FR2584811A1 - Device for measuring deformations of a body, in particular a wind tunnel model - Google Patents

Abstract

A body C is provided with emitting or reflecting luminous points P1, P2, P3. An optical system casing 10 includes an input slot, followed by an optical system ending in two cylindrical lenses 14 and 24 which interact with photodetector arrays 15 and 25 which act in perpendicular directions. The signals from the arrays 15 and 25, which are of the charge transfer diode type, are gathered in a line SPD in order to be applied to a processing device DTS, calculating the centre of gravity CB for the various luminous spots detected by the arrays.

Description

Device for measuring deformations of a body, in particular a model in an aerodynamic wind tunnel.

The invention relates to the use of optoelectronic devices to perform, remotely and without contact, deformation measurements on a body.

The measurement of deformations is essential in particular when studying the behavior of a body subjected to mechanical stresses: wind behavior of engineering structures, such as bridges, buildings, cooling towers of power plants large dimensions; study of the behavior of satellite structures in a state of weightlessness, or in a simulated thermal environment; study of models, in particular of aircraft models, in an aerodynamic wind tunnel, application to which we will return later.

Several solutions have already been proposed for determining the deformation of a body.

It has in particular been envisaged to use imaging systems, which use cameras to film the deformation in real time. These cameras have two-dimensional photosensitive surfaces whose resolutions are 625 x 625 dots for a Vidicon camera or 500 x 500 dots for a detector camera in charge transfer circuits (whose short name is
DTC or CCD from the English Charge Coupled Device).

This insufficient resolution limits the precision of the results obtained, and such cameras do not allow very small displacements or micro-deformations to be measured. Other problems arise with the duration of the measurement, because of the essentially sequential functioning of these cameras.

We have also proposed in the French Patent Application published under No. 2 255 611 a device for stereoscopic observation of a body, making it possible to analyze its displacements and deformations. In this device, we scan line by line the surface to be studied, by means of a light brush, and the light backscattered by retroreflectors arranged on the body is captured, in order to calculate the displacements of the retroreflectors resulting from the deformation.

The system constructed in this way is clearly complex; it also has the disadvantage of using several mobile optical devices, including a mirror rotating at high speed. In addition, the resolution obtained is not exceptionally good, and above all the analysis of the surface of the body to be studied is sequential. This again poses a problem of measurement time, that is to say of temporal resolution, note that a mechanical scanning of an image cannot be very fast.
To fix ideas on an example, we will now return to the problem of measuring deformations undergone by an aircraft model in an aerodynamic wind tunnel.

In such an application, the volume to be analyzed fits into a cube of 4 meters on each side. This cube must be analyzed with a recoil limited to about 4 meters. It immediately follows that the optical system used must be of short focal length. This latter constraint in turn means that it is not possible in practice to use several juxtaposed charge transfer photodetector circuits.

In addition, the required resolution is 20,000 points in each dimension. On the other hand, it is necessary that the analysis can be done up to frequencies of the order of a kilohertz, because the model is generally subjected to different vibrational phenomena of a hundred hertz. Finally, it is necessary that the measurements made at different points of the model can be considered to be simultaneous, thus constituting each time a kind of instantaneous photograph of the model.

This presentation of a particular problem with which the Applicant was confronted clearly shows the difficulties that can be encountered during deformation measurements.

More generally, the present invention aims to provide an opto-electronic device for dimensional measurements at a distance which has high performance, both in terms of spatial resolution and measurement time.

One of the aims of the invention is thus to provide an elementary spatial resolution of several thousand points, a resolution which can be increased by a factor of 10 by specific processing methods of the invention.

Another aim of the invention is to obtain such a result from a single integrated circuit forming a photodetector member.

The invention also aims to carry out measurements of the aforementioned type which can, while being practically simultaneous, remain valid in the presence of vibratory phenomena up to frequencies of the order of a hundred hertz.

Yet another object of the invention is to provide a device as defined above, which is capable of operating with an optic of short focal length, therefore of great depth of field.

Of course, it is required that these aims be achieved by using purely electronic means, to the exclusion in particular of any mechanical scanning by prism or rotating mirror.

The opto-electronic device of which the invention makes use is of the type comprising - means defining on the body to be measured a series of light points (by emission or by reflection), - photodetector means with discrete elements, placed at a distance from the body , and arranged, in combination with optical means, to perceive said light points whatever their position on the body, as well as to deliver at least two digital angular information in correspondence of each discrete element perceiving a light point; and - digital electronic means suitable for processing this digital information in order to determine the deformations of the body.

