to project the main signal, called the object signal, onto a scene to be illuminated in real time;

[0077] The optical device 30, which is used to transmit the main signal to the matrix photodetector 50, is advantageously adapted to form an equivalent light source for the reference signal located in or near a predetermined area at the aperture stop of the optical imaging device 40.

[0078] An optical imaging device 40 adapted to receive the portion of the object signal that is backscattered by the illuminated scene and transmit it so as to form an image of the illuminated scene on the receiving plane of the matrix photodetector 50 and transmit a reference signal.

[0079] The O-matrix photodetector 50 includes a matrix of detection pixels having indices i and j, which is designed to receive backscattered object signals and reference signals that interact to form heterodyne signals, wherein each detection pixel is advantageously adapted to determine the distance to the illuminated scene based on the difference frequency of the detected heterodyne signals.

[0080] Imaging system 1 includes an optical source 10 for a coherent, continuous, and frequency-modulated signal, referred to as the primary signal. Optical source 10 is preferably a laser source that emits the continuous primary signal. Exemplarily, the primary signal may have an optical frequency located in the infrared domain.

[0081] Furthermore, the primary signal, for example in this example, is linearly frequency modulated based on a starting frequency f0 during a repetition period T with a bandwidth of B. In this example, the signal is a linearly frequency modulated signal, i.e., a sine wave, whose instantaneous frequency changes linearly with time. Exemplarily, the primary signal can be expressed as follows:

[0082]

[0083] Optical source 10 has a coherence length that is typically greater than the difference in the optical path between the reference channel and the object channel. The reference channel is the path followed by the reference signal between optical source 10 and matrix photodetector 50. The object channel is the path followed by the object signal from optical source 10 to the scene and the path followed by the object signal backscattered by the scene to matrix photodetector 50. This difference in the optical path may initially correspond to twice the maximum distance between imaging system 1 and the scene.

[0084] The optical source 10 may thus include a laser source 11 of the type of vertical cavity surface-emitting laser (VCSEL), which generally has a coherence length of about one meter in the case of emission in the near-infrared domain (0.7 to 2 μm), or even a laser source of the type of edge-emitting laser (EEL), which may have a coherence length of about ten meters or even one hundred meters.

[0085] Between the optical source 10 and the optical projection device 20, passive optical elements can be located on the optical path of the main signal. Thus, in Figure 1A In the example, the shaping lens 12 can be located at the output of the optical source 10 and can allow the collimated beam to be aligned while simultaneously increasing its lateral dimension to a diameter of, for example, several millimeters. Furthermore, a spatial filtering device may be present to remove high spatial frequencies. Thus, the main signal propagates in a collimated manner, having a Gaussian distribution and a diameter of approximately several millimeters, for example, 5 mm.

[0086] Imaging system 1 includes an optical projection device 20 adapted to project a portion of a main signal onto a scene to be illuminated in real time. "Scene" should be understood as the scene illuminated by the object signal. The portion of the main signal to which this signal is projected is called the object signal. Thus, the entire scene is illuminated in real time by the object signal, which therefore has an angular aperture capable of illuminating the entire scene in real time. Preferably, the scene is also uniformly illuminated, i.e., the object signal has an intensity with a constant angular distribution.

[0087] In this example, the optical projection device 20 or the optical transmission device 30 includes at least one optical element adapted to divide a primary signal into an object signal on one hand and a reference signal on the other. Figure 1A In an example, this could involve, for instance, a semi-reflective strip or a beam-splitter cube. In this case, the beam-splitter cube 31 transmits the portion of the primary signal that becomes the target signal and reflects the portion of the primary signal that becomes the reference signal. The intensity distributions of the target signal and the reference signal are preferably different, thus being 90% for the target signal and 10% for the reference signal. Figure 1B In the case of the same functionalized substrate, the input waveguide coupled to the optical source 10 extends to the first diffraction grating matrix 22 that projects object signals to the scene, and to the second diffraction grating matrix 35 that projects reference signals to the matrix photodetector 50.

[0088] To illuminate the scene instantly and uniformly in this situation, the optical projection device 20 shapes the optical signal. For this purpose, the optical projection device includes at least one optical element 21 adapted to disperse the object signal. The angular aperture of the object signal ensures that the entire scene is uniformly illuminated by the object signal. The angular aperture of the object signal is preferably approximately equal to the field of view of the optical imaging device 40 (e.g., approximately 20°). Furthermore, it may include at least one optical element adapted to maintain a constant angular distribution of intensity, thereby achieving a flat distribution of intensity, rather than a Gaussian distribution. Of course, the individual optical elements may be separate or overlapping.

[0089] exist Figure 1A and 1B In this example, a diffuser 21 (for transmission) is positioned at the output of the beam-splitting cube 31 or the first diffraction grating matrix 22 to transform the intensity distribution of the object signal from a Gaussian distribution to a flat distribution. Furthermore, in this case, the same diffuser 21 disperses the object signal. A lens (not shown) may be located downstream (or upstream) of the diffuser to give the angular aperture a predetermined value, in this case, approximately equal to the field of view of the optical imaging device 40.

[0090] The imaging system 1 also includes an optical transmission device 30 adapted to transmit a portion of the main signal to the matrix photodetector 50. This transmitted portion of the main signal is called a reference signal. In the case of heterodyne detection, the transmitted reference signal corresponds to the signal of the local resonator (LO).

[0091] In this example, the optical transmission device 30 is adapted to uniformly illuminate the matrix detector 50 with the reference signal. For this purpose, in Figure 1A In the example, in this case, it includes a diffuser 32 located downstream of the beam-splitting cube 31, adapted to make the angular distribution of the intensity constant, thereby obtaining a flat distribution of the intensity of the reference signal, no longer a Gaussian distribution. Figure 1B In the example, the second diffraction grating matrix 35 is constructed such that the spatial intensity distribution of the reference signal is uniform. Of course, the spatial intensity distribution of the reference signal does not have to be uniform; it can have different distributions, such as a Gaussian distribution.

[0092] Furthermore, the optical transmission device 30 is adapted to ensure that all detection pixels of the matrix photodetector 50 receive the reference signal. This is related to the fact that all detection pixels also receive the backscattered object signal. Additionally, a heterodyne signal is formed at each detection pixel, enabling the distance image to be determined based on the entire detection pixel matrix. Figure 1AIn the example, the optical transmission device 30 includes at least one converging lens 33 disposed downstream of the diffuser 32, which shapes the beam of the reference signal into a predetermined angular aperture. This is pre-defined, taking into account the various optics of the optical imaging device 40, so that the reference signal effectively illuminates each detection pixel. As a variation, the optical transmission device 30 may not include a lens that illuminates all detection pixels. The diffuser 32 simply needs to be properly arranged, taking into account its diffusion cone, the diameter of the incident reference signal on the diffuser 32, the aperture of the optical imaging device 40, and the size of the detection matrix of the photodetector 50. Figure 1B In the example case, the diffraction grating of the second matrix 35 is designed such that it has an angular aperture that illuminates all the detection pixels.

