METHOD AND DEVICE FOR VISUALIZING LASER IMPULSE IMPULSE IMPULSE ON A TARGET

Description

[0001] The present invention relates to the field of guidance, for example the guidance of aerial munitions such as missiles or rockets. BACKGROUND OF THE INVENTION

[0002] Guided munitions typically consist of control surfaces whose orientation allows the munition to be steered, and a guidance system connected to the control surfaces to orient them so that the munition is directed toward the target. One commonly used guidance system is the seeker (or "homing head" in common parlance), which includes a sensor designed to detect a signal from the target and a processing circuit that interprets the detection data from the sensor into commands for the control surfaces.

[0003] There are munitions equipped with semi-active laser homing devices that detect the reflection of laser pulses from a laser designator onto the target. The laser designator can be mounted on the vehicle that fired the munition or on another vehicle; it can also be operated by a dismounted soldier located some distance from the target. In addition, an observer may be required to observe the intended target to confirm that it is indeed the correct one. The main characteristics of the laser designator are: the emission wavelength (typically 1064 nm), the energy emitted per pulse (typically a few tens to a few hundred millijoules), the duration of a pulse (typically a few tens of nanoseconds), the pulse repetition period or PRI (typically 50 to 120 ms).

[0004] The success of guidance relies largely on the ability to visualize the impact of laser pulses on the target (this impact is commonly referred to as a "laser spot") and to measure the PRI, as the latter is a characteristic that allows us to discriminate the origin of the different laser spots that may be simultaneously present on a battlefield.

[0005] Indium gallium arsenide (InGaAs) CMOS sensors are used to visualize this spot, but they do not allow for the determination of the PRI. To detect the PRI, a laser spot tracker detector based on avalanche photodiodes must be added to the CMOS sensor, which increases the complexity, size, power consumption, and cost of the visualization device.

[0006] A method for detecting laser pulse impacts in an image, which allows the pulse repetition period to be determined using a rolling shutter readout circuit, is known from documents WO-A-2021 / 198631 and US-A-2021 / 247231. The pulse repetition period is determined by a Fourier transform. SUBJECT OF THE INVENTION

[0007] The invention aims in particular to enable the visualization of laser pulse impacts and to estimate the PRI in a simple, reliable and relatively inexpensive way. SUMMARY OF THE INVENTION

[0008] For this purpose, the invention provides a method for visualizing the impacts of laser pulses emitted by a laser designator, according to a predetermined repetition period, on a target present in a scene, by means of a device comprising: a silicon CMOS type optronic sensor having a predetermined number of lines, a "Rolling Shutter" type reading circuit, and an electronic image processing circuit.

[0009] The process includes the following steps: expose each line of the sensor for an exposure time (t EXPOSURE) and read each line for a readout time (t READOUT) such that the exposure time is greater than or equal to the readout time of each line but less than or equal to the product of the readout time of each line and the number of lines, detect in each image frame provided by the readout circuit a transition between lines of the sensor exposed to a signal resulting from a reflection of one of the laser pulses and lines not exposed to said signal, estimate from this transition a pulse timing instant in at least two images and deduce the pulse repetition period.

[0010] Thus, the invention exploits a drawback of CMOS sensors combined with a rolling shutter readout circuit, namely the existence of a readout phase shift from one line to the next (a drawback that results in artifacts and geometric distortions in certain moving scene images). Indeed, with such an arrangement, the elementary detectors (commonly called "pixels" or "photosites") have no memory: the sensor exposure (exposure time tEXPOSITION) is performed sequentially, line of elementary detectors after line of elementary detectors, and each line is read (readout time tREADOUT) as soon as its exposure is complete, as illustrated in the figure 1 It is thanks to this time phase shift (equal to the reading time) that the images will contain an area exposed to the reflection of the pulses and an unexposed area, because the luminous flux associated with the reflection of a laser pulse is brief and will be interrupted before all the lines have been exposed. Detecting the transition is therefore relatively easy, and knowing the time phase shift between the lines, it is possible to estimate the IR. The invention also relates to a device for implementing this method.