The invention firstly provides for making use of an initial phase of individual acquisition of the positions of the various light points of the body, followed by a tracking phase during which these points are treated together.

The photodetector means preferably used are two linear arrays of photosensitive elements which are preferably of size smaller than that of the spot created on them by the light points. The two bars are arranged in the focal plane of the optical means and oriented in two substantially perpendicular directions. As for them, the associated optical means are means of cross anamorphosis transforming the image of each light point into two segments of straight lines substantially orthogonal whose respective positions on the bars are linked to the spatial coordinates of the light points on the model.

Finally, for their part, the digital electronic means comprise means of pretreatment capable of carrying out, simultaneously on each strip, a determination of the barycenters of each of the different lighting spots perceived by this strip; and. processing means with tracking allowing continuous monitoring of the movements of each of the light points of the body, from the barycenters thus identified on each of the bars.

The initial acquisition phase can be carried out using a sequential ignition of the various light points of the body. It can also take into account a priori knowledge of the positions of the light points, in a chosen state of the body. This is made immobile, during the initial phase, as much as possible.

For the tracking phase, the light points are normally on simultaneously, preferably permanently.

According to another aspect of the invention, the digital electronic preprocessing means are of the wired type. They understand . means for determining space windows respectively centered on each of the lighting spots, and. means for determining the barycenter of the light intensities felt by the different photodetector elements of each window.

 It is advantageous to eliminate the windows of width less than that covered by a predetermined number of photodetector elements adjacent to each other.

As far as they are concerned, the optical means of cross anamorphosis are at least partially made up of two optical devices of the cylindrical or spherical cylindrical type, in particular of the cylindrical double Gauss type, at least for certain applications.

In a particular embodiment, the optical means of cross anamorphosis comprise two devices each consisting of: - a first group of two lenses, at least one of which has a cylindrical entry face; the first lens being convergent and the second diverging; - a pupil conjugated with the entrance pupil by the first group of lenses; -a converging, concave-convex lens; and a group of output lenses consisting of at least one converging lens having a cylindrical output face and preferably two such lenses.

A preferred embodiment of these optical anamorphic cross means will be defined below.

In an application such as that of aircraft models in the wind tunnel, the'corpus to be measured, that is to say the model, is cooperative. It is then possible to define the light points by light-emitting diodes, or even by the ends of optical fibers also coupled to a light source such as a laser.

For other applications, the light points will be defined by reflection on the body, in particular using retroreflectors, which will receive a light beam produced by a source associated with the measurement device and preferably placed in the immediate vicinity of the optical means d cross anamorphosis.

By suitably arranging two devices according to the invention, a three-dimensional measurement can be carried out in real time.

Other characteristics and advantages of the invention will appear on examining the detailed description below, and the appended drawings, in which - Figure 1 is a block diagram of a first embodiment of the invention; - Figure 2 is a block diagram of a second embodiment of the invention; - Figure 3 is a block diagram of a preferred embodiment of the invention, applicable to aircraft models in an aerodynamic wind tunnel; - Figure 4 is an exploded diagram of the cross anamorphic lens incorporated in the device of Figure 3; - Figures 4A and 4B are views in two perpendicular planes, of the optical characteristics of the elements of Figure 4; - Figure 4C is a table giving the numerical values associated with the device of Figures 4A and 4B ;; - Figures 5A and 5B are layout diagrams of the components inside an optical box of the device; - Figure 6 is the block diagram of the electronic part of the device according to the invention; and - Figure 7 is the detailed diagram of part of the electronics of Figure 6.

The accompanying drawings contain a lot of information of a certain nature. They are incorporated into the description not only to supplement it, but also to help define the invention to the extent necessary.

In Figure 1, we have designated by C a body whose displacements and especially the deformations we want to measure.

This body, assumed here to be cooperative, is provided with light-emitting diodes such as P1, P2 and P3.

The radiation from one of these diodes, for example P1, is perceived by an optical box BO, with which is associated a device for processing DTS signals.

The optical box BO first of all has an entry slot 10, followed by an objective 11. A semi-reflective strip 12 subjects the light radiation coming from the points such as P1 of the body to two different paths.

The direct path passes through a blade 13 forming a filter at 0.9 micrometer, and followed by a cylindrical lens, for example a convex plane 14 which forms an image of the point P1 on one or, preferably, several elements of a photodetector array. 15. The lens 14 performs an anamorphosis of the image transforming each light point into a line segment so that an image is always formed on the bar 15 regardless of the position of the point P1, within the exploration domain .