[0093] Preferably, the receiving planes of the diffuser 32 and the matrix photodetector 50 are not conjugate to prevent the formation of an image of the diffuser 32 on the matrix photodetector 50, which would degrade the quality of the heterodyne signal to be detected. However, the diffuser 32 may be conjugate on a plane called the intermediate plane located at or near the aperture stop of the optical imaging device 40 to form an equivalent light source 36 for the reference signal, such as a reference... Figures 7A to 7D As detailed in the description.

[0094] The imaging system 1 also includes an optical imaging device 40 adapted to receive a portion, called a backscattered object signal, of an object signal backscattered by the scene and to transmit the backscattered object signal to the matrix light detector 50. The optical imaging device is particularly adapted to form an image of the illuminated scene on the receiving plane of the matrix light detector 50. Furthermore, the illuminated scene and the receiving plane of the matrix light detector 50 are conjugate to the closest depth of field.

[0095] Backscattered object signal s or (t) relative to the object signal s o (t) has a delay τ. This allows it to be injected into the backscattered object signal s. or The frequency shift f of the scene in (t) D Consider it as zero (f) D In the case of (=0), it is expressed as:

[0096]

[0097] In this case, the optical imaging device 40 or the optical transmission device 30 includes at least one optical element adapted to combine the reflected and scattered object signal and the reference signal, i.e., to transmit them along the same optical axis to the matrix photodetector 50. Figure 1AIn this example, this could involve a semi-reflective strip or a combined cube 34. In this case, the combined cube 34 reflects the reference signal along the optical axis to the matrix photodetector 50 and transmits the backscattered object signal along the same optical axis. The two optical signals thus propagate towards the matrix photodetector 50 along the same optical axis in a common channel. Consequently, the two optical signals can be superimposed along the same optical axis, improving the quality of the resulting heterodyne signal. Figure 1B In this case, the functionalized substrate 2 and the second diffraction grating matrix 35 are transparent to the backscattered object signal, allowing the backscattered object signal to be transmitted and superimposed on the reference signal. In practice, the functionalized substrate 2 transmits the backscattered object signal without actually interfering with it because the refractive index difference between the diffraction grating and waveguide of the reference signal on one side and the supporting substrate on the other side is small (thus limiting the diffraction of the backscattered object signal), and because the surfaces of the waveguide and diffraction grating are small compared to the size of the incident backscattered object signal.

[0098] The optical imaging device 40 has free-space optical elements and includes at least one lens and an aperture stop 42 defining a physical pupil. It should be noted that the aperture stop 42 may not be a physical object separate from the lens, but may be defined by the profile of the lens. It includes an optical axis orthogonal to the receiving plane of the matrix photodetector 50. In this case, it is adapted to receive the backscattered object signal and a reference signal, allowing them to be transmitted along the same optical axis to the matrix photodetector 50 and thereby allowing better superposition of the two optical signals, thus improving the combination of the two optical signals through interference, which improves the intensity of the heterodyne signal. In this case, the optical imaging device 40 includes a plurality of lenses 41, 43, with the aperture stop 42 arranged between them.

[0099] The imaging system 1 also includes a matrix photodetector 50 having a matrix of detection pixels extending in a receiving plane. The receiving plane of the matrix photodetector 50 is located in a plane conjugate to the scene (to the nearest depth of field) via the optical imaging device 40. In other words, an image of the scene is formed in the receiving plane of the matrix photodetector 50. Each detection pixel is designed to receive a reference signal and a backscattered object signal, which interfere with each other to form a signal called a heterodyne signal. This could involve a photodetector of the CMOS or CCD type.

[0100] heterodyne signal s het (t) originates from the reference signal s ref (t) and the backscattered object signal s or The interaction between (t) includes a constant component (intended to be filtered) and a periodic AC component that forms the useful signal. het (t). This periodic alternating component can be expressed as follows:

[0101]

[0102] Among them, the difference frequency f b It equals Bτ / T, and therefore equals 2BD / cT, where c is the speed of light. Determine the difference frequency f. b Therefore, the distance D between the scene and the matrix photodetector 50 is obtained.

[0103] The detection pixel can be formed by a microlens arranged opposite to the photodiode. For example... Figure 2A and 2B As shown, each detection pixel has a lateral dimension d defined by a microlens. px The photosensitive area of ​​a photodiode has a lateral dimension d. pd In this case, the detection plane is considered to be the plane in which the microlens extends.

[0104] The detection pixels are advantageously suited for determining the difference frequency of the detected heterodyne signal. For this purpose, each pixel includes, for example, a pn, pin, or avalanche type photodiode and a photoelectric conversion device to convert the detected heterodyne signal into an electrical signal, such as a voltage. The conversion device may include a filter for the detected heterodyne signal. het DC component of (t) het And only the AC component is retained. het The system includes a filter, an amplifier for the amplitude of the filtered heterodyne signal, a detector for the periodic variation of the filtered heterodyne signal, and a counter, thereby determining the value of the difference frequency of the heterodyne signal. The frequency value determined by each detected pixel can then be transmitted to a calculation unit, which derives a distance map from it. The difference frequency can be determined simultaneously for all detected pixels to obtain an instantaneous distance map (distance image) of the scene subsequently illuminated (global shutter mode). As mentioned earlier, the imaging system 1 may also have a rolling shutter operation.

[0105] Therefore, during operation, optical source 10 emits a coherent, continuous, and frequency-modulated main signal, a portion of which (the object signal) is projected onto the scene by optical projection device 20 to illuminate the scene in real time. Optical transmission device 30 transmits a portion of the main signal (the reference signal) to matrix photodetector 50. Optical imaging device 40 receives the backscattered object signal and forms an image of the illuminated scene on the detection pixel matrix. In this case, it also receives the reference signal transmitted to the detection pixel matrix. The reference signal s ref (t) and the backscattered object signal s or (t) phase interference and form a heterodyne signal s at each detection pixel. het (t). The difference frequency f of the detected heterodyne signal is then determined for each detected pixel. b Then determine the distance map of the illuminated scene.

[0106] Figure 2A and 2B The detection pixel P of the matrix light detector 50 is schematically shown. ij Reference signal s at some points in the middle ref (t) and the backscattered object signal s or Example of the light field (spatial intensity distribution) of (t). Figure 2A It is a cross-sectional view. Figure 2B This is a top view taken when the lateral dimension of the speckle particles is larger than the lateral dimension of the detection pixel. Of course, without departing from the scope of this invention, the lateral dimension of the speckle particles may be smaller than the lateral dimension of the detection pixel.