[0011] Other features and advantages of the invention will become apparent from the following description of a particular and non-limiting embodiment of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Reference will be made to the attached drawings, including: There figure 1 is a timing diagram illustrating the operation of a rolling shutter readout circuit, with a representation of the exposure time and readout time of each line, highlighting the time lag between the exposure of one line and the next; The figure 2 is a temporal pattern, analogous to that of the figure 1 illustrating the detection of a laser pulse; The figure 3 is a view of the image corresponding to this detection, with the sensor lines represented; The figure 4 is a temporal pattern, analogous to that of the figure 1 illustrating another detection of a laser pulse; The figure 5 is a view of the image corresponding to this other detection with the representation of the sensor lines; The figure 6 is a temporal representation of the succession of image frames and the detection moments of laser pulses; The figure 7 is an illustration of how the impact zone of pulses is detected in the scene; The figure 8 is a flowchart of a visualization device according to the invention. DETAILED DESCRIPTION OF THE INVENTION

[0013] The invention relates to a method and device for visualizing a scene and the impacts of laser pulses emitted by a laser designator on a target present in the scene. The main characteristics of the pulses emitted by the laser designator are: the emission wavelength (typically 1064 nm), the energy emitted per pulse (typically a few tens to a few hundred millijoules), the duration of a pulse (typically a few tens of nanoseconds), the pulse repetition period or PRI (typically 50 to 120 ms).

[0014] In the described embodiment, the preceding data is known to the user of the device: the user is, for example, faced with a scene containing several targets, each of which is illuminated by a laser designator projecting laser pulses that form a laser spot. The user then uses the device and method of the invention to visualize the laser spot and determine the pulse repetition period to ensure that the visualized laser spot corresponds to the one the user is looking for. It should be noted that the invention is also useful for a user facing a single target illuminated by a single designator. The device of the invention can thus be used by the laser designator operator, a separate observer, and also in a munition guided by said laser spot.

[0015] With reference to the figure 8 The device of the invention comprises: an optronic sensor 1, an electronic reading circuit 2, and an electronic image processing circuit 3.

[0016] As is known, the optronic sensor 1 is a silicon CMOS type with an epitaxial layer of sufficient thickness to make the optronic sensor sensitive to a wavelength of 1064 nm. The optronic sensor 1 comprises an array of elementary detectors (or pixels or photosites), namely photodiodes, arranged in rows and columns. The optronic sensor 1 has a predetermined number of rows, Z, here equal to five for simplicity in the description and figures. Typically, each elementary detector struck by a photon during the exposure phase is likely to produce a charge that will be recovered by the readout circuit 2 during the readout phase.

[0017] The readout circuit 2, known in itself, is of the "Rolling Shutter" type and allows the charges of the elementary detectors to be retrieved line by line. More precisely, the readout circuit 2 can be controlled to expose each line for an exposure time tEXPOSITION and read each line for a reading time tREADOUT (see on the... figures 1 , 2 And 4 The exposure time tEXPOSITION is adjustable by the readout circuit 2, while the readout time tREADOUT is fixed. The readout circuit 2 provides image frames (visible on the figure 6 ), each image frame corresponding to an image such as those represented on the figures 3 , 5 And 7 .

[0018] The processing circuit 3 includes at least one processor and a memory containing at least one program executable by the processor to implement the method of the invention.

[0019] The process according to the invention comprises the steps of: direct the optronic sensor 1 towards the scene containing the target onto which the designator projects the laser pulses; expose each line of the sensor for the exposure time t EXPOSURE and read each line for the reading time t READOUT such that the exposure time t EXPOSURE is greater than or equal to the reading time t READOUT of each line but less than or equal to the product of the reading time t READOUT of each line and the number Z of lines to obtain a video stream of the scene and more precisely of the target with the laser spot; detect in each image frame provided by the reading circuit 2 a transition between lines of the optronic sensor 1 exposed to a signal resulting from the reflection of one of the laser pulses and lines not exposed to said signal, estimate from this transition a pulse instant in at least two images and deduce the repetition period of the pulses.

[0020] More specifically, in the preferred embodiment, the image processing circuit 3 is programmed to detect and isolate the laser spot in the images. To facilitate the detection of the laser spot in the image, the playback circuit 2 can be controlled during image capture so that each image frame of the video stream has a duration tFRAME less than or equal to half the repetition period of the pulses to be detected, and that the exposure time tEXPOSITION is greater than or equal to half the duration tFRAME. The first condition ensures that at least one image is available on which the laser spot does not appear (this image therefore being representative of an ambient background), and the second condition maximizes the probability of integrating the laser spot. The laser spot can then be isolated from the images, and the ambient background can be subtracted from each image frame so that only the laser spot remains in the image.As illustrated on the . figure 7 An image I2 representing the background of the scene is subtracted from the images I1 of a video stream representing the scene with the laser spot. The images I1-I2 of the resulting video stream then essentially consist of the laser spot.