From the blade 12, the other path includes a deflection plane mirror 22, followed by a filter 23 similar to the filter 13, then by a cylindrical lens 24 disposed in the direction perpendicular to the lens 14. The cylindrical lens 24 forms an image of the point P1 on one or, better, several elements of a photodetector array 25, which is perpendicular to the array 15. (The perpendicularity is understood here in the optical sense, that is to say account given the various references that may have taken place previously.)
In this example, the bars 15 and 25 are solid state linear multi-element detectors, also called charge transfer detectors, which each have 2,048 sensors.

The output signals of the bars 15 and 25 are grouped together in a transmission line SPD, which terminates in the device for processing the DTS signals. This includes an interface, if necessary with memory, marked IM. A sequencer SQ defines the clock signals specific to the dialogue of the interface IM with the bars 15 and 25. A calculating member CB determines the barycenter of the spots, as will be seen below.

Figure 2 has a structure similar to Figure 1, and will not be described in detail.

It is simply noted that the two measurement channels are completely independent. Thus, the optical box BO comprises two rectangular entry skylights 10 and 20, followed by two objectives 11 and 21, then filters 13 and 23 already mentioned. The rest follows as before.

Figure 3 is an interesting variant, particularly applicable to aircraft models in an aerodynamic wind tunnel.

Such a wind tunnel, whose outline is shown schematically in
SA, internally receives a MA model.

This is mounted on a support SM, crossed by a main bundle of FOP optical fibers.

Inside the model, the optical fibers are distributed in FO, to reach different points that one wishes to locate in a characteristic way. This has been schematized here by taking points
P1, P2 and P3 at the ends of the wings and at the front of the aircraft. Of course, practice can involve a much higher number of points.

The main optical fiber FOP receives, through an optic forming a condenser OC, the radiation from a light source SL, such as a laser for example.

Through its support SL, the model is movable vertically in the direction VT, as well as in yaw according to the curved arrow LA, or in roll according to the curved arrow RO.

A porthole F formed in the wall of the aerodynamic wind tunnel SA allows an optical box BO to observe at least part of the light points carried by the model.

This box is constituted as before, using two bars 15 and 25, but it is noted that the other elements are here grouped in two spherocylindrical objectives denoted respectively 110 for the measurement channel in direction X and 120 for the channel of measurement in direction Y. It is assumed that the bars 15 and 25 are respectively loaded with measurements in the directions X and Y.

As before, the SPD output signals are applied to the DTS signal processing device.

The detail of a sphero-cylindrical objective is illustrated in FIG. 4. It consists of six elements 121 to 126, among which there are four cylindrical lenses 121, 122, 125 and 126, the radii of curvature of which are chosen to reduce field aberrations up to an angle of incidence of the light rays of 60 ". The other lenses are spherical. The image of the diaphragm 123 by the lenses 121 and 122 constitutes the entrance pupil of the optical device.

Figure 4A illustrates a sectional view of the device of Figure 4, in the sense that all the lenses are active.

Figure 4B; which is a sectional view in the perpendicular direction, shows that the entry faces of the lenses 121 and 122 are planar. It also shows that the entry face of the lens 125 is planar, while it is the exit face of the lens 126 which is planar.

In FIG. 4A, we see the photodetector strip, here 25, in its entire length, in the image plane 127 of the cross anamorphosis optic. In FIG. 4B, we only see the trace of the bar 25 in section in the same plane 127.

FIG. 4C gives the detailed characteristics of the cross anamorphic optics used preferentially according to the invention. This figure starts from the numerical references already given to the lenses. Each lens is broken down into its input diopter and its output diopter: the column NO, which defines the number of diopters, thus assigns dioptres 1 and 2 to the first lens 121. Of course, the diaphragm 123 and the plane image 127 are considered to be unique diopters.

The second column of FIG. 4C each time identifies the nature of the surface of the diopter: the code 20 indicates that it is a cylindrical surface, the code 1 that it is a surface of revolution , and the code 0 that it is a flat surface, which is the case of the diaphragm 123 and the exit plane 127.

The RCO column indicates the curvature in mm. The column
SEPN indicates the thickness of the diopter located before the surface concerned, in millimeters.

Columns IND1 to IND3 show the refractive index of the optical medium located in front of the surface concerned, respectively for the wavelengths 0.000824 mm, 0.000850 mm, and 0.000800 mm.

The DISP column indicates the dispersion of the material constituting the lens.

The HMAX column indicates the height of the optical element constituting the lens.