[0107] Here, in Figure 2A Several detection pixels P are shown in the figure. ij P i+1,j P i+2,j (and Figure 2B P in ij P i+1,j P i,j+1 Each of them includes a microlens 52 disposed opposite to the photodiode 51. In this case, the microlenses 52 are adjacent and have a lateral dimension d. px (In this case, it defines the lateral dimension of the detection pixel), and the photodiode 51 (the effective detection area of ​​the incident photon) has a lateral dimension d. pd In this case, the dimension is smaller than the dimension d. px The receiving plane of the matrix photodetector 50 is considered to be the plane corresponding to the matrix of the microlens 52.

[0108] Reference signal s ref (t) can have a light field that instantly and uniformly illuminates the entire detection matrix and thus the entire surface of each detected pixel. However, the backscattered object signal s or (t) can have a speckle pattern, that is, it shows the spatial intensity distribution of the laser grain size, referred to as speckle particles. or (t). This speckle pattern originates from the scene being illuminated by coherent light from scattering objects. The backscattered optical signal s or The individual rays of (t) thus interfere with each other, thereby forming a speckle pattern composed of several speckle particles (bright areas) surrounded by dark areas. The speckle particles are randomly phase-shifted relative to each other. More specifically, the backscattered object signal s or (t) have the same frequency, but have a non-uniform spatial phase distribution: the phase is approximately constant in each speckle particle, but randomly different between different speckle particles.

[0109] The speckle particles have an average lateral dimension d of approximately 2×λ×NO. sp, where λ is the backscattered object signal s or The wavelength of (t) (e.g., in the visible or near-infrared domain, equal to the wavelengths of the primary and reference signals), and NO is the number of apertures of the optical imaging device 40, NO = f di / d do In this case, f di It is focal length and d do This is the lateral dimension (e.g., diameter) of the diameter of the input or output pupil of the optical imaging device 40. Therefore, it depends on the number of apertures NO and the lateral dimension d. px speckled particles can be like Figure 2B The speckle pattern extends over one or more detection pixels, or extends less than one detection pixel. Advantageously, the speckle particles are larger than the detection pixels to improve the detection of heterodyne signals and prevent interference from multiple speckle particles in the same detection pixel on the difference frequency f. b The determination is due to the phase shift between these speckle particles. Moreover, the number of spaces can be large, for example, at least 2, and the size of the detection pixel can be up to 10 μm.

[0110] It is important to note that the optical field of the reference signal, whose optical intensity is spatially uniform at time t, differs significantly from the optical field of the backscattered object signal, as characterized by the speckle pattern. It should also be noted that in this case, these patterns are schematic to enhance clarity. Thus, the size of the speckle particles does not decrease with transmission through the microlens 52 because it involves diffractive optics rather than geometric optics. In other words, the speckle particles have approximately the same size regardless of whether the plane of the microlens 52 or the plane of the photodiode 51 is considered.

[0111] The detection pixel P is now limited to the receiving plane. ij The backscattered object signal s in or Spatial distribution S of speckle particles in (t) ij or (t)(Geometric Distribution). When the considered detection pixel is located in the dark region of the speckle pattern, a local value of zero or near zero is assumed. Furthermore, it has a non-zero value when speckle particles are present in the considered detection pixel. Similarly, the spatial distribution S of the bright region defining the reference signal in the receiving plane is... ij ref (t). In Figure 2A and 2B In this case, the spatial distribution S ij ref (t) is uniform and has the same non-zero value at each detected pixel in the receiving plane.

[0112] However, it appears that the backscattered object signal s orSpatial distribution S of speckle particles in (t) ij or (t) and reference signal s ref Spatial distribution of the bright region S of (t) ij ref The difference between (t) can affect the performance capability of the imaging system 1, especially the quality of the heterodyne signal detected at each detected pixel in terms of, for example, the detected intensity (DC and AC components), the intensity of the AC component only, and / or the signal-to-noise ratio.

[0113] In practice, in heterodyne detection, the reference signal allows for amplification of the backscattered object signal. Heterodyne signal s het The useful part of (t), that is, for the detected pixel P ij AC component ij het The strength of (t) is approximately Among them, E ij =F ij / S ij It is local lighting, F ij The incident signal at the detection pixel P ij Illuminated surface S ij The luminous flux (number of photons) on the surface.

[0114] However, if, for example, we consider Figure 2B The detection pixel P shown in the figure ij P i+1,j P i,j+1 P i+1,j+1 Its area is approximately (2×λ×NO). 2 In this case, speckle particles illuminate the surface S. ij or (t), S i+1,j or (t), S i,j+1 or (t) and S i+1,j+1 or (t), each of which is approximately the area (d) of each detected pixel. px ) 2 One-quarter. In contrast, the reference signal s ref (t) Illuminate the entire surface of the detection pixel so that surface S ij ref (t), S i+1,j ref (t), S i,j+1 ref (t) and S i+1,j+1 ref (t) are each equal to (d) px ) 2Therefore, the reference signal s is visible. ref (t) Uniform illumination of the detection pixel is provided by the AC component of the heterodyne signal. ij het The suboptimal intensity of (t) is expressed as this, because the spatial distribution S ij ref (t) is uniform, and the spatial distribution of speckle particles S is not considered. ij or (t). Therefore, this also affects the intensity I. ij het (t), and also affects what can be defined as AC ij het (t) / √(AC ij het (t)+DC ij het The signal-to-noise ratio (SNR) of (t) ij het (t). In fact, it appears that a portion of the reference signal was unintentionally used to amplify the dark regions of the backscattered object signal.

[0115] Furthermore, within the scope of this invention, it is intended to take into account the spatial intensity distribution I of the backscattered object signal. ij or (t), changing the spatial intensity distribution of the reference signal I ij ref (t) is used to optimize the spatial distribution of the parameters of interest representing the heterodyne signal, such as its intensity (AC and DC components), the intensity of its AC component only, or even the signal-to-noise ratio.

[0116] Spatial intensity distribution of the reference signal I ij ref The change in (t) results in the formation of a bright region surrounded by a dark region, the arrangement and shape of which depend on the spatial intensity distribution I of the backscattered object signal. ij or (t). The equivalent "speckle pattern" of the reference signal is therefore associated with the effective speckle pattern of the object signal being displayed in reverse. It can be associated with the spatial variation of the reflectivity of the illuminated scene. Thus, the surface S illuminated by the bright area of ​​the reference signal... ij ref (t) can be approximately equal to the surface S illuminated by the speckle particles of the backscattered object signal. ij or (t), and can be at least partially superimposed. By reducing the illuminated area S for all or some of the detected pixels. ij ref (t), the formation of these bright regions thus achieves the optimization of the spatial distribution of the selected parameters of interest.

