Deflectometry measurement method

The deflectometry method uses a display and camera system to capture fringe images in multiple configurations, calculating a single common solution for precise shape and position determination, addressing the limitations of existing methods and achieving interferometric accuracy without additional input data.

JP2024544771APending Publication Date: 2024-12-04ワイズ·ライト
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
JP2024533108
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Priority Date
2021-12-02
Filing Date
2022-11-28
Publication Date
2024-12-04

AI Technical Summary

Technical Problem

Existing deflectometry methods are unable to accurately measure the shape of objects with precision close to that of interferometric methods, are expensive, require additional input data, or are sensitive to errors, making them unsuitable for precise shape measurement of optical components like telescope mirrors or lenses.

Method used

A deflectometry method using a display and camera system that captures images of fringes in multiple configurations, calculates possible solutions, and identifies a single common solution to determine the object's shape and position, eliminating the need for additional input data and reducing measurement uncertainty.

Benefits of technology

The method achieves shape measurement accuracy comparable to interferometry with reduced uncertainty, using inexpensive components and avoiding approximation errors, suitable for measuring reflective and transparent objects with high precision.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2024544771000001_ABST
    Figure 2024544771000001_ABST
Patent Text Reader

Abstract

The present invention relates to a method for making measurements from a reflective or transparent object (2) by means of a deflectometer (1), the deflectometer (1) comprising a display (10) for displaying dynamic fringes, a camera (20) configured to acquire images of the fringes reflected by the reflective object (2) and / or transmitted through the transparent object (2), and a computer (30) configured to execute a program to perform measurements based on the images acquired by the camera (20), the method comprising: The method includes the steps of: successively acquiring images of fringes reflected and / or transmitted by the object (2) in at least two configurations of an optical system defined by the camera (20); executing a program based on the acquired images to calculate possible solutions for the shape of the object (2) and the position of the object (2) in the optical system defined by the display (10), the object (2) and the camera (20) in each of the configurations; and acquiring a final solution for the shape and position of the object (2) by identifying a single solution that is common to the previously calculated possible solutions.
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present invention relates to the technical field of optical measuring instruments, in particular to the field of deflectometers. [Background technology]

[0002] From the prior art it is known to use deflectometry as a means for measuring geometric defects. With reference to FIG. 1 of the present application, by displaying fringes on a display (10), it is possible to observe the reflection of these fringes produced by the reflective object (2) to be measured. By observing the deformation of the reflected fringes, the presence of geometric defects can be inferred from the geometric defects on the reflective surface of the object (2) to be measured.

[0003] The disadvantage of this method is that it does not allow to measure the shape of the object itself, but only defects of the shape of the object. Indeed, the tilt of the surface (which induces a deformation of the projected fringes) can only be quantitatively measured if the shape of the surface is already known. The prior art consists in considering the object approximately flat in order to estimate the tilt. Moreover, calculating the shape based on the tilt requires an integral calculation step, which depends on arbitrary constants. Therefore, the evaluation of the tilt by this approximation method provides an infinite number of solutions for the shape of the object.

[0004] Although the defects detected by this method can be of very small dimensions (often less than one micrometer), the measurement of shape is not quantitative: the measurement is purely qualitative, since only defects of shape are obtained.

[0005] Known improvements of this method make it possible to use deflectometry as a semi-quantitative measurement means. In this case, the measurement method uses additional input data (such as the distances between the camera, the object and the display and their mutual orientation angles). Thus, integral constants can be determined, which allow the measurement to provide the shape of the object. The disadvantage is that the measurements of the additional input data must be very small so that this method provides satisfactory results. The reason is that these data must be transformed so that they are represented in a single system ("alignment"), which requires very precise determination of the position of the auxiliary measuring instruments used.

[0006] Or, therefore, the measurement of the shape defects is applied to a known nominal shape of the object to be measured, which makes it possible to obtain the actual shape of the object to be measured. This method has the disadvantage of necessarily knowing the theoretical shape of the object to be measured in order to be able to obtain its actual shape, which induces a certain approximation in the quality of the measurement.

