REALITY ENHANCEMENT PROCEDURES AND MONITORING SYSTEM

Description

[0001] The invention relates to a method for displaying a supervision image, and a supervision system for supervising a device. This device can be an industrial tool, a machine, a vehicle, a living being, a building, etc.

[0002] Supervision may take place either during normal operating conditions of the device, or when the device is undergoing testing, or any other specific circumstances.

[0003] The invention can in particular be used for the supervision of engines, and more particularly of turbomachines or rocket engines, when they are subjected to tests.

[0004] In the industrial world, the supervision of the operation of a device, for example a manufacturing unit, is generally carried out using a supervision system, under the responsibility of a supervision manager.

[0005] Such a monitoring system generally includes video surveillance means, comprising one or more cameras, the images of which are displayed on one or more screens.

[0006] The monitoring system may also include screens on which diagrams are displayed, for example fluid circulation diagrams, and on which the values ​​of critical parameters of the manufacturing unit are displayed in real time.

[0007] The monitoring system may also include a tracking system, which monitors various parameters of the device and issues alert messages indicating any malfunctions detected. These messages can be displayed on the fluid flow diagrams mentioned above.

[0008] The supervisor analyzes the images received from the cameras and the messages received in real time. Based on these messages, he can, for example, change the viewing angle of one of the cameras or zoom in on specific parts of the supervised manufacturing unit.

[0009] In the event of an anomaly or malfunction, a large number of error or anomaly messages are displayed almost simultaneously by the monitoring system. Given the considerable amount of information displayed, it is then difficult for the supervisor to assimilate the information quickly enough, select the relevant information and make the optimal decision.

[0010] In particular, important information is distributed across several screens (video screens showing images taken by cameras, screens showing fluid circulation diagrams, screens possibly displaying alert messages and values ​​of critical variables of the device, etc.), which complicates the work of the supervisor in understanding the situation and making the necessary decisions.

[0011] Also, a first objective of the present invention is to propose a display method for supervising a device by means of a supervision system, which allows an optimized display of information relating to a supervised device, so as to facilitate the understanding thereof by the person responsible for supervision.

[0012] To achieve this objective, according to a first aspect of the invention, a method of displaying a supervision image is proposed for the supervision of a device by means of a supervision system, the supervision system comprising a display and at least one real camera arranged so as to be able to film all or part of the device.

[0013] The display method is defined in claim 1.

[0014] Preferably (but not necessarily), the display method is implemented in real time, which allows the supervision image displayed on the display to be updated in real time. The display then displays a film or video sequence which presents the evolution of the device in real time.

[0015] The display process uses two types of images: acquired images, which are images acquired by the (real) camera(s), and computer-generated images representing the device. Computer-generated images are calculated using a three-dimensional digital model of the device (or at least part of it).

[0016] Very schematically, the display method according to the invention therefore consists of acquiring an image of the device using a real camera (the active camera), calculating a synthetic image representing all or part of the device, and combining the acquired image and the synthetic image to form a supervision image which is then displayed.

[0017] It is understood that the combination of the acquired image and the synthetic image is carried out in such a way as to preserve in the supervision image the most important information relating to the device.

[0018] Thus, the combined image obtained makes it possible to present on a single display or screen, at any time, the most relevant representation of the device.

[0019] In normal operation, the displayed supervision image may be an external, overview of the device, as acquired by one of the cameras.

[0020] However, the display method according to the invention multiplies the display possibilities, by making it possible to add to the information coming from the cameras (the acquired images), additional information which is integrated into the synthetic image. The supervision image thus obtained constitutes an extremely rich image, particularly useful for understanding the behavior of the supervised device. The synthetic image can be used, for example, to represent hidden internal parts of the device.

[0021] Thanks to the richness and amount of information contained in the supervision image, it is generally possible to reduce the number of screens required to ensure supervision of the device.

[0022] Furthermore, advantageously the supervision image is formed as a function of the state of the device, since either the determination of the image parameters, or the calculation of the synthetic image, or the formation of the supervision image, is (or are) carried out as a function of the state parameter or state parameters of the device.