[0117] For this purpose, the imaging system 1 according to the present invention includes a phase correction device 60, which comprises:

[0118] A spatial phase modulator 61 (i.e., SLM, spatial light modulator) is arranged upstream of the optical imaging device 40 on the path of the reference signal and is adapted to apply a spatial phase distribution called a correction to the reference signal. Spatial phase distribution;

[0119] The computing unit 62, which is connected to the photodetector and the spatial phase modulator 61, is adapted to:

[0120] Based on the spatial intensity distribution of the backscattered object signal in the receiving plane detected by the representative matrix photodetector 50 ij or (t i The first spatial distribution of )

[0121] The optimized, corrected spatial phase distribution to be applied to the reference signal by the spatial phase modulator 61 is determined.

[0122] This ensures that the reference signal has a spatial intensity distribution I in the receiving plane, which is called the correction. ij ref,opt (t i The spatial intensity distribution of the heterodyne signal is optimized to represent the spatial distribution of the parameters of interest, which are selected from the intensity I of the heterodyne signal. ij het (t i (DC and AC components), intensity of AC component only ij het (t i (or signal-to-noise ratio SNR) ij het (t i ).

[0123] In other words, the first spatial distribution is determined by considering the spatial distribution of the backscattered object signal in the receiving plane. This may involve the spatial intensity distribution I of the backscattered object signal itself. ij or (t), or (in the case where only the backscattered object signal is not detected) the spatial intensity distribution I of the heterodyne signal. ij het (t). This can also involve the AC component. ij (t i Spatial distribution or signal-to-noise ratio (SNR) ij (t i The spatial distribution of ) or even the difference frequency f bThe spatial distribution. Thus, this first spatial distribution takes into account the speckle pattern of the backscattered object signal.

[0124] The subsequent aim is to optimize the spatial distribution of the parameters of interest associated with the heterodyne signal. This could involve the intensity of the heterodyne signal (DC and AC components), the intensity of the AC component of the heterodyne signal, or even the signal-to-noise ratio.

[0125] For this purpose, the spatial phase distribution of the reference signal is altered according to a first spatial intensity distribution (which takes into account the speckle pattern of the backscattered object signal), thereby transforming it from an initial spatial distribution (e.g., uniform) to a modified spatial distribution (non-uniform) to optimize the spatial distribution of the parameter of interest. This results in the spatial intensity distribution of the reference signal tending to the spatial distribution of the speckle pattern of the backscattered object signal, especially when the parameter of interest is the intensity (DC and AC components) of the heterodyne signal. In other words, the reference signal will have bright regions whose geometric distribution is correlated with the geometric distribution of the speckle grains of the backscattered object signal. The spatial distribution of the parameter of interest is thus optimized, which improves the performance capability of the imaging system 1.

[0126] Figure 3A and 3B This is a schematic partial view of an imaging system 1 according to various embodiments, wherein it includes a free-space optical structure ( Figure 3A ) and guiding optical structure ( Figure 3B Such a phase correction device 60 in )

[0127] Figure 3A An imaging system 1 according to one embodiment is described in free-space optical construction. It is related to… Figure 1A The difference described herein is roughly that it includes, in this example, a spatial phase modulator 61 positioned on the path of the reference signal between the beam-splitting cube 31 and the combined cube 34, and a computing unit 62 connected on one side to the matrix photodetector 50 and on the other side to the spatial phase modulator 61. In this example, the spatial phase modulator 61 is separate from the diffuser 32 and positioned downstream of the diffuser. As a variation, the spatial phase modulator may be located upstream of the diffuser 32, or even adjacent to or overlapping with the diffuser. Preferably, as detailed below, the spatial phase modulator 61 is located in a plane conjugate to the intermediate plane in the central region defined by the light rays located at the edge of the non-vignetting domain of the backscattered object signal, to form the equivalent light source 36 of the reference signal (see Figures 7A to 7D This improves the superposition of the wavefront of the reference signal and the wavefront of the backscattered object signal. This, in turn, limits the generation of interference fringes between the two incident signals, which could reduce the detection capability of the detection pixel for the heterodyne signal.

[0128] The spatial phase modulator 61 can be a liquid crystal projection modulator. It can include a modulation pixel matrix with indices m and n. The number of modulation pixels can be less than or equal to one-tenth of the number of detection pixels. Thus, it can be an electrically controlled spatial light modulator (ESLM), in which liquid crystal is disposed between two sheets of transparent material. The modulation pixels are controlled by transparent microelectronic elements, such as thin-film transistors (TFTs).

[0129] The computing unit 62 is connected to the matrix photodetector 50 and the spatial phase modulator 61. The computing unit includes a programmable processor capable of executing instructions stored on an information storage medium. The computing unit also includes components for determining the spatial intensity distribution I of a reference signal to be applied to the detection plane. ij ref Corrected spatial phase distribution And a memory necessary for optimizing the spatial distribution of the parameters of interest representing the heterodyne signal. The memory is also adapted to store computational information. Figure 6A and 6B Two examples of methods for determining a distance map of a scene using an imaging system 1 according to one embodiment are shown, and these maps are described below.

[0130] Figure 3B This describes an imaging system 1 according to another embodiment in a guiding optical configuration. It is related to... Figure 1B The main difference lies in that the first diffraction grating matrix 22, the spatial phase modulator 61, and the matrix photodetector 50 are implemented in or on the same functionalized substrate 2 made of, for example, a silicon-based semiconductor material. However, both the optical transmission device 30 and the optical imaging device 40 include free-space optical elements. The matrix photodetector 50 can thus be a photodetector implemented in the functionalized substrate 2, such as... Figure 3B As shown. In this case, this is a simplified illustration for clarity: in reality, waveguides are typically very thin, about hundreds of nanometers, while photodiodes are implemented in semiconductor layers about a few micrometers thick. As a variation, the matrix photodetector 50 can be fabricated on a substrate other than substrate 2 and then transferred thereon. Again, as a variation, it can obviously be implemented in a functionalized substrate that is different from and separate from substrate 2.

[0131] Optical source 10 transmits a primary signal into the waveguide of functionalized substrate 2. This primary signal propagates to a first diffraction grating matrix 22, which projects the primary signal onto diffuser 21. Another portion of the primary signal is transmitted to spatial phase modulator 61. This can be formed by projection grating groups, each associated with a phase shifter. A reference signal is extracted from functionalized substrate 2 by deflecting mirror 37, which redirects the reference signal to lens 33 and subsequently a combined cubic 34. The reference signal and the backscattered object signal are then transmitted by optical imaging device 40 to a photodiode matrix implemented on the same functionalized substrate 2. Matrix photodetector 50 is connected to computing unit 62, which is also connected to spatial phase modulator 61.

[0132] Figure 4A yes Figure 3B A detailed view of an embodiment of such a functionalized substrate 2 for the imaging system 1 is shown. It is shown in longitudinal cross-section and top view. As shown in the top view, the optical projection device 20 includes a waveguide in which the main signal propagates. A beam splitter divides the main signal into multiple arms, each of which includes an object signal for projecting onto the scene (e.g., such as...). Figure 4A As shown, several diffraction gratings (through at least one shaping lens).