[0007] The last measurement method using deflectometry consists of making a geometric reconstruction of the object by triangulation. This method requires the mobility of the display, whereas in other measurement methods the display and the camera are fixed. With reference to FIG. 2, which illustrates the prior art: - starting from an incident ray (ri) that is known and captured by the camera (10), then by determining the optical path of an emitted light ray (re) defined by a first pixel (P1) displayed by the display (10) when the display (10) is in a first position (101) and a second pixel (P2) displayed by the display (10) when the display (10) is in a second position (102), Thus, the intersection point of the incident ray (ri) and the emitted ray (re) provides a point O of the object (2).

[0008] By performing such a geometric reconstruction for each of the camera's incident rays, it is possible to obtain each point of the surface of the reflective object to be measured, and thus its shape.

[0009] However, the disadvantage of this method over triangulation is that it gives a given result, the quality of which is mediocre, and inasmuch as the method is based on the calculation of the intersection points of light rays (extremely sensitive to the slightest errors in the propagation direction of these rays), the accuracy of the measurement is at most 10 micrometers. This approach does not take advantage of one of the main features of deflectometry, namely its inherent differential aspect resulting from its dependence on the tilt of the object to be measured.

[0010] Thus, the prior art methods are not adapted to measure objects, the knowledge of which is fundamental, of their shape to the nearest micrometer. For example, with respect to telescope mirrors or lenses, these deflectometry methods are therefore not suitable.

[0011] Interferometry is another measurement method (it makes it possible to obtain the shape of an object with an uncertainty much smaller than one micrometer) and is also a reference method for measuring the geometric quality of optical mirrors. However, interferometers are expensive instruments, very sensitive to vibrations, and it requires complementary elements to measure shapes other than flat or spherical. These elements themselves are often very expensive. These can be compensating parts (they are custom manufactured and are commonly called "nulls"), or computer-generated holograms.

[0012] Another optical measurement method is stereodeflectometry, in which a camera must acquire images of the object to be measured according to different angular configurations, which has the disadvantage of limiting the field that can be measured by the camera.

[0013] The last method is to measure using machines for measuring in three dimensions, which perform spatial measurements using sensors at multiple points of the object to be measured. These machines are on the one hand expensive and on the other hand very slow, since the spatial resolution of the shape to be measured is directly connected to the number of points to be measured. In addition, such machines operate in a regulated environment, especially by temperature, so that the measurements are correct. Summary of the Invention [Problem to be solved by the invention]

[0014] The object of the present invention is to overcome the disadvantages of the prior art by proposing a simple and inexpensive measurement method that makes it possible to obtain the shape of an object with a quality approaching that of the prior art interferometric methods.

[0015] Another object of the invention is to design a measuring instrument adapted for the implementation of such a method, which instrument should also be simple and cheap to use. [Means for solving the problem]

[0016] For this purpose, methods have been developed for carrying out measurements from reflective or transparent objects by deflectometry, which include: a display showing dynamic fringes; a camera configured to acquire images of the fringes reflected by a reflective object and / or transmitted through a transparent object; a computer configured to execute a program for carrying out measurements on the basis of images acquired by the camera; Includes.

[0017] According to the invention, the method comprises the steps of: - successively acquiring images of the fringes reflected and / or transmitted by the object in at least two configurations of the optical system defined by the display, the object and the camera; - executing a program and calculating, based on the acquired images, in each of the configurations possible solutions for the shape of the object and the position of the object in the optical system defined by the display, the object and the camera; obtaining a final solution for the shape and position of the object by identifying a single solution common to the previously calculated possible solutions; Includes.

[0018] The measurements performed in each of the configurations provide an infinite number of solutions due to the unknown integration constant. Of all these possible solutions, only one is the actual solution, called the "final solution". Comparing the possible solutions in the different configurations eliminates inconsistent solutions, since a single solution is common to the two configurations. Thus, the shape of the object and its position in the optical system is obtained.