Finally, the NAT column indicates each time the type of material, which is here either air or a glass with sulfur hexafluorre.

The overall focal length of this cross anamorphic lens is 20 mm.

Furthermore, for cylindrical diopters 1, 3, 9 and 11, four additional lines indicate the characteristics in the perpendicular plane.

These cross anamorphic optics can be mounted as indicated at 110 and 120 in FIG. 5A, in a side wall of an optical box BO. One of the optics works directly with the bar 15, placed on a support S15 (FIG. 5B).

The optic 120 works with the bar 25, by means of an angle gear by means of a mirror 22.

The bar 25 is fixed on a support S25.

The bars are connected by different supply lines and useful signals constituting the assembly
SPD to the electronic circuit of FIG. 6 which will now be described.

Conventionally, a box 200 supplies the bars 15 and 25 with the voltages and currents they need.

 I1 starts from the bar 15 a multiple link 201, and likewise from the bar 25 a multiple link 202, which goes to a block 210 responsible for controlling the bars ("leaders") using clock signals, all by performing a video adaptation.

For this, a frequency synthesizer 205 is provided, which controls a logic 206, which on the one hand develops different clock signals X '"T" "Ra' çRb intended for the bars and therefore passing to the block 210 by a bus 208. On the other hand, the logic 206 defines addresses of binary elements on the arrays by another bus 214 which goes to a block 220 responsible for the extraction of the barycenters of the light spots.

The connection between the blocks 210 and 220 is made by a bus 212, which separates the video signals relating to the elements of even rank and those relating to the elements of odd rank, as will be understood by those skilled in the art familiar with the operation of the DTC strips.

From block 220, an interface 280 processes digital signals of coordinates X. and Y., on 16 bits i i in parallel, in a bus 295, which can go to a microcomputer 290 with which a monitor 295 is associated.

The whole form of treatment with continuation.

The clock phases indicated above with respect to bus 208 have the following meaning - wX is the clock which regulates the integration time vdes
photodetectors, is the clock for reading photodetectors; - Ra is the clock for returning to the reference level of the diode for reading the packets of charges for the even channel; - <PRb is the reset clock to the reference level of the read diode for the odd channel.

 I1 is now referred to in FIG. 7.

The V2i + l and V inputs. respectively represent
2i odd and even video information, which arrives alternately on the outputs of bars 15 and 25.

It can also be seen that the clock generation block 206 will itself generate the clock phase signals and and vT already mentioned, and this under the control of a bus 214 forming an input / output channel.

The odd video information of the bars goes to a video adaptation circuit denoted 221A. The information of even rank goes on a video adaptation circuit denoted 221B. These circuits 221A and 221B are respectively followed by threshold circuits 223A and 223B.

In parallel with this, the outputs of blocks 221A and 221B also go to a circuit 230 forming a multiplexer, which processes the video level information coming from each of the points of the bars in order, on the one hand, to apply them to an analog / digital 232, on the other hand also transmit them to a circuit 225 forming a detector of the maximum of each spot, this circuit 225 also receiving the outputs of the threshold circuits 223A and 223B.

The stages 221 are intended to align the useful video signals with respect to the reference signals of the dark level, as defined in the threshold circuits 223.

The analog multiplexer 230 puts the signals of each pixel in series (which come in parallel for the even and odd ranks) before their digitization on six bits is done by the converter 232.

The video information coming from the adaptation circuits 221 are made discrete by the threshold circuits 223. After that, the maximum and spot detector circuit 225 uses analog comparators to detect the presence of spots, and indicate when we meet the pixel that defines the maximum of the spot. At this time, a memory circuit of the first in-first-out (FI-FO) type referenced 226 stores the address which is on the channel 214, which ensures the memorization of the position of the maximum of each spot.

A spot counter 228 is incremented each time a new lighting spot is encountered, a spot which is defined by a certain number of adjacent pixels, that is to say neighboring pixels, which are found to be exicted, taking account in particular threshold effects that were applied upstream.

Finally, there is a circuit 236 which is used to validate a calculation grid on either side of the maximum of the spot, in order to extract the barycenter from the latter.

The circuit 225 having defined by a pulse the fact that it has encountered a maximum pixel, we can summarize the operation by saying that the member 226 stores the address corresponding to this pixel, while the spot counter 228 is incremented by 'a unit and that the validation circuit 236 defines a segment on either side of the maximum, in terms of number of pixels to be taken into consideration.