[0023] For this purpose, it is naturally appropriate to preselect specific sets of image parameters, and / or a specific method of calculating the synthetic image, and / or a specific method of forming a supervision image, which are adapted so that, when the device is in a certain predetermined state, the supervision image formed is particularly relevant for interpreting the state of the device and enabling its supervision.

[0024] This operating mode allows a particularly relevant synthetic image to be submitted to the supervisor, and thus typically reduces the time spent by the supervisor analyzing the state of the supervised device, identifying malfunctions, and triggering appropriate corrective actions. IMAGE SETTINGS

[0025] Image parameters (camera orientation and position relative to the device; camera optical parameters) are the parameters that determine what appears in a camera's image. They specify where the camera is positioned, how it is oriented, and what its optical parameters (zoom, etc.) are.

[0026] The acquired image parameters define what will appear in the image acquired by the active camera, and the synthetic image parameters define which image will be calculated during the synthetic image calculation step c).

[0027] Image parameters can therefore concern a real camera, or a virtual camera producing an image by being positioned in a virtual scene.

[0028] Step a2) of determining the image parameters may be a passive step, during which the elements chosen (active camera) or determined (image parameters) either take default values, or are carried over and retain the value taken during the previous iteration of the program implemented by the supervision system.

[0029] The choice of the active camera in some cases alone determines both sets of image parameters when for both sets of parameters (acquired image and synthetic image) the current image parameters of the active camera are taken. This is what happens most often.

[0030] However, there may also be some differences between the acquired image settings and the synthetic image settings.

[0031] For example, the acquired image parameters and the synthetic image parameters can be different: to compensate for an error in positioning and / or orientation and / or in the optical parameters of the active camera; and / or because the computer-generated image represents only a portion of the active image. COMPUTER-GENERAL IMAGE

[0032] Computer-generated images are images of the device or a part of the device as could be provided by a virtual camera. They are in fact rendered images obtained (in a manner known per se) from a virtual scene in which the three-dimensional digital model of the device (or a part of it) is located, and in which one or more 'virtual cameras' are positioned.

[0033] In some cases it may be appropriate that there are differences between the virtual scene and the real device.

[0034] For example, the virtual scene may possibly include digital models of objects other than the device under test. For example, in the case of a rocket engine operational test, it may include models of the engine's fuel tanks, etc., if the representation of these objects makes it easier for the supervisor to understand the image.

[0035] On the other hand, possibly in the computer-generated image the device may be incomplete (or, equivalently, parts of the device may be made transparent on the screen).

[0036] In other cases, the device can be positioned in the virtual scene differently from reality; for example in an 'exploded' position, in order to allow a more readable representation of the constitution of the device.

[0037] Furthermore, the computer-generated image is generally enriched by additional information that is superimposed on the visual rendering of the digital model of the device. This additional information can notably vary in real time. This additional information is normally integrated into the computer-generated image in the vicinity of the point in the image to which it relates.

[0038] This additional information may be a representation, positioned on the device itself, of the field of variations of certain parameters of the device.

[0039] This representation can be done in 'false colors', that is, with different colors conventionally assigned to different ranges of values ​​of the parameter represented.

[0040] This additional information may finally include symbolic information, particularly verbal.

[0041] Thanks to the additional information it includes, and / or thanks to the fact that it represents parts of the device not visible (or less clearly visible) by real cameras, the computer-generated image allows the supervisor to better understand the physical phenomena taking place in the device and consequently to understand more easily how it evolves. COMBINATION OF THE ACQUIRED IMAGE AND THE COMPUTED IMAGE

[0042] In the supervision image, the acquired image and the synthetic image can be combined in different ways.

[0043] These two images can be superimposed using partially transparent layers. It is also possible for the computer-generated image and the acquired image to be complementary; for example, in the supervision image, all pixels relating to a specific part of the device can come from the computer-generated image, while the other pixels come from the acquired image.

[0044] The supervision image can combine different layers corresponding respectively to the acquired image and the synthetic image, using transparency.

[0045] Naturally, the display process is all the more efficient as in the supervision image, the acquired image and the synthetic image combine in a realistic manner.