[0019] In this way, the method makes it possible to obtain the shape of the object as well as its shape defects. The method does not use additional input data. It therefore makes it possible to obtain measurements of the object based on a single object. The method is simplified and therefore approximation errors are avoided.

[0020] In methods that use the reflection of the fringes and thereby the slope of the feature to be measured, the uncertainty in the measurement is comparable to the uncertainty of measurements obtained by interferometry.

[0021] In an embodiment, the method includes a step of calibrating the deflectometer by successively acquiring images of the fringes in at least two configurations (preferably at least three configurations) of the optical system defined by the display positioned directly facing the camera. This calibration method is simple and makes it possible to guarantee the reliability and performance of the method. The calibration, which only involves the display and the camera, makes it possible to correctly identify the paths of the respective incident rays of the camera and the actual shape of the display, taking into account the subsequent measurements.

[0022] To make this method suitable for measuring reflective objects, additional steps are required: moving the display and / or camera so that they are no longer aligned, and having a reference mirror to reflect the display projection towards the camera.

[0023] To verify correct operation of the deflectometer at the end of the calibration, measurements of the reference mirror have been previously obtained by another means, such as an interferometer, and the calibration step includes the additional steps of measuring the reference mirror by the deflectometer and comparing the measurements obtained by the deflectometer with the previous measurements.

[0024] The present invention also relates to a deflectometer for performing measurements on reflective or transparent objects, the deflectometer comprising: a display showing dynamic fringes; a camera configured to acquire images of the fringes reflected by a reflective object and / or transmitted through a transparent object; a computer configured to execute a program for carrying out measurements on the basis of images acquired by the camera; The present invention relates to a deflectometer, including a deflectometer.

[0025] According to the invention, the display and / or the camera are mounted on a frame with the ability to move between at least two positions, such that the camera captures images according to at least two configurations of an optical system defined by the display, the object, and the camera; The measurement of an object is the calculation of its shape, and the computer program is configured to calculate the geometry of the object solely based on the images acquired by the camera in at least two configurations, the calculation being performed by integration of the slopes measured in each configuration.

[0026] Thus, the deflectometer integrates inexpensive components.

[0027] Finally, such a deflectometer does not require the use of additional input data and is therefore simple to use: it is sufficient to have the object to be measured in the optical system defined by the display, the object and the camera in order to be able to perform the measurement: there are no steps that limit the precise positioning or alignment of the object or another element.

[0028] To ensure repeatability of the display and / or camera positioning in each configuration, they are mounted on the frame by an isostatic connection, meaning that this connection does not generate deformations of the mechanical structure in question that would result in measurement errors. In addition, isostatic mounting can be cheaper than mechanical precision systems (e.g. precision screws, etc.).

[0029] To facilitate the movement of the display and / or the camera, this is ensured by a two-stage system, which includes a first stage that can be moved relative to the frame, transporting a second stage on which the display and / or the camera are fixed, the second stage being a free position in which the second stage does not impede the movement of the first stage relative to the frame; an indexed position, in which the second stage prevents the first stage from moving and is isostatically connected to the frame; Thus, an operator does not need to carry the display and / or camera to move them from one configuration to another.

[0030] In a preferred embodiment, the isostatic connection is of the Boys connection type because this connection is simple to implement.

[0031] In this embodiment, the connection of the Boys is realised by three spherical partial pins positioned in a triangle on the underside of the second stage opposite the frame, each intended to abut between the two inner faces of a V positioned opposite the upper side of the frame.

[0032] To simplify the mechanical construction of the kinematic connection, the display and / or the camera can be moved along a linear axis.