[0133] The other arm is the part that transmits the main signal to the spatial phase modulator 61. Figure 4B An example of such a guided optical modulator 61 is shown. It may be formed of multiple arms, each arm comprising several diffraction gratings 63. These project a reference signal to the free-space optical elements of the optical transmission device 30. Each diffraction grating 63 is associated with a basic optical phase shifter 64. This basic optical phase shifter is conventionally of the electrorefractive effect or thermo-optical effect type. In both cases, the phase change is obtained by changing the refractive index of the material forming the core of the waveguide under consideration. This change in refractive index can be obtained by altering the density of free carriers in the case of an electrorefractive phase shifter, or by altering the temperature applied to the arm in the case of a thermo-optical phase shifter.

[0134] Figure 5A and 5B This schematically illustrates that, in the case of an imaging system 1 according to one embodiment, at some locations in the detection pixels of the matrix light detector 50, the reference signal s ref (t) and the backscattered object signal s or Example of the light field (spatial intensity distribution) of (t). In this case, the spatial phase modulator 61 will optimize the corrected spatial phase distribution. An application is made to a reference signal such that the spatial intensity distribution I of the reference signal is... ij ref (t) is correlated with the spatial intensity distribution of the backscattered object signal, thereby optimizing the spatial distribution of the parameters of interest that represent the heterodyne signal.

[0135] As in Figure 2A and 2B In the backscattered object signal, there is a speckle pattern, and the speckle particles have a lateral dimension d of approximately 2×λ×NO in the receiving plane. sp Their dimensions d sp It can be larger than the detection pixel d px The size of each detection pixel P can be smaller or equal in size. ij On surface S ij or The object signal is illuminated by backscattering on (t).

[0136] However, the reference signal no longer has a uniform spatial intensity distribution I ij ref (t), such as Figure 2A and 2B As shown, instead of a uniform spatial intensity distribution, it exhibits bright regions surrounded by dark areas. This equivalent "speck pattern" of the reference signal originates from the spatial intensity distribution I of the backscattered object signal applied by the spatial phase modulator 61. ij or Optimized modified spatial phase distribution of (t) Therefore, the bright regions of the reference signal have a geometric distribution that is more or less similar to the geometric distribution of the speckle particles: their size d zb ref Approximately the size d of the speckle particles sp Furthermore, they are at least partially superimposed on the speckle particles.

[0137] Furthermore, as mentioned above, the area S illuminated by the bright region of the reference signal ij ref Decreasing (t) helps increase the strength I of the heterodyne signal. ij het (DC and AC components), improve the strength of the AC component. ij het (t), and improve signal-to-noise ratio (SNR) ij het (t). The performance of imaging system 1 is thus improved.

[0138] Figure 6A This illustrates a distance map D for determining a scene using an imaging system 1 according to one embodiment. ij (t i The flowchart shows the steps of the method. Of course, the steps are separated for clarity, but several of them are performed simultaneously or actually simultaneously.

[0139] Based on the corrected spatial phase distribution to be applied to the reference signal Spatial intensity distribution of the reference signal in the receiving plane I ij ref The associated transformation function H directly optimizes the spatial intensity distribution I of the reference signal. ij ref (t i Based on this, the determination method in this situation is implemented. Furthermore, in Figure 6A The method for determining this involves focusing on the intensity of the heterodyne signal, and aims to make the spatial intensity distribution I of the reference signal... ij ref (t i Spatial intensity distribution of the backscattered object signal I ij or (t i Similar to ), this method is particularly suitable for speckle particles with a size d. sp Larger than the size d of the detected pixel sp The situation.

[0140] During preparatory step 090, the transformation function H is determined such that... This step can be implemented through digital simulation or parametric studies, taking into account the optical components of the optical transmission and imaging devices. The conversion function H is stored in the memory of the computing unit 62. Thus, knowledge of this conversion function H allows for the determination of the spatial phase distribution based on the correction applied to the reference signal. Determine its spatial intensity distribution I in the receiving plane ij ref .

[0141] During step 100, the projected target signal s o (t) to illuminate the scene instantly and preferably uniformly.

[0142] Iteratively at an increasing fixed time t i Perform the following steps. At each determined time t i The method determines the distance to map D. ij (t i Then, determine time t. i Transition to the next moment t i+1 At each defined time t i The optimization is performed in two phases, denoted as k=0 and k=1, in relation to the spatial distribution of the parameters of interest. Steps 110 to 113 involve the first phase of the optimization phase (k=0), and steps 120 to 122 involve the second phase (k=1).

[0143] During step 110, and simultaneously with step 100, for example by means of the shutter, the reference signal s is paused.ref (t i,k＝0 The transmission of the reference signal. Furthermore, the strength I of the transmitted reference signal. ref (t i,k＝0 The value is zero. Step 111 expresses the fact that the spatial phase modulator 61 remains inactive.

[0144] During step 112, the matrix photodetector 50 receives and detects the incident optical signal, i.e., the backscattered object signal s. or (t i Furthermore, the spatial intensity distribution I of the detected signal. ij det (t i,k＝0 The spatial intensity distribution I corresponding to the backscattered object signal ij or (t i,k＝0 Therefore, it has a speckle pattern. This spatial intensity distribution I ij or (t i,k＝0 The data is transmitted to the computing unit 62 and stored in the memory.

[0145] During step 113, the calculation unit 62 determines the reference signal to be applied such that its spatial intensity distribution I ij ref (t i,k＝1 ) equals the detected spatial intensity distribution I ij or (t i,k＝0 ) optimized and corrected spatial phase distribution In other words, calculate the following:

[0146] Then the second phase of the optimization phase begins (k=1). During step 120, the reference signal s is allowed. ref (t i,k＝1 The reference signal is transmitted through optical transmission device 30 and phase correction device 60.

[0147] During step 121, the spatial phase modulator 61 modulates the optimized and corrected spatial phase distribution. An optimized, corrected spatial intensity distribution I is applied to the reference signal, resulting in the reference signal having an optimized spatial intensity distribution in the receiving plane. ij ref,opt (t i,k＝1 This is roughly equal to the spatial intensity distribution I of the backscattered object signal (and therefore the speckle pattern). ij or (t i,k＝0In this case, equality roughly relates to the geometric distribution of bright and dark regions, and the total luminous flux of the reference signal (integrated over the entire surface of the detection pixel matrix) may or may not be equal to the total luminous flux of the backscattered object signal.