[0046] The fact that these two images combine realistically means that when these two images are combined, together they form an image in which the part of the device represented by the computer-generated image appears in a normal, or at least normal, relative position to the part of the device (or the rest of the device) represented by the image acquired by the camera.

[0047] It may happen that when implementing the method, the image initially calculated in step c) cannot be realistically combined with the image acquired by the camera.

[0048] To overcome this problem, in one embodiment of the method according to the invention, during step d) the synthetic image is modified and / or recalculated so that the acquired image and the synthetic image combine realistically to form the supervision image.

[0049] First of all, it should be noted that it is not essential that the acquired image and the synthetic image correspond perfectly. Advantageously, the supervision image retains its interest even if there are certain differences between the acquired image and the synthetic image.

[0050] These differences can first of all, if they are not very significant, be overcome by resizing the computer-generated image, that is to say by applying a dilation or contraction in its two directions, and possibly a rotation.

[0051] If the differences between the acquired image and the synthetic image are significant, this may be due to errors in the actual image acquisition parameters of the active camera.

[0052] In practice, the image parameters of cameras are known to within the measurement errors of these parameters. Therefore, there may be a discrepancy between the theoretical image parameters and the actual image parameters for the acquired image.

[0053] In this case, in step d) it is necessary to identify from the acquired image, the actual image parameters thereof; to update from the actual parameters of the acquired image, the synthetic image parameters; to recalculate the synthetic image taking into account the updated synthetic image parameters, and finally to recalculate the supervision image.

[0054] In the display process, a certain number of choices are generally left under the control of the supervisor: in particular, the choice of the active camera, the image of which is displayed on the display, and its parameters (position, orientation, optical parameters).

[0055] In some cases, it may be decided to display a supervision image in which the device is filmed with a camera position or orientation, or even camera optical parameters, which do not correspond to the current parameters of the different cameras.

[0056] In this case, the method for displaying a supervision image can be implemented in the following manner: in step a2) image parameters are determined which are not those of any of the real cameras at the instant considered; and before acquiring the image acquired in step b2), during a step b1) the parameters of the active camera are adapted so that they correspond substantially to the image parameters determined in step a2).

[0057] Thus, in this implementation mode, the image parameters are imposed on the camera so as to take the new desired values. In other words, the desired type of supervision image is chosen in advance (the specific position and orientation of the camera relative to the device, its zoom level, etc.); the active camera is then chosen according to this choice, and the active camera is positioned, oriented and adjusted according to this choice.

[0058] Furthermore, the way in which the acquired image and the computer-generated image are combined is generally specified by the supervisor during the implementation of the supervision system. For example, the supervisor may have at his disposal functions allowing him to display the computer-generated image, and in it, to make transparent (or invisible on the screen) the components of the device that he does not want to see, as well as functions for displaying specific parameters of the device.

[0059] Furthermore, in addition to these possibilities, the method for displaying a supervisory image in accordance with the invention has the ability to automatically modify the display in certain circumstances.

[0060] For this purpose, the monitoring system is informed of the evolution of the device thanks to step a0 of the method, during which one or more device status parameters are acquired. The monitoring system can, for example, receive device status parameters transmitted by a device monitoring system (called a health monitoring system), preferably periodically and in real time.

[0061] Following the acquisition of this or these state parameters and in accordance with the invention, at least one of the following operations involved in obtaining the supervision image is carried out as a function of the state parameter(s) of the device: in step a2), the image parameters are determined, and / or in step c), the synthetic image is calculated, and / or in step d), the supervision image is formed.

[0062] The device status parameter(s) are time-varying parameters. These parameters may be or include information about a device fault or malfunction. This information may be, for example, an alert or alarm message generated by the device's tracking system.

[0063] According to the invention, the image parameters are adjusted, and the synthetic image and the supervision image are recalculated, preferably in real time, after any variation in a state parameter of the device justifying a change in the display of the supervision system. Thus, the supervision manager is informed in real time, on the display, of any changes in the supervised device.

[0064] In particular, in step a2) the image parameters are modified, and in step c) the synthetic image is calculated, and in step d), the supervision image is formed again, when a state parameter of the device acquired in step a0) crosses a predetermined threshold.