This segment could be of fixed length. In practice, it is preferable to adjust it to take into account different characteristics of the spot, including the amplitude of its maximum pixel, and possibly the amplitude of the neighboring pixels.

The output of the analog converter 232 is applied to a digital delay circuit 234, which can be a buffer memory. This delay is intended to allow the detector circuit 225 to operate to detect a maximum. I1 is of course also intended to allow a return to the desired number of pixels, as defined by the calculation grid of the member 236: we cannot in fact know in advance what will be the maximum, and therefore determine whether or not the current pixel will intervene in a barycenter calculation.

Thus, the information digitized for the pixels belonging to each spot are selected by the validation circuit of the calculation grids 236, to be applied to calculation bodies produced in material form.

These calculating members firstly comprise a member 242 which performs a sum of multiplications of the form x. multiplied by yi, where yi is digital information
ii representing the light intensity perceived at each pixel, while x. is digital information representing the position of this pixel.

The calculation unit 242 is based on a multiplier / accumulator circuit operating on 16 bits. A memory 246 of the FIFO type, that is to say first-in-first-out, stores the information which will be called digitizers, for a reason which will be understood later.

The circuit 244 performs a summation of the form
This circuit can be a simple 16-bit accumulator circuit. Here again, its output is applied to a memory 248 of the FIFO type, which stores denominators.

At the end of reading of the bar concerned, which is one of the bars 15 and 25, a microprocessor 280, connected to the input-output channel 214, will first of all read the number of spots detected, as it is shown by counter 228.

Then, sequentially, for each task, it reads the address of the maximum in the circuit 226.

 I1 then acquires the numerator and the denominator associated with the grid for calculating the task concerned, in memories 246 and 248 respectively.

The microprocessor 280 is then able to calculate the barycenter of the task according to the formula barycenter = grid address x 16 + (numerator x 16)
denominator
Those skilled in the art will understand that the result is expressed in sixteenths of a pixel.

Typically, the calculation of a barycenter by the means just described takes about 90 microseconds.

Thus, for an application aiming to follow a dozen points in real time, it takes a total of approximately 2 milliseconds to read the entire DTC bar which is associated, and obtain the addresses of the ten spots with an accuracy of - + 1 / 16th pixel.

Those skilled in the art will also have understood that it is advantageous that the width of a light spot is greater than the size of the pixel, in order to increase the accuracy of the measurement.

Of course, it will also be possible to use the microprocessor 280 or else the members 221, 223 and 225 to omit parasitic points, defined for example by the fact that the length of the segment to which they belong is less than a configured length. It is thus possible to remove certain highlights from the image which are not due to the diodes or other light sources placed on the model.

The experiments carried out by the Applicant have shown that such filtering does not alter the precision of the calculation of the barycenter of the interesting spots.

In the above, we have referred to a single bar; of course, the same operations are carried out for the second bar, operating in the perpendicular direction.

For the application to aircraft models in an aerodynamic wind tunnel, only one bar is required for each direction of measurement. For other applications, it is conceivable to have, for each direction, a linear network of photosensitive elements consisting of several arrays of photodetectors in series, the continuity between the arrays being produced with optical elements of the semi-transparent plate type.

It will be understood that the arrangements which have just been described make it possible to carry out a two-dimensional analysis without mechanical means. A high resolution is reached in the field, since in the example chosen, the selected array having 2048 photo-elements, the gain is about a factor of 100 compared to the arrays of photodetectors also available. The signal processing which has just been described makes it possible to achieve a resolution much lower than the pixel.

Finally, the speed of analysis, which has also been called measurement time, obtained according to the invention, is much higher than what the prior art allows since, for example, ten points of the target can be located in less than 2 milliseconds.

Furthermore, the cross anamorphosis optic which has been described above is very advantageous in that it allows the implementation of the device of the invention in a field where it is necessary to operate with a wide field, up to 'to 70, while retaining acceptable aberrations.

What has just been described allows the simultaneous acquisition of a fairly large number of pixels.

 It is still necessary to be able to process these pixels simultaneously.

For this, the invention makes use of a separate microcomputer, illustrated at 290 in FIG. 6, and with which a display monitor 298 can be associated.

The computer 290 receives on a bus 295 the barycenter position information defined by the microprocessor 280.

In an initial phase, the target's light points are lit sequentially. This makes it possible to acquire the points one by one, at their two coordinates. We can of course use for this purpose any information known in advance on these points, which is particularly easy in the case of an aircraft model at rest in the aerodynamic wind tunnel.