[0148] During step 122, the matrix photodetector 50 receives and detects the incident optical signal, which is now the backscattered object signal and the reference signal that form the heterodyne signal. Furthermore, the spatial intensity distribution I of the detected signal... ij det (t i,k＝1 The spatial intensity distribution I corresponding to the heterodyne signal ij het (t i,k＝1 This depends on the spatial intensity distribution I of the backscattered object signal. ij or (t i,k＝1 Spatial intensity distribution of the reference signal I ij ref,opt (t i,k＝1 ).

[0149] Furthermore, the reference signal possesses a “speckle pattern” associated with the backscattered object signal. The total luminous flux remains constant, but it is distributed in bright regions roughly superimposed on the speckle grains. The area S illuminated by the bright regions of the reference signal is larger than that illuminated by uniform illumination. ij ref,opt (t i,k＝1 Therefore, the intensity of the heterodyne signal (DC and AC components), the intensity of the AC component of the heterodyne signal (the useful signal) alone, and its signal-to-noise ratio are all improved by the non-uniform illumination of the reference signal associated with the speckle pattern of the backscattered object signal. The performance capability of imaging system 1 is thus improved. Of course, this assumes that the matrix photodetector is not saturated when the incident light signal is received.

[0150] During step 130, imaging system 1 calculates the difference frequency f of the heterodyne signal at each detected pixel. b And determine the distance of the scene to map D. ij (t i Then, determine time t. i Transformation into the next moment t i+1 The method can then be continued by repeating step 110, especially when the scene is dynamic.

[0151] Figure 6B It is used to determine the distance map D of a scene using an imaging system 1 according to one embodiment. ij (t iThe flowchart shows the steps of another method. In this case, for clarity, the steps are also separated, but several of them are performed simultaneously or practically simultaneously.

[0152] Based on iteratively changing the spatial intensity distribution of the reference signal I ij ref (t i This allows for the optimization of the spatial distribution of the parameters of interest to implement a deterministic approach in this case. This could be the intensity of the heterodyne signal (DC and AC components), the intensity of the AC component of the heterodyne signal alone, or even the signal-to-noise ratio. This method can be applied to the speckle particle size d. sp Larger than the size d of the detected pixel px The situation, and size d sp Less than d px The situation...

[0153] During step 200, the projected target signal s o (t) to illuminate the scene simultaneously and preferably uniformly.

[0154] During step 210, simultaneously with step 200, the reference signal s is transmitted. ref (t). The reference signal propagates to the matrix photodetector 50 (through the optical imaging device 40) via the optical transmission device 30 and the phase correction device 60.

[0155] The subsequent steps 220 to 224 constitute part of the stage for optimizing the spatial distribution of the parameters of interest. They are performed iteratively, with the exponent changing from the initial value k = 0 to the final value kf. This iterative optimization is performed for each specific time t. i Implementation. When it is complete, determine time t. i Transformation into the next moment t i+1 .

[0156] During step 220, the spatial phase modulator 61 modulates the corrected spatial phase distribution. A reference signal is applied such that the reference signal has a spatial intensity distribution I in the receiving plane. ij ref (t i,k For the initial iteration k=0, the corrected spatial phase distribution It can be uniform, or it may include random phase changes, or it may even be equal to the phase change at the previous time t. i-1 Determined optimized spatial phase distribution

[0157] During step 221, the matrix photodetector 50 receives and detects the incident optical signal, thus forming a spatial intensity distribution I. ij het (ti,k The heterodyne signal consists of the backscattered object signal and the reference signal. This depends on the spatial intensity distribution I of the backscattered object signal. ij or (t i The corrected spatial intensity distribution of the reference signal I ij ref (t i,k It is important to note that the backscattered object signal can have a spatial intensity distribution I that remains unchanged during each iteration of the optimization phase (when the scene is static). ij or (t i ), or may have minor variations.

[0158] During step 222, the spatial distribution of the parameter of interest is determined. This could be the intensity of the heterodyne signal (DC and AC components), the intensity of the AC component of the heterodyne signal alone (the useful signal), or the signal-to-noise ratio (SNR). In this example, it is the SNR. ij het (t i,k ).

[0159] During step 223, the optimization criterion C(t) is determined. i,k When optimization is based on a predetermined number of iterations, this involves the exponent k: the value of the exponent is then compared with a predetermined value kf (during step 224), and the optimization phase is repeated if that value is not reached. This may involve the spatial distribution SNR of iteration k. ij het (t i,k Spatial distribution SNR of the previous iteration k-1 ij het (t i,k The local deviation. This deviation can be the sum of the squares of all other local deviations, or the maximum value of the local deviation (e.g., ...). Figure 6B (As shown).

[0160] During step 224, the optimization criterion C(t) will be optimized. i,k The value of ) is different from the predetermined value C. th Compare. Repeat the optimization loop (k becomes k+1), and implement steps 220 to 224 again until the optimization criterion reaches the predetermined threshold.

[0161] It is important to note that during step 220, the spatial phase distribution of iteration k+1... The computational unit 62 is based on the optimization criterion determined in iteration k, more specifically based on the optimization criterion C. ij (t i,k (In this example, based on SNR) ij het (ti,k )-SNR ij het (t i,k-1 The spatial distribution of the difference is used to determine the local phase. The computation unit 62 then uses an optimization algorithm (gradient descent, etc.) to change the local phase until the optimization criterion reaches a predetermined threshold.

[0162] Finally, during step 230, imaging system 1 calculates the difference frequency f of the heterodyne signal at each detected pixel. b And determine the distance of the scene to map D. ij (t i Then, determine time t. i Transition to the next moment t i+1 The method is determined by repeating step 220.

[0163] It should be pointed out that, Figure 6A and 6B The two examples of the determination method shown are provided illustratively. Other examples of how this method can be implemented are also provided. Thus, in Figure 6A Within the scope of the method, it can be implemented in Figure 6B Similar iterative optimization is shown: Thus, through iterative optimization, the spatial intensity distribution I of the reference signal is... ij ref (t i,k The spatial intensity distribution of the backscattered object signal (I) can be considered. ij or (t i,k ).

[0164] Figures 7A to 7D An advantageous configuration of the imaging system 1 is shown, wherein an equivalent light source 36 for forming a reference signal is formed in an intermediate plane located at or near the aperture stop. These figures illustrate the receiving plane of the matrix photodetector 50, the optical imaging device 40 (lenses 41, 43 and aperture stop 42), and the equivalent light source 36 for the reference signal.

[0165] Furthermore, the optical transmission device 30 and, if applicable, the optical imaging device 40, generate an image of the spatial phase modulator 61 and / or diffuser in the intermediate plane. The spatial phase modulator 61 and / or diffuser 32 are therefore located in a plane conjugate with the intermediate plane. This configuration is similar to that described in French patent application FR 2000408 filed on January 16, 2020. The equivalent light source 36 is positioned longitudinally along the optical axis, and its lateral dimension can be defined by optical shaping elements. The equivalent emission angle can be defined by the diffuser 32 of the optical transmission device 30 ( Figure 3A ) or through grating matrix 61 ( Figure 3B () to limit.