[0033] With the aim of encouraging the use of the deflectometer at all times, the movement of the display and / or camera must be - a motor-driven horizontal sliding connection ensuring the movement of the first stage relative to the frame; a motor-driven vertical slide connection that allows the second stage to be retracted between a free position and an indexed position, the vertical slide connection having a clearance. This is ensured by:

[0034] For the purpose of automating the measurements, the motors of the horizontal and vertical slide connections are connected to a computer, and the computer program is configured to automatically control the motors to move the display and / or camera between positions. [Brief description of the drawings]

[0035] [Figure 1] 1 is a diagram illustrating the deflectometry principle. [Diagram 2] 1 is a diagram illustrating a prior art deflectometry method. [Diagram 3] 1 is a diagram illustrating the calibration of the method according to the invention. [Figure 4] 2 is a diagram illustrating the method according to the invention during the measurement of a transparent object. [Diagram 5] 1 is a diagram illustrating another calibration step of the method. [Figure 6] 1 is a diagram illustrating multiple solutions obtained during measurements in a first configuration. [Figure 7] 4 is another diagram illustrating some solutions obtained following measurements in a first configuration; [Figure 8] 11 is a diagram illustrating multiple solutions obtained during measurements in a second configuration. [Figure 9] 1 is a diagram illustrating a comparison of the solutions obtained in a first and second configuration; [Figure 10] 1 is a diagram of a deflectometer camera according to the present invention. [Figure 11]1 is a diagram of a display of a deflectometer according to the present invention. [Figure 12] This is a diagram of the Boys connections. [Figure 13] Another Boys connection diagram. [Figure 14] 1 is a diagram illustrating a two-stage system for transporting a display and / or a camera. [Figure 15] 1 is another diagram illustrating a two-stage system. [Figure 16] 1 is a diagram illustrating a cradle for positioning a display. [Figure 17] 1 is a diagram illustrating the measurement of an object by stereodeflectometry. [Figure 18] 4 is a diagram illustrating the measurement of the same object according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0036] With reference to Fig. 1, deflectometry is a method for measuring an object (2) by implementing a deflectometer (1) including a display (10) that projects fringes, the reflection of which by the object (2) is acquired by a camera (20). The displayed fringes are a predetermined dynamic sinusoidal pattern. This predetermined sinusoidal pattern makes it possible to know, for each pixel captured by the camera (20), from which pixel it presents on the display (10). This corresponds to a prior art technique widely known as phase shift measurement.

[0037] The reflected pattern is distorted by the actual shape of the object (2), so the computer (30) interprets the deformation of the reflected fringes to computationally determine the shape of the object (2).

[0038] By shape of the object (2) this means the measurement of its geometry, its dimensions and its shape defects, approximately down to the scale of one micrometer. The deflectometer (1) according to the invention also makes it possible to obtain the position of the object (2) within the optical system defined by the display (10), the object (2) and the camera (20).

[0039] 1 illustrates a typical case of measuring by deflectometry, where the object (2) is reflective. In this case, the measurement consists of obtaining the shape of the reflective surface of the object (2). However, it is also conceivable to proceed to measure a transparent object (2), in which case the measurement consists of obtaining the overall shape of the object (2) and not just the shape of one of its surfaces.

[0040] 3, the calibration step of the method consists of placing a camera (20) directly facing the display (10). During this step, the display (10) and / or the camera (20) are moved in several configurations and an acquisition of the displayed fringes is performed each time.

[0041] This calibration step makes it possible to perfectly know, in each of the configurations, the shape of the display and the optical path of the incident light ray (ri) coming from the display (10) to the camera (20). That is, based on each pixel of the display, the path of the incident light ray (ri) is determined and connects it to the camera (20). Three light rays (ri1, ri2, ri3) are represented, which correspond to three pixels (P1, P2, P3).

[0042] This step also makes it possible to know the movements carried out in the system between each configuration. A comparison of the images acquired by the camera (20) between each configuration makes it possible to determine, through calculations carried out by a computer program (30), which refined transformations allow to move from one configuration to another. These refined transformations will be used during the subsequent step of measuring the object (2). The refined transformations are compositions of translations and rotations, which completely describe the movements of a rigid solid.

[0043] Precise knowledge of the paths of these incident rays (ri) is necessary for the quality of the subsequent measurements, and the amplitude of the displacements between the different configurations must also be as large as possible. In fact, the larger the amplitude, the more precisely the paths of the incident rays (ri) passing through all positions are determined, but it is important not to design the deflectometer (1) too large. A total amplitude of displacement of approximately 2 meters is a reasonable compromise between the size of the system and the quality of the calibration.