[0065] The supervision image is therefore itself a function of the anomaly detected.

[0066] The invention also comprises a supervision system for supervising a device according to claim 4.

[0067] The image parameters for the supervision image are typically a single image parameter set, used both to acquire images in step b2), and to compute synthetic images in step c).

[0068] It may also be two separate sets of image parameters, namely a set of acquired image parameters and a set of synthetic image parameters, then used respectively for the acquisition of images in step b2) and for the formation of synthetic images in step c).

[0069] It is understood that the determination means are capable in the latter case of defining acquired image and synthetic image parameters which are compatible, that is to say which allow that in step d), the acquired image and the synthetic image produced on the basis of these parameters can be combined in a realistic manner.

[0070] Equivalently to points b2), c) and d) above, it can be stated that the computing unit is configured to: b2) acquiring images from the active camera, according to the image parameters determined by the determination means; c) calculating a synthetic image of the device or of a part of the device such as could be provided by a virtual camera taking into account the image parameters determined by the determination means; d) forming an image called a supervision image from the acquired image and the synthetic image in such a way that the acquired image and the synthetic image combine realistically to form the supervision image.

[0071] The following improvements may also be adopted, individually or in combination: the acquisition means may be capable of adjusting at least one parameter of at least one real camera so as to make said parameter equal to an image parameter determined by the calculation unit. the calculation and / or image formation means may be capable of modifying and / or recalculating the synthetic image as a function of the acquired image so that the acquired image and the image thus modified and / or recalculated combine realistically in step d) to form the supervision image. in the supervision system, the monitoring means may be anomaly detection means; the determination means may be capable of determining the image parameters (in step a2) as a function of an anomaly detected on the device.The monitoring means may in this case comprise anomaly detection means comprising an anomaly interpreter, capable of selecting a main anomaly information item from a large number of available anomaly information items, to calculate the modified image as a function of this main anomaly information item. The calculation means may be capable of calculating the synthetic image in step c) as a function of an anomaly detected on the device.

[0072] Within the scope of the invention, a computer program is also proposed comprising instructions for executing the steps of the display method as defined previously, when said program is executed by a computer connected to at least one real camera arranged so as to be able to film all or part of the device.

[0073] Within the scope of the invention, there is also provided a recording medium readable by a computer on which is recorded a computer program comprising instructions for executing the steps of the display method as defined previously when said program is executed by a computer connected to at least one real camera arranged so as to be able to film all or part of the device.

[0074] The invention will be better understood and its advantages will appear better on reading the detailed description which follows, of embodiments shown as non-limiting examples. The description refers to the appended drawings, in which: there figure 1 is a schematic view of a supervision system according to the invention; the figure 2 is a diagram illustrating the steps of the method according to the invention, in one embodiment; the figure 3is a schematic view of a virtual scene used for image calculation, when implementing the method according to the invention; and the figure 4 is a schematic view illustrating the formation of the supervision image from an acquired image and a synthetic image.

[0075] A supervision system 10 according to the invention, for the supervision of tests carried out on a rocket engine 20, is shown schematically in the figure 1 .

[0076] This supervision system 10 allows a supervision manager, in charge of the tests, to control the proper conduct of tests planned for the engine 20 from a control station.

[0077] Engine 20 is a test rocket engine, comprising a nozzle 21, subjected to operational tests in a conventional manner. It is placed for this purpose on a test bench not shown.

[0078] The engine 20 is equipped with various sensors 22 which constitute means for monitoring the rocket engine (or device) 20. The sensors 22 measure engine state parameters such as, for example, temperatures, pressures, accelerations, etc.

[0079] The engine 20 is also placed under the surveillance of different cameras 24.

[0080] The sensors 22 are connected to a computer or electronic control unit ('ECU') 26. The function of this computer is to continuously monitor, in real time, the evolution of the engine 20 during the operating tests. The computer 26 thus constitutes a "health monitoring system" of the engine 20. Such a system is described for example by French patent No. FR2956701.