[0166] This causes the wavefront of the incident reference signal at each detection pixel to be closer to, or even identical to, the wavefront of the backscattered object signal in terms of shape and principal orientation. This, in turn, limits the generation of interference fringes between the two optical signals, which can degrade the quality of the detected heterodyne signal.

[0167] The field of view (FOV) of the optical imaging device 40 is defined as the angle at which the matrix photodetector 50 is sensitive to the backscattered object signal. In this case, it is defined as the angle measured at the center of the output pupil between two incident rays of the backscattered object signal that pass through the same point O and reach two end pixels opposite each other relative to the optical axis. The end pixels are the detection pixels located at the edges of the detection matrix. Furthermore, the output pupil is the image of the aperture stop through a lens located downstream of the aperture stop.

[0168] Next, as in Figure 7B More specifically shown is the central region Zc that defines the light rays at the edge of the vignetting-free region (the term "vignetting-free region" is used in the publication by JEGreivenkamp entitled Field Guide to Geometrical Optics, SPIEPress, Bellingham, WA (2004)). It is defined here as the region laterally defined by the edge rays of the complete optical field, i.e., the rays of the backscattered object signal that pass through the contour of the aperture stop and reach the end pixels of the detection pixel matrix. These end pixels are located at the edges of the detection pixel matrix, i.e., the pixels furthest from the optical axis. This central region takes the form of two cones sharing the same base defined by the contour of the aperture stop. The apexes of the cones are M and M', located on the optical axis, upstream and downstream of the aperture stop, respectively. They are the longitudinal end points of the central region.

[0169] Figure 7A The receiving plane of the optical imaging device 40 and imaging system 1 according to an alternative embodiment is shown in detail and in a longitudinal cross-sectional view, wherein the equivalent light source 36 of the reference signal formed by the optical transmission device 30 is located in a central region Zc upstream of the aperture stop in this case. The central region Zc is more specifically located in... Figure 7B As shown in the image.

[0170] For the two end pixels Px a and Px b On one hand, it shows the light rays at the edge of the non-vignetting domain of the backscattered object signal (solid line), and on the other hand, it shows the light rays originating from the two opposite ends of the equivalent light source 36 and received by each of the end pixels (dashed line).

[0171] Therefore, for the end pixel Px aIn this case, it receives two rays R that pass through the edge of the aperture stop (in the longitudinal plane) from the backscattered object signal. a or1 R a or2 And two rays R from the reference signal originating from the edge of the equivalent light source 36. a ref1 R a ref2 Light beam R a or1 and R a ref1 Both pass through the same edge of the aperture stop and are therefore superimposed. For the second end pixel Px... b The received light rays will not be described in detail due to their similarity.

[0172] The optical transmission device 30 is adapted such that the equivalent light source 36 has a divergent (or equivalent emission) cone at each of its points, covering the entire detection pixel matrix. Thus, each of the detection pixels effectively receives a reference signal in addition to the backscattered object signal. This results in the proper formation of a heterodyne signal at each detection pixel.

[0173] Since the equivalent light source 36 is arranged between points M and M' and has a lateral dimension at least equal to the lateral dimension of the central region, at each detection pixel, the receiving cone of the reference signal is approximately the same as the receiving cone of the backscattered object signal.

[0174] Figure 7B and Figure 7A The same, except that no ray R from the reference signal originating from the equivalent light source 36 is shown. a ref R b ref In this example, the light rays R at the edge of the region without vignetting are highlighted. a or1 R a or2 and R b or1 R b or2 (In this case, in the longitudinal plane) the central region Zc is defined laterally and longitudinally. The point M upstream of the aperture stop is located on the optical axis and lies within the ray R. a or1 and R b or2 The point M' downstream of the aperture stop is located on the optical axis and at the intersection of the two points, while the point M' is located on the optical axis and at the intersection of the two points. a or2 and R b or1The intersection of the two. The largest lateral dimension of the central region Zc lies in the plane of the aperture stop.

[0175] Figure 7C and Figure 7A The same, except that only the detected pixel Px is considered. a The received light ray, i.e., the light ray R of the backscattered object signal. a or1 and R a or2 and the ray R of the reference signal a ref1 and R a ref2 (Reference R not shown) a ref1 This is because the ray is related to ray R. a or1 (Overlap). Each point of the equivalent light source 36 illuminates at least the entire receiving plane, and in this example, only the receiving plane is illuminated. Thus, the same detection pixel receives not only the backscattered object signal but also reference signals originating from all points of the equivalent light source 36.

[0176] It is important to note that, compared to the case where the equivalent source 36 would be located outside the central region Zc, the differences between the wavefronts of these optical signals have been reduced: in particular, the differences in the wavefronts incident on the detection pixel Px have been reduced. a The deviation angle θ between the principal directions of these wavefronts a The value. The principal direction is defined at the detected pixel under consideration, in this example at the end pixel Px. a At that point, the average direction of the angular cone of the incident beam (reference signal or backscattered object signal).

[0177] Thus, the equivalent light source 36 forming the reference signal in the central region Zc reduces the deviation angle θ at the considered detection pixel. This, in turn, increases the dimension between the fringes that can be formed between the backscattered object signal and the reference signal. The dimension between the fringes is initially approximately λ / 2sinθ, where λ is the wavelength of the incident optical signal. The increase in the dimension between the fringes is further achieved by having angular cones at each detection pixel with incident optical signals (reference signal and backscattered object signal) that are very similar to or even identical to each other.

[0178] This improves the intensity of the detected heterodyne signal. In fact, the significant deviations between the principal directions of the wavefront at angle θ and / or between the angle cones of the incident optical signal lead to a reduction in the size between the fringes, which can be approximately the size of the detection pixel d. pxThe size, or even smaller. Furthermore, the fact that several interference fringes (bright and dark areas) exist at the scale of each detection pixel during the photodiode acquisition time can lead to a reduction in the detection quality of the heterodyne signal.

[0179] It is particularly advantageous that the equivalent light source 36 is positioned as close as possible to the aperture stop, and its width is at least equal to or preferably approximately equal to the width of the central region, and thus, in some cases, equal to the width of the aperture stop.

[0180] For this reason, Figure 7D Show Figure 7C The illustrated variant of the structure, in this example, places the equivalent light source 36 approximately in the plane of the aperture stop 42. It can be seen that in the detection pixel Px... a At the point, the end ray R a or1 and R a ref1 Overlap, light rays R a or2 and R a ref2 They also overlap. This is also the case for rays defining the principal directions of these signals. The deviation angle θ is approximately zero. Furthermore, since the equivalent light source 36 approximately covers the entire surface of the aperture stop 42, the angular cones of the two incident optical signals at each detection pixel are approximately equal. In addition, the wavefronts of the backscattered object signal and the reference signal are approximately the same. The dimension between the fringes can exceed the dimension of the detection matrix, so that interference fringes between these signals can be considered that might degrade the quality of the heterodyne signal detected by each detection pixel. Then, solid shadow interference fringes are referenced. The performance capability of the imaging system 1 is further improved.