[0044] This calibration step makes it possible to obtain equations for all incident rays (ri) on the camera (20) in the system of optical systems with a lateral accuracy of approximately 1 micrometer and an angular accuracy of approximately 1 microradian.

[0045] This calibration step also makes it possible to measure flatness defects of the display, which makes it possible to improve subsequent measurements of the shape of reflective objects by using this display (20). It is therefore possible to use a standard display (20) that is commercially available, which is an economic advantage. This is the case, for example, for computer monitors or ordinary televisions.

[0046] Calibration must be performed periodically, but it is not necessary to redo the calibration before proceeding to measure each new object (2). Indeed, the stability of the calibration depends only on the geometry of the rigid mechanical structures, which depend only on the temperature, and optionally on the creep of plastic materials that may be involved mainly in the manufacture of displays and camera lenses.

[0047] Referring to FIG. 4, once the calibration step has been performed, it is possible to proceed to measure the transparent object (2) by placing the transparent object (2) between the display (10) and the camera (20).

[0048] However, if the object (2) to be measured is reflective, it is necessary to modify the alignment of the system so that the fringes reflected by the object (2) are observable by the camera (20).

[0049] Referring to FIG. 5, therefore, either the display (10) or the camera (20) is moved and a reference mirror (3) is placed so that the camera (20) has within its field of view an image of the display (10) realised by the mirror (3).

[0050] As the camera (20) is moved relative to the display (10), a new sophisticated transformation is applied in the optical system. In order to be able to precisely identify the reference mirror (3) without introducing new unknowns into the optical system, the reference mirror (3) is previously known, i.e. its exact shape has been measured, for example by interferometry or by using another deflectometer that has already been calibrated.

[0051] A sophisticated transformation corresponding to the movement of the camera (20) relative to the display (10) can be determined based on acquiring images of the fringes of the display (10) when the reference mirror (3) is in any of at least three different positions, allowing the camera (20) to have these fringes in its field of view. There are no restrictions on the quality or repeatability of the positioning of the reference mirror (20) in this step; it is a loose positioning.

[0052] 6, the reference mirror (3) is then removed and the object (2) is positioned so that the camera (20) has fringes seen by reflection in its field of view. The camera (20) first acquires the reflected fringes in a first configuration, with the camera (20) in a first position (201).

[0053] Since the display of the fringes is a predetermined pattern, and since the incident rays (ri) to the camera (20) are all identified during a calibration step, the starting point is known for each incident ray (ri) from the display (10). The unknowns are the shape and position of the object (2) that allowed its reflection in the direction of the camera (20).

[0054] The computer program (30) is programmed to carry out calculations that make it possible to identify possible solutions regarding the shape and position of the object (2). As explained, for each configuration of the optical system defined by the display (10), the object (2) and the camera (20), an infinite number of solutions are possible. Figure 6 shows the first configuration (S 1,i , S 1,j , S 1,k ), each of which can correspond to a different emitted ray (re) and also to different tilts and positions on the surface of the object (2).

[0055] The same determination is performed for each incident ray (ri) on the camera (20), which makes it possible to obtain the complete shape of the object (2), as represented in FIG.

[0056] FIG. 7 illustrates diagrammatically multiple shape and position solutions in space obtained by measurements in a first configuration.

[0057] 8, the camera (20) is moved and positioned at a second position (202) and a new acquisition is performed in this second configuration of the optical system.

[0058] Similarly, knowing the incident ray (ri) from the display (10) and its starting point makes it possible to obtain multiple solutions of the second configuration. To facilitate reading FIG. 8, three solutions of the second configuration (S 2,i , S 2,j , S 2,k ) is shown.

[0059] 9, the final step of the method according to the invention is to compare the sets of solutions obtained in each configuration and determine which is the unique solution common to each configuration, which corresponds to the actual shape and position of the object (2) to be measured.

[0060] To improve the accuracy of the measurements, it is advantageous to perform the measurements in more than two configurations. Thus, the measurement uncertainty is significantly reduced by the so-called "least squares" minimization procedure used.