[0021] We observe that the laser spot is off-center and does not occupy the entire image field. However, for the PRI estimation to be robust, it is desirable for the laser spot to occupy the entire image (in other words, the entire area imaged by the optronic sensor 1). The image processing circuit 3 is programmed to define a window (or "Region of Interest") around the laser spot to obtain the "PRI window" image.

[0022] Note that a pointing misalignment of the designator (resulting, for example, from atmospheric turbulence, a designator instability, or other factors) could cause the laser spot to move outside the window defined during the windowing operation. To make the PRI measurement more robust to pointing misalignment, defocusing can be performed to limit the influence of such misalignment during the windowing operation, as this defocusing enlarges the laser spot in the image. The defocusing required is proportional to the pointing misalignment encountered. However, such defocusing degrades the signal-to-noise ratio and therefore reduces the maximum measurement range, which must be taken into account if it is chosen. The image processing circuit 3 is then programmed to analyze each frame of the video stream.Two scenarios are possible regarding the detection of the laser pulse by the optronic sensor 1 when the exposure time tEXPOSITION is greater than or equal to the readout time tREADOUT of each line but less than or equal to the product of the readout time tREADOUT of each line and the number Z of lines. This is shown in Figure 1. figure 2 The instant of appearance t of a laser spot relative to the exposure time t EXPOSI-TION and the readout time t READOUT of each line of the optronic sensor 1. We see that the instant of appearance occurs after the exposure time of lines 1 to 3 and during the exposure time of lines 4 and 5. The resulting image, extracted from the video stream and visible on the figure 3 , therefore includes an area not exposed to the impulse (occupying the upper three-fifths of the image) and an area exposed to the impulse (occupying the lower two-fifths of the image).

[0023] We represented at the figure 4 the instant of appearance t of a laser spot relative to the exposure time t EXPOSI-TION and the reading time t READOUT of each line of the optronic sensor 1 according to a temporal sequence different from that of the figure 2 We see that the moment of appearance occurs during the exposure time of lines 1 to 3 and before the exposure time of lines 4 and 5. The resulting image, extracted from the video stream and visible on the figure 5 , therefore includes an area exposed to the impulse (occupying the upper three-fifths of the image) and an area not exposed to the impulse (occupying the lower two-fifths of the image).

[0024] The transition between the two zones is therefore a precise indicator, accurate to the readout time, of the laser pulse (all the more precise the shorter the laser pulse duration – here a ratio between 100 and 1000 – compared to the product of the number of sensor lines and the sum of the exposure and readout times of each line). The transition is detected by comparing the value of each pixel in each image line to a threshold adjusted according to the noise level of the optronic sensor 1. Preferably, each image frame in the video stream has a duration less than or equal to half the repetition period of the pulses to be detected, in order to further facilitate the detection of the transition.

[0025] It is therefore possible to estimate the laser time t in an image n from the following equations: for the case corresponding to figures 2 And 3 t 0 , n + t EXPOSITION + L 1 , n − 1 * t READOUT ≤ t laser , n t laser , n ≤ t 0 , n + t EXPOSITION + L 1 , n * t READOUT for the case corresponding to figures 4 And 5 t 0 , n + L 1 , n − 1 x t READOUT ≤ t laser , n ≤ t 0 , n + L 1 , n x t READOUT with, in both cases, t 0,n the initial time of image frame n, L 1,n the index of the last line of the sensor preceding the transition detected in image frame n, t laser,n the time of dating of the laser pulse in image frame n, This estimate is accurate to within the reading time.

[0026] Furthermore, in the video stream, we have several image frames and we will try to detect a transition in each of the corresponding images in order to calculate the repetition period of the pulses.

[0027] We can thus see on the figure 6 that pulse 1 was emitted during the capture of frame 1, pulse 2 was emitted during the capture of frame 3, pulse 3 was emitted during the capture of frame 6, and pulse 4 was emitted during the capture of frame 9.

[0028] The detection of transitions in image frames 1 to 9 will allow us to determine the laser pulse time t in each of frames 1, 3, 6 and 9 and to calculate the pulse repetition period using the following formula: PRI = t laser , n + m − t laser , n / m with t laser,n the time of the laser pulse in frame n, t laser,n+m the time of the laser pulse in frame n+m, m the number of frames separating frames n and n+m.