Thereafter, one can operate by continuity to follow the pairs of coordinates X and Y relative to the different points, which allows a considerable simplification of the processing, and especially an acceleration of this one, compared to the case where it would be necessary for each times treat sequentially each of the light points that the target carries.

The invention thus allows an attitude restitution of the aircraft model (or of any other body which it is desired to measure). However, due to the high precision and the low response time that it provides, the invention also allows dynamic monitoring of the positions of the points, in order to access the deformations that this model or this body can undergo. And this is true even if we take into account relatively high frequency vibrations, which can reach a hundred hertz.

Up to now, a measurement made according to two coordinates has been considered. By using at the same time another device according to the invention, oriented in a different direction relative to the body, it is of course possible to make three-dimensional measurements thereon.

Of course, the invention is not limited to the embodiment described. On the contrary, it extends to any variant which results from the description and from the claims below.

Claims (15)

Claims. 1. Opto-electronic device for remotely contactlessly measuring deformations of a body, of the type comprising - means defining on the body (C) a series of light points (P1-P3); - photodetector means (BO) with discrete elements placed at a distance from the body, and arranged, in combination with optical means, to receive said light points whatever their position on the body, as well as to deliver at least two digital angular information (SPD) in correspondence of each discrete element perceiving a luminous point; and - digital electronic means, suitable for processing this digital information to determine the deformations; characterized, in combination - by the fact that, after an initial phase of individual acquisition of the positions of the various light points of the body, these are then treated together in a tracking phase; - by the fact that the photodetector means are two linear arrays of photodetector arrays (15, 25), arranged in the focal plane of the optical means and oriented in two substantially perpendicular directions, the size of each of the photosensitive elements being less than that of the spot created on them by the light points, the associated optical means (10-14; 20-24) being means of cross anamorphosis transforming the image of each of the light points into two substantially orthogonal line segments; - by the fact that the digital electronic means include. pre-treatment means (DTS) suitable for carrying out, simultaneously on each strip, a determination of barycenter in each of the different lighting spots perceived by this strip; and. tracking processing means (MTP) allowing the continuous monitoring of the movements of each of the light points of the body, from the barycenters noted on each of the bars.  2 Device according to claim 1, characterized in that the initial acquisition phase is carried out using a sequential lighting of the various light points of the body, while these are lit simultaneously, preferably permanently , during the chase phase. 3. Device according to one of claims 1 or 2, characterized in that the initial phase is carried out using a priori knowledge of the positions of the light points1 in a selected state of the body. 4. Device according to one of claims 1 to 3, characterized in that the body is immobile during the initial phase. 5. Device according to one of claims i to 4, characterized in that the digital electronic preprocessing means (DTS) are of the wired type. 6. Device according to one of claims 1 to 5, characterized in that the digital electronic preprocessing means comprise means for determining spatial windows respectively centered on each of the lighting spots, and means for determining the barycenter of the intensities of illumination felt by the different photodetector elements of each window. 7. Device according to claim 6, characterized by the elimination of the windows of the width less than that which covers a predetermined number of adjacent photodetector elements. 8. Device according to one of the preceding claims, characterized in that the optical means of cross anamorphosis at least partially consist of two devices of the cylindrical or sphero-cylindrical type (110 120). 9. Device according to claim 8, characterized in that the optical anamorphic cross means are of the cylindrical double-gauss type (110, 120). 10. Device according to claim 9, characterized in that the optical means of cross anamorphosis comprise two devices each consisting of: - a first group of two lenses, at least one of which has a cylindrical entry face; the first lens being convergent and the second diverging; - a pupil, conjugated with the entrance pupil by the first group of lenses; - a converging, concave-convex lens; and - a group of output lenses consisting of at least one converging lens having a cylindrical output face. 11. Device according to claim 10, characterized in that the outlet group comprises two converging lenses having a cylindrical outlet face. 12. Device according to claim 11, having substantially the characteristics of the table of Figure 4C. 13. Device according to one of the preceding claims, characterized in that, the body being cooperative, the light points are defined on the side of the body itself, by light-emitting diodes, preferably by ends of optical fibers coupled elsewhere. to a light source such as a laser (SL). 14. Device according to one of claims 1 to 12, characterized in that the light points are defined by reflection on the body, preferably by retroreflectors, of a light beam produced by a source associated with the device. 15. Device according to claim 14, characterized in that the source is placed in the immediate vicinity of the optical means of cross anamorphosis. i6. Device according to one of the preceding claims, characterized in that it comprises two sets of photodetector means and associated cross anamorphosis optics, which allows three-dimensional measurement in real time.