[0061] The present invention also relates to a deflectometer (1) designed to implement the method described above.

[0062] 10 and 11, such a deflectometer (1) includes a display (10) and a camera (20). At least one of these elements (and preferably both) is movably mounted relative to a frame (40). It is necessary that at least one of these elements is movable so that the optical system can be mounted in two different configurations.

[0063] If the display (10) and the camera (20) are movable, it allows to increase the number of different configurations and to be more easily adapted to the different object shapes (2) to be measured. It is therefore possible to measure concave and convex mirrors with one and the same deflectometer (1).

[0064] The frame (40) is preferably made from two parts so that it can be easily moved from a position where the display (10) and camera (20) are aligned for a calibration step, to a remote position where the display (10) and camera (20) are remote for the step of measuring the reflective object (2).

[0065] For the sake of simplicity, the mobility of the display (10) and / or the camera (20) is an axial translation along the optical axis of the camera (20). The measurement taken is therefore a coaxial deflectometry. In the illustrated embodiment, this mobility is obtained by a two-stage system. The first stage (41) is mounted in a sliding connection to the frame (40). This sliding connection can be of any suitable type, an economical means is to use a profile that engages with a complementarily shaped roller (411).

[0066] Advantageously, this mobility is motorized, for example by a motor (413) driving a pulley and belt system or a worm screw.

[0067] The first stage (41) has embedded therein a second stage (42), which is a free position in which the second stage (42) does not impede the movement of the first stage (41) relative to the frame (40); an index position, in which the second stage (42) prevents the movement of the first stage (41) and thus has an isostatic connection with the frame (40); It is possible to retreat between It is the second stage (42) that carries the display (10) or the camera (20).

[0068] The connection between the first stage (41) and the second stage (42) is not a sliding connection in the strict sense, because the positioning of the second stage (42) must be free with respect to the first stage (41), which therefore involves a lot of clearance for an imperfect kinematic connection.

[0069] Indeed, in the index position, the second stage (42) is isostatically connected to the frame (40) so that the index position is always reproducible, identical to the nearest micrometer, each time the second stage (42) moves from the free position to the index position. If the first stage (41) prevented the second stage (42) from being positioned without constraints relative to the frame (40), the configuration of the optical system would not be exactly reproducible and measurements would be impaired.

[0070] Performing a movement of the display and / or camera in an axial translation (coaxial deflectometry principle) makes it possible to measure objects of larger dimensions as if the measurement of the object was performed according to different angles (stereodeflectometry principle).

[0071] With reference to Figures 17 and 18, measurements of one and the same reflective object (2) are illustrated according to a stereodeflectometry method in Figure 17 and according to an embodiment of the present invention in Figure 18.

[0072] In these figures, the camera (20) is moved according to two positions. The measurable zone (ZM) is at the intersection of the fields of view (CV1, CV2) of the respective cameras (201, 202). To simplify the geometrical construction of the figures, a mirror display (10') is represented, which is the image of the display (10) as seen by the camera (20). This concerns the planar symmetry of the display (10) with respect to a plane locally tangent to the object (2).

[0073] In the stereodeflectometry method illustrated in Fig. 17, a zone must be measured according to several angular configurations in order to obtain usable measurements. The measurable zone (ZM) has a limited surface area. In addition, the effect of stereoscopic vision, and therefore the sensitivity of the measurement, increases with the angular difference of observation, but at the same time limits the measurable zone (ZM). That is a significant disadvantage of stereodeflectometry, which is eliminated by the embodiment illustrated in Fig. 18.

[0074] With reference to Fig. 18, the camera (20) is now moved in an axial translation (T) without changing the viewpoint of the camera (20). It is observed that the measurable zone (ZM) at the intersection of the fields of view (CV1, CV2) of the respective cameras (201, 202) is much larger. The method (according to which the movement path of the display and / or the camera is an axial translation) therefore makes it possible to measure objects of larger dimensions.

[0075] To achieve the isostatic connection between the frame (40) and the second stage (42), several solutions can be considered.