[0029] In the example of the figure 6 , we can choose frames 1 and 3, 1 and 6, 1 and 9, 3 and 6, 3 and 9, and / or 6 and 9 to calculate the PRI.

[0030] Of course, the invention is not limited to the embodiment described but encompasses any variant falling within the scope of the invention as defined by the claims.

[0031] In particular, the device may have a different structure than the one described.

[0032] The number and arrangement of circuits and the distribution of functions between circuits can be modified.

[0033] Furthermore, it is possible to determine the position of the laser spot not by detection but by prior knowledge of the targeted area in the image and the angular extent of the laser spot.

[0034] All or part of the following characteristics may be omitted: Transition detection is performed using a threshold adjusted according to the sensor noise level; the duration of the image frames is less than or equal to half the pulse repetition period; the background ambient background is subtracted to isolate the signal in all or part of the image frames; the image frame is recentered around the impacts prior to the transition detection; and defocusing is used to facilitate the recentering of the image frame around the impacts.

[0035] Therefore, defocusing is not necessary if the risk and / or extent of misalignment are low. Similarly, windowing is unnecessary if the sensor field is specifically focused on the laser spot during image capture (for example, using a lens array with adjustable focal length).

[0036] The ratio between the duration of the laser pulse and the product of the number of sensor lines and the sum of the exposure and readout times of each line can be smaller or larger than that mentioned, depending on the accuracy required.

[0037] We can focus on only a portion of the images, for example the one showing the configuration of the figure 3 or that of the figure 5 .

Claims

1. Method for visualising impacts of laser pulses emitted by a laser designator, according to a predetermined repetition interval (PRI), onto a target present in a scene, by means of a device comprising: a silicone-based CMOS-type optronic sensor (1) having a predetermined number of rows (Z), a rolling shutter-type reading circuit (2), and an electronic image processing circuit (3); the method comprising the steps of: - exposing each row of the sensor for an exposure time (tEXPOSURE) and reading each row for a reading time (tREADOUT) so that the exposure time is no less than the reading time of each row, but no more than the product of the reading time of each row and the number of rows, - detecting, in each image frame provided by the readout circuit, a transition between the rows of the sensor exposed to a signal resulting from a reflection of one of the laser pulses and of the rows not exposed to said signal, - estimating, from this transition, a pulse dating instant in at least two images and in order to infer the pulse repetition interval.

2. Method according to claim 1, wherein the detection of transition is performed by using an adjusted threshold according to a noise level of the sensor.

3. Method according to claim 2, wherein each image frame has a duration less than or equal to half of the pulse repetition interval.

4. Method according to claim 3, wherein the image processing circuit is arranged to remove an ambient backdrop from each image frame, so as to isolate said signal.

5. Method according to any one of the preceding claims, wherein the pulse dating instant is estimated from the following formulas: t 0 , n + t EXPOSURE + L 1 , n − 1 * t READOUT ≤ t laser , n t laser , n ≤ t 0 , n + t EXPOSURE + L 1 , n * t READOUT t 0 , n + L 1 , n − 1 × t READOUT ≤ t laser , n ≤ t 0 , n + L 1 , n × t READOUT with t0,n, the initial instant of the image frame n, L1,n, the index of the last row of the sensor preceding the transition detected in the image frame n, tlaser,n, the dating instant of the laser pulse in the frame n.

6. Method according to claim 5, wherein the pulse repetition interval (PRI) is estimated from the following equation: PRI = t laser , n + m − t laser , n / m with tlaser,n, the dating instant of the laser pulse in the frame n, tlaser,n+m, the dating instant of the laser pulse in the frame n+m, m, the number of frames separating the frames n and n+m.

7. Method according to any one of the preceding claims, comprising the step of recentring the image frame around the impacts, prior to the detection of the transition.

8. Method according to claim 7, comprising the step of performing a defocusing to facilitate the recentring of the image frame around the impacts.

9. Device for visualising a scene and impacts of laser pulses emitted by a laser designator, according to a predetermined repetition interval (PRI), onto a target present in the scene, comprising: a CMOS-type optronic sensor (1) having a predetermined number of rows (Z), a rolling shutter-type readout circuit (2) which can be controlled to expose each row for an exposure time (tEXPOSURE) and reading each row for a reading time (tREADOUT), so that the exposure time is no less than the reading time of each row, but no more than the product of the reading time of each row and the number of rows, and an electronic image processing circuit (3) arranged to implement the method according to any one of the preceding claims.