[0081] The supervision system 10 comprises a calculation unit 12, to which a screen (or display) 14 and a keyboard 16 are connected. The supervision system further comprises a second control screen 14' and a second keyboard 16'.

[0082] The calculation unit 12 is designed to execute a display computer program, allowing the display on the screen 14 of supervision images of the engine 20. For this purpose, the calculation unit 12 comprises determination means 12a, capable of determining the active camera and the image parameters; acquisition means 12b, for acquiring the images from the cameras, calculation means 12c for calculating synthetic images of the device, and finally image formation means 12d, for forming the supervision images from the images provided by the acquisition means and the calculation means.

[0083] The acquisition means 12b are also capable of transmitting to the cameras (and to the positioning means which allow their positioning and orientation) desired position parameters and optical parameters. They therefore also constitute means for adjusting camera parameters. INITIALIZING THE DISPLAY PROGRAM Creation of the 3D scene

[0084] The display program of the computing unit 12 uses as data a three-dimensional digital model 20' of the engine 20. When the program is initialized, a virtual scene is defined in which the digital model 20' of the engine 20 is arranged. Virtual cameras 24', having optical (virtual) characteristics identical to those of the real cameras 24, are positioned in the virtual scene. The virtual scene thus obtained is illustrated by the figure 4 . Program settings

[0085] Additionally, a set of predefined behaviors are programmed.

[0086] Each predefined behavior includes: a condition or criterion, which is a function of state parameters measured by the sensors 22 (and possibly other quantities such as time, etc.), and the image parameter values ​​to be used if the condition or criterion thus defined is satisfied.

[0087] Thus, when the system is operated, as soon as the predefined condition or criterion is satisfied, the display and possibly the active camera immediately adopt the predefined image parameters for these conditions. This allows the display most suited to the circumstances to be immediately displayed to the supervisor.

[0088] The supervision system 10 can for example be programmed to have, in the event of detection of an increase in the vibration level of an engine bearing, the following predefined behavior: the image parameters are adjusted so that the active camera is oriented towards the location of the bearing and acquires an image with the maximum magnification factor (zoom); the camera closest to the bearing is chosen as the active camera, brought closer to the bearing, oriented towards it; its magnification factor is made maximum; an image is then acquired by this camera; the calculation means 12c calculate a synthetic image of the bearing, in which the bearing is represented with a specific color; the image formation means 12d combine this synthetic image of the bearing with the acquired image representing the rest of the engine, so that the supervision image obtained shows the synthetic image of the bearing within the overall image of the engine.When forming the supervision image, the synthetic image is placed on a layer placed in front of the layer containing the image acquired by the camera, so that in the supervision image, the synthetic image remains fully visible.

[0089] The supervision system 10 can for example also be programmed to have the following predefined behavior.

[0090] The computing means can, using theoretical digital models, calculate in real time the variations of certain parameters in the engine, based on available measurements, for example: Energy dissipation; heat transfers.

[0091] As a result, the calculation means integrate into the computer-generated image a colorized representation of the field of variations of the parameter thus calculated. This colorized representation can be in the form of a color gradient although there is only a point instrumentation and a limited number. EXECUTION OF THE DISPLAY PROGRAM

[0092] The main steps of the display program are illustrated by the figure 2 .

[0093] The display program is an iterative program, which continuously executes steps S0) to e) described below in a loop, thus making it possible to update the supervision image each time step e) is carried out. Step a0 Updating supervision information

[0094] The initial step a0 of this program consists first of all in acquiring updated values ​​of the state parameters of the engine 20. This information is transmitted to the calculation unit 12 in real time by the computer 26. This information is acquired by means 12a for determining image parameters of the calculation unit 12.

[0095] This information includes engine parameter values, and / or alert messages from the computer 26 when abnormal values ​​are recorded for certain parameters.