[0076] Referring to FIG. 12, the second stage (42) may include three spherically shaped elements (43) (e.g., pins or curved head screws, etc.) supported by simple gravity against the inner surface of a V (45) (i.e., a "V" shaped wedge) disposed on the frame (40). For the Boys' connection to work, the Vs (45) are not parallel to each other and preferably the axes of the Vs (45) must converge to one central point of the connection. Similarly, the spherically shaped elements (43) may be positioned on the first stage (41) and the Vs (45) may be positioned on the second stage (42).

[0077] Referring to Figures 13 to 15, an alternative solution is to replace each V (45) with a pair of spherical partial abutments (44).

[0078] Although the two mountings are equivalent from a kinematic point of view, the use of the V-shape (45) allows greater freedom of positional difference during docking of the second stage (42) on the frame (40), which simplifies the design of the sliding connection with clearance connecting the second stage (42) with the first stage (41).

[0079] 14 and 15, the free connection between the first stage (41) and the second stage (42) can be seen in more detail. The free connection in the preferred embodiment takes the form of an axis (414) fixed on the first stage (41) and inserted into an oblong orifice (421) of the second stage (42). There is a large clearance between the axis (414) and the orifice (421), between 1 mm and 5 mm.

[0080] The oblong orifice (421) is vertically oriented and its length is sufficient to allow the V-shape (45) or spherical abutment (44) to disengage from the pin (43), thus breaking the isostatic connection of the preferred embodiment, since it is normally maintained by gravity.

[0081] The transition from the indexed position to the free position can be obtained by an actuator or eccentric, preferably motor-driven, mounted on the first stage (41), making it possible to raise the second stage (42).

[0082] Of course, if the isostatic connection is of a different construction (e.g., of the "point, line, plane" type), the kinematics of the retraction between the free position and the index position and the motor drive can be adapted.

[0083] 16, the frame (40) includes several cradles (B1-B5), each configured to isostatically receive a second stage (42). In the illustrated embodiment, each cradle (B1-B5) therefore includes three pins (43).

[0084] In this embodiment, the display (10) is movable between several cradles (B1-B5) distributed over a length of approximately 2 m, the large length making it possible to obtain with high precision the path of the incident light ray (ri).

[0085] The camera (20) itself is movable between several cradles distributed over a length of approximately 25 cm. This length is sufficient so that the configurations used during the measurement steps are sufficiently different and sufficient for the method to determine the shape of the object (2) with the expected accuracy. Moreover, this length is low enough so that the part of the frame (40) carrying the camera (20) is small, which makes it possible to facilitate the transition from the registered position used for calibration to the remote position used for the measurement of the reflective object (2).

[0086] In a preferred embodiment, the display (10) and camera (20) therefore: - in a reproducible manner, to the nearest micrometer, since the indexing is performed by an isostatic connection that does not deform the parts by the introduction of excessive forces - in an assisted manner, since the mobility and reversal are motorized; It is possible to transition from one configuration of the optical system to another, - Because the system does not use mechanical precision systems (such as micrometer tables or precision ball screws), the system is much cheaper than prior art solutions.

[0087] Preferably, the computer program (30) is programmed such that each step of the method is automated, the program comprising: - performing measurements in a configuration; - retracting the display (10) and / or the camera (20) in the free position; - moving the display (10) and / or the camera (20) to bring it to the next index position; - placing the display (10) and / or the camera (20) in the following cradles; and - Carrying out measurements on new configurations and so on until all desired measurements have been performed; then determining the measurements of the object (2) by comparing them with each possible solution of the configuration; Includes instructions for:

[0088] Of course, any optional adjustment of the camera (20) (e.g. diaphragm or lens aperture, etc.) can be motorized and connected to the computer (30) and controlled by a program, which also improves the automation of the deflectometer (1).

[0089] The same applies to the movement of the reference mirror (3) in at least three different arbitrary positions during the calibration step.