[0181] The above describes specific embodiments. Different variations and modifications will be apparent to those skilled in the art.

Claims

1. An FMCW type LIDAR imaging system (1), comprising: A coherent light source suitable for transmitting the main signal of continuous frequency modulation (10); An optical projection device (20) for projecting a portion of the main signal, called the object signal, onto the scene to be illuminated in real time. Optical transmission device (30) for transmitting a portion of the main signal, referred to as the reference signal, to the optical imaging device (40). The optical imaging device (40) is adapted to receive the speckle pattern portion of the object signal backscattered by the scene, referred to as the backscattered object signal. Matrix photodetector (50), which includes detection pixels (P ij The matrix detects that each pixel receives the backscattered object signal and the reference signal in a receiving plane conjugate with the scene via the optical imaging device (40), wherein the backscattered object signal and the reference signal form a frequency difference (f) b The heterodyne signal representing the frequency of the distance (D) between the matrix photodetector (50) and the scene; The imaging system is characterized in that it includes a phase correction device (60), which comprises: A spatial phase modulator (61), which is disposed upstream of the optical imaging device (40) on the path of the reference signal, is adapted to apply a spatial phase distribution, referred to as a modified spatial phase distribution, to the reference signal; A computing unit (62), which is connected to the matrix photodetector (50) and the spatial phase modulator (61), is adapted to: According to the first spatial distribution representing the spatial intensity distribution of the backscattered object signal in the receiving plane detected by the matrix photodetector (50), Determine the optimized, corrected spatial phase distribution to be applied to the reference signal by the spatial phase modulator (61). The reference signal is configured in the receiving plane to have a spatial intensity distribution called a modified spatial intensity distribution, which optimizes the spatial distribution of the parameters of interest representing the heterodyne signal. These parameters are selected from the intensity of the heterodyne signal, the intensity of the AC component of the heterodyne signal, or the signal-to-noise ratio.

2. The imaging system (1) according to claim 1, wherein, The detected pixel has a lateral dimension (d) less than 2×λ×NO. px ), where λ is the wavelength of the backscattered object signal and NO is the number of apertures of the optical imaging device (40).

3. The imaging system (1) according to claim 1, wherein, The optical projection device (20) and the optical transmission device (30) include free-space optical elements.

4. The imaging system (1) according to claim 3, wherein, The spatial phase modulator (61) is a liquid crystal phase modulator.

5. The imaging system (1) according to claim 1, wherein, The optical projection device (20) includes a guiding optical element implemented in a functionalized substrate (2), the functionalized substrate including a diffraction grating matrix adapted to receive the main signal via a waveguide and project the object signal onto the outside of the functionalized substrate (2).

6. The imaging system (1) according to claim 5, wherein, The phase correction device (60) includes a guiding optical element implemented in the functionalized substrate (2), and the spatial phase modulator includes a diffraction grating matrix adapted to receive the reference signal via a waveguide and project the reference signal onto the outside of the functionalized substrate (2) after the spatial phase distribution to which the correction is applied.

7. The imaging system (1) according to claim 5, wherein, The optical transmission device (30) includes a free-space optical element adapted to transmit the reference signal projected by the spatial phase modulator (61) to the optical imaging device (40).

8. The imaging system (1) according to claim 5, wherein, The detection pixel matrix is ​​implemented in or on the functionalized substrate (2).

9. The imaging system (1) according to claim 1, wherein, The optical transmission device (30) and the optical imaging device (40) are adapted to transmit the reference signal and the backscattered object signal along the same optical axis to the matrix photodetector (50).

10. The imaging system (1) according to claim 1, wherein: The optical imaging device (40) includes an aperture stop and at least one free-space optical element, thereby defining a field of view for the matrix photodetector and a central region (Z) laterally defined by light rays propagating to the edges of the backscattered object signal of the detection pixels, referred to as end pixels, located at the edges of the detection pixel matrix. C ); The optical transmission device (30) and the optical imaging device (40) are adapted to form an image of the reference signal in an intermediate plane orthogonal to the optical axis of the optical imaging device (40), thereby forming an equivalent light source (36) of the reference signal. The equivalent light source (36) contains the light rays at the edge of the backscattered object signal in the non-vignetting region in the central region (Z). C )middle; The equivalent light source (36) has an emission angle at each point that is at least equal to the field of view of the optical imaging device (40) for the reference signal.

11. The imaging system (1) according to claim 10, wherein, The equivalent light source (36) has a central region (Z) that is at least equal to the light rays at the edge of the backscattered object signal without vignetting. C The lateral dimension of ).

12. A method for determining a distance map of a scene using the imaging system (1) as claimed in claim 1, wherein, The parameter of interest is the intensity of the heterodyne signal, and the method includes the following steps: a / Projecting the object signal onto the scene via the optical projection device (20) to illuminate the scene in real time; b / The first spatial intensity distribution of the incident optical signal representing the backscattered object signal is detected by the matrix photodetector (50); c / The calculation unit (62) determines the corrected spatial phase distribution to be applied to the reference signal by the spatial phase modulator (61); d / The modified spatial phase distribution is applied to the reference signal through the spatial phase modulator (61); e / The spatial intensity distribution of the heterodyne signal is detected by the matrix photodetector (50); f / Through the computing unit (62), the spatial distribution of the parameter of interest is determined based on the spatial intensity distribution of the detected heterodyne signal; If applicable, repeat steps c / to f / by changing the modified spatial phase distribution until the determination criterion as a function of the spatial distribution of the parameter of interest reaches a predetermined threshold. g / When the determination criterion reaches the predetermined threshold, the distance map is determined.

13. The method according to claim 12, wherein: During step b / , the transmission of the reference signal is paused so that the incident optical signal is the backscattered object signal; During step c / , the modified spatial phase distribution is determined based on a predetermined transformation function that expresses the spatial intensity distribution of the reference signal in the receiving plane as a function of the spatial phase distribution applied by the spatial phase modulator (61), such that the spatial intensity distribution of the reference signal in the receiving plane is approximately equal to the spatial intensity distribution of the detected backscattered object signal. During step d / , the transmission of the reference signal is no longer paused.

14. The method according to claim 12, wherein: During step b / , the detected optical signal is the heterodyne signal; Repeat steps c / to f / , and change the spatial phase distribution according to the predetermined optimization algorithm and the optimization criteria in the previous iteration until the optimization criteria reach the predetermined threshold, thereby optimizing the spatial distribution of the parameter of interest.

Images