[0096] Step a0 of the program also consists of acquiring values ​​or instructions specified by the supervisor, for example the choice of the active camera, or of certain image parameters for the displayed image. The specified parameters may in particular be acquired image parameters influencing the active camera, and / or synthetic image parameters influencing the synthetic image to be integrated into the displayed image. Steps a1) and a2) Determination of the active camera and image parameters

[0097] The first processing steps a1) and a2) are carried out by the determination means 12a of the calculation unit 12. These determination means determine the active camera and determine the acquired image and synthetic image parameters. This step therefore consists, with a view to forming a supervision image which will then be displayed on the screen 14, in determining: among the different cameras 24, the active camera, which will provide a photo of the device or a part thereof; the acquired image parameters that the image acquired by the active camera must have, namely the position and the shooting orientation of the latter relative to the engine 20, and the optical parameters that the camera must take, for example the zoom factor; and the synthetic image parameters that the synthetic image must have, namely the position and the shooting orientation of the latter relative to the digital model of the engine 20, the specific position(s) of the latter in space (which may differ from the position of the corresponding real parts), and the optical parameters that the virtual camera must take to produce an image of the virtual scene, such as for example the zoom factor.

[0098] Most often, the acquired image parameters are identical to the synthetic image parameters. However, the program executed by the system 10 may be provided so that the image parameters are different for the acquired image and for the synthetic image. This situation may arise, for example, in the case where the desired supervision image is an overall image of the engine 20, but it is desired to insert into this overall image a synthetic image representing only one component of the engine, for example a small one. The image parameters of the synthetic image are then specified such that the synthetic image represents only the component in question.

[0099] On the other hand, most often, the determination means use as image parameters (acquired and synthesis) the image parameters of the active camera 24 used previously. The supervisor who operates the supervision system 10 specifies the active camera by default (used when starting the system); moreover, as indicated previously, he may at step S0 have specified a new camera as the active camera, or have modified the image parameters.

[0100] Thus, taking into account the instructions of the supervisor, the determination means define the parameters for the acquired and synthetic images.

[0101] A particularly interesting property of the supervision system 10 is its behavior in the event of a malfunction of the engine 20 being tested.

[0102] If a specific malfunction message (part of a predetermined list) is transmitted to the unit 12 by the computer 26, or if certain parameters of the engine 20 exceed certain predetermined values, in step a2) the program executed by the calculation unit 12 itself automatically defines new image parameters.

[0103] For example, if certain specific malfunction messages concerning a critical component of the engine 20 are received, the determination means automatically take into account (without intervention by the supervisor) image parameters which make it possible to specifically display on the screen the component experiencing the malfunction.

[0104] The acquired image parameters are thus adjusted, for example, so that the active camera is closest to the faulty component experiencing the malfunction; the camera is then positioned as close as possible to this component and oriented towards it. The zoom factor (the only adjustable optical parameter on the 24 cameras) is adjusted so that the faulty component is visible in the best conditions on the active camera image. Step b) Image acquisition

[0105] This step has two sub-steps. b1) Adjustment of the active camera

[0106] After the image parameters have been determined (step a2), in case the parameters of the active camera do not match the acquired image parameters, the active camera is adjusted so that its parameters match the desired acquired image parameters determined by the determining means 12a.

[0107] In the example presented, the cameras 24 are movable in translation along the vertical axis on racks 28. They are also orientable by rotation around a horizontal axis (arrows R).

[0108] Also, to adjust the parameters of the active camera 24, the acquisition means 12b (functioning as camera parameter adaptation means) transmit to the active camera 24 and to the positioning means thereof the desired acquired image parameters. The positioning means then position and orient this camera, and the latter adjusts its internal parameters (in particular the zoom factor), in accordance with the transmitted parameters. Step b2) Image acquisition acquired

[0109] Once the camera is positioned and adjusted to the acquired image parameters (if applicable), the acquisition means 12b proceed to acquire an image from the active camera 24. Step c) Calculation of the synthetic image

[0110] On the other hand, the system 10 calculates the synthetic image following the synthetic image parameters. This image represents a part of the engine 20 - or possibly the entire engine 20 -.

[0111] The characteristics of the computer-generated image are generally predetermined, or specified by the supervisor based on the parameters he wishes to monitor.

[0112] Thus, a default display is defined for each of the engine components. This default display can include the representation of certain parameters; for example, for the engine nozzle, the vibration rate is displayed by default.

[0113] The supervisor may also request the display of specific parameters.

[0114] However, the characteristics of the synthetic image may also, under certain circumstances, be defined based on the state information of the engine 20.

[0115] Thus, in the event that certain predetermined malfunction messages are issued, or if certain parameters exceed certain predetermined values, the calculation means 12c take into account specific characteristics predetermined for the synthetic image. These specific characteristics are of different natures; they may be: To make transparent certain components through which one wishes to be able to see (visible / non-visible component); To position certain components in specific positions, so that the view obtained is more meaningful (for example, one can specify that the engine components are positioned in 'exploded' view so that the internal parts of the engine are visible); To display values ​​of the parameters that are considered the most important for the supervisor, if he receives the malfunction message in question. The parameter values ​​can be displayed in digital form, or more advantageously with conventional false colors, on the engine itself (or only on some of its components); To represent the faulty component(s) according to certain colors having a predetermined conventional meaning, possibly flashing.

[0116] For example, if the temperature at a point on the nozzle exceeds a predetermined value, the calculation means modify the characteristics of the synthetic image so that the temperature information on the nozzle is displayed, i.e. the synthetic image is calculated so as to represent the variations in the temperature field on the nozzle, as shown in figure 4 (This figure shows isothermal curves surrounding a hot spot P of the nozzle). Step d) Formation of the supervision image

[0117] Once the acquired image and the synthetic image have been obtained, the 12d image forming means of the unit 12 form the supervision image.

[0118] Most often, the parameters of the acquired image and the computer-generated image match, so that the computer-generated image can be easily combined with the acquired image; for example, because the image parameters are exactly the same.

[0119] However, it may happen that the combination of the acquired image and the synthetic image does not produce a realistic image.

[0120] After combining the two images, the display program evaluates whether the combination is satisfactory. This evaluation can be done by extracting edges from both images, and evaluating the correspondence of edges between the two images.

[0121] If the combination of the acquired image and the synthetic image is satisfactory, the supervision image obtained is sent to screen 14.

[0122] Otherwise, the supervision image is recalculated. This recalculation can be done in different ways.

[0123] If the differences are minimal, basic image processing such as applying scale factors ('Scale' functions) in both directions is applied to the synthetic image, so as to better match it to the acquired image. Only step d) of determining the supervision image by combining the acquired and synthetic images is done again.

[0124] If the differences are greater, the synthetic image is recalculated. The computing unit then determines new synthetic image parameters, closer to the actual image parameters of the acquired image (step a2).

[0125] These new computer-generated image parameters can be calculated in different ways.

[0126] They can first be calculated from the acquired image itself. For this purpose, targets (for example circular patches made of retroreflective material) can be placed on the engine 20, and the calculation unit comprises means for determining the image characteristics of an image simply from the positions of the targets appearing in the image.

[0127] Alternatively, the synthetic image parameters may be recalculated from the acquired image, without the need to integrate targets into the scene. For this purpose, the computing unit may comprise means for determining the image characteristics of an image by performing contour extraction in the acquired image, and applying a shape recognition algorithm to the contours thus extracted.

[0128] When the new synthetic image parameters have been set, the calculation means 12c again calculate a synthetic image on the basis of these new parameters (step c). The supervision image is then formed by combining the acquired image and the new synthetic image (step d).

[0129] If the combination of the acquired image with the new synthetic image is satisfactory, the new supervision image is then transmitted to the screen 14. Otherwise, the step of improving the synthetic image is carried out again.

[0130] An example of the formation of a supervision image appears on the figure 4 . This represents on the one hand an image 50 acquired by a camera 24; this image represents the entire engine 20.

[0131] There figure 4also presents a computer-generated image 52. On this image is represented only the nozzle 21 of the engine. This nozzle is represented by means of the digital model 21' of the nozzle. On the digital model 21' are represented the temperature variations on the surface of the nozzle, in the form of isothermal curves.

[0132] There figure 4 finally shows the supervision image 54 formed by combining the acquired image 50 and the synthetic image 52. In image 54, the representation of the nozzle comes from the synthetic image 52 (and therefore from the digital model 21', on which the isothermal curves are represented); all the rest of the image comes from image 50 acquired by the camera.

[0133] In a variant of the display program, the determination means systematically determine the synthetic image parameters from the acquired image, for example by using targets as indicated previously.

[0134] The computing unit 12 operates in real time, i.e. all of the steps a1) to e) are carried out in real time.

Claims

1. A method for displaying a supervision image for the supervision of a device (20) by means of a supervision system (10), the supervision system comprising a display (14) and at least one real camera (24) arranged so as to be able to film all or part of the device; the method comprising the following steps: a1) a camera called an active camera is selected from several cameras, if the system comprises several cameras; a2) two sets of image parameters are determined, specifically a set of acquired image parameters and a set of synthesis image parameters, each set of image parameters comprising a relative camera position and orientation relative to the device, and one or more optical parameter(s) of camera; b2) an acquired image (50) is acquired showing all or part of the device by means of the active camera, according to the acquired image parameters; c) a synthesis image (52) of the device or part of the device is calculated such as a virtual camera could provide taking into account the synthesis image parameters; d) an image called supervision image (54) is formed from the acquired image and the synthesis image such that the acquired image and the synthesis image are combined realistically to form the supervision image; and e) the supervision image is displayed on the display; the method being characterized in that: it further comprises a step a0) during which at least one state parameter of the device is acquired; and in that at step a2) the image parameters are determined as a function of said at least one state parameter of the device, and are modified when said at least one state parameter of the device acquired at step a0) exceeds a predetermined threshold, the supervision image being formed again.

2. The display method according to claim 1, wherein at step a2) image parameters which are not those of any one of the real cameras are determined at the relevant instant; and prior to acquiring the acquired image at step b2), during a step b1) the parameters of the active camera are adapted such that they correspond substantially to the image parameters determined at step a2).

3. The display method according to claim 1 or 2, wherein the synthesis image is modified and / or recalculated as a function of the acquired image such that at step d) the acquired image and the synthesis image are combined realistically to form the supervision image.

4. A supervision system (10) for the supervision of a device, comprising a computing unit (12), a display (14), and at least one camera (24); supervision system in which the computing unit comprises: a) determination means (12a), capable of determining image parameters for an image (54) called supervision image, said image parameters comprising a relative camera position and orientation relative to a device filmed by said at least one camera, and one or more optical parameter(s) of camera; and making the choice of a camera called an active camera from several cameras, when the system comprises a plurality of cameras; b2) acquisition means (12b) capable of acquiring images of the active camera according to the image parameters determined by the determination means; c) calculation means (12c) capable of calculating a synthesis image of the device or part of the device such that a virtual camera could provide taking into account the image parameters determined by the determination means; d) image formation means (12d) capable of forming said supervision image from an acquired image (50) and a synthesis image (52) such that the acquired image and the synthesis image are combined realistically to form the supervision image (54); and wherein the display is capable of (e) displaying the supervision image; the system being characterized in that it further comprises monitoring means (22) of the device, capable of acquiring at least one state parameter of the device; and in that the determination means are capable of determining the image parameters as a function of said at least one state parameter of the device, and are capable of modifying the image parameters when said at least one state parameter of the device acquired exceeds a predetermined threshold, the image formation means being capable of forming the supervision image again.

5. The supervision system according to claim 4, wherein the acquisition means (12b) are capable of adjusting at least one parameter of at least one real camera (24) so as to render said parameter equal to an image parameter determined by the computing unit.

6. The supervision system according to claim 4 or 5, wherein the calculation and / or image formation means are capable of modifying and / or recalculating the synthesis image as a function of the acquired image such that the acquired image and the now modified and / or recalculated image are combined realistically at step d) to form the supervision image.

7. A computer program comprising instructions for execution of the steps of the display method according to any one of claims 1 to 3 when said program is executed by a computer connected to at least one real camera (24) arranged so as to be able to film all or part of the device (20).

8. A recording medium readable by a computer on which is recorded a computer program comprising instructions for execution of the steps of the display method according to any one of claims 1 to 3 when said program is executed by a computer connected to at least one real camera (24) arranged so as to be able to film all or part of the device (20).