[0090] The deflectometer (1) and method according to the invention make it possible to perform shape and dimensional measurements on an object (2) by using only the display (10) and the camera (20) therefor, without any other additional input data. The deflectometer (1) can be used in environments not specifically provided for metrology and does not need to be positioned on an optical table due to its insensitivity to vibrations. The measurements performed are of metrological quality comparable to interferometers and are adapted to the control of parts for optical imaging, while the means implemented are inexpensive.

[0091] The method and the deflectometer (1) can be configured differently from the examples and illustrations given without departing from the scope of the invention as defined by the claims.

[0092] In particular, the movability of the display (10) and / or the camera (20) is not necessarily translational, and the markings of different configurations can be shaped differently from the indexing presented.

[0093] In embodiments not shown, to conserve equipment, only the display (10) or only the camera (20) is movable between several configurations.

[0094] In particular, the camera 20 is lighter and smaller to move than the display, and therefore in some embodiments, it is only the camera 20 that is movable.

[0095] Moreover, the technical features of the different embodiments and variants described above can be combined with each other, either completely or in part, so that the method and the deflectometer (1) can be adapted in terms of cost, functionality and performance. [Explanation of symbols]

[0096] 1 Deflectometer 2 objects 3 Reference Mirror 10. Display 10' Mirror Display 20 Camera 30 Computers 40 frames 41 First Stage 42 Second Stage

Claims

1. A method for carrying out measurements of a reflective or transparent object (2) by means of a deflectometer (1), comprising: A display (10) for displaying dynamic fringes; a camera (20) configured to acquire images from fringes reflected by the reflective object (2) and / or transmitted through the transparent object (2); a computer (30) configured to execute a program to perform measurements based on the images acquired by the camera (20); Including, - acquiring successively images from fringes reflected and / or transmitted by said object (2) in at least two configurations of an optical system defined by said display (10), said object (2) and said camera (20); - executing said program based on the acquired images, and calculating possible solutions for the shape of said object (2) and the position of said object (2) in the optical system defined by said display (10), said object (2) and said camera (20) in each of said at least two configurations; obtaining a final solution for the shape and the position of the object (2) by identifying a single solution that is common to previously calculated possible solutions.

2. 2. The method of claim 1, further comprising the step of calibrating the deflectometer (1) by successively acquiring images from fringes in at least two configurations of an optical system defined by the display (10) positioned directly facing the camera (20).

3. 3. The method of claim 2, wherein the calibrating step comprises the additional steps of moving the display (10) and / or the camera (20) out of alignment and placing a reference mirror (3) to reflect a projection from the display (10) towards the camera (20).

4. 4. The measurement method according to claim 3, wherein measurements of the reference mirror (3) have been previously obtained by another means, such as an interferometer, and the calibrating step comprises the additional step of measuring the reference mirror (3) by means of the deflectometer (1) to verify correct operation of the deflectometer (1) by comparing the measurements obtained by the deflectometer (1) with the previous measurements.

5. A deflectometer (1) for making measurements from a reflective or transparent object (2), comprising: A display (10) for displaying dynamic fringes; a camera (20) configured to acquire an image of the fringes reflected by the reflective object (2) and / or transmitted through the transparent object (2); a computer (30) configured to execute a program to perform measurements based on the images acquired by the camera (20); the display (10) and / or the camera (20) are mounted on a frame (40) for movement between at least two positions, and the camera (20) captures images according to at least two configurations of an optical system defined by the display (10), the object (2) and the camera (20); A deflectometer (1), wherein the measurement of the object (2) is a calculation of the shape of the object, and the program is configured to calculate the geometric shape of the object (2) based solely on the images acquired by the camera (20) in the at least two configurations, the calculation being performed by integration of the slopes measured in each of the at least two configurations.

6. The deflectometer (1) according to claim 5, wherein the display (10) and / or the camera (20) are mounted on the frame (40) by an isostatic connection.

7. The movement of the display (10) and / or the camera (20) is ensured by a two-stage system, which comprises a first stage (41) that is moved relative to the frame (40) and the display (10) and / or the camera (20) transport a second stage (42) that is fixed on the first stage, The second stage (42) comprises: a free position in which the second stage (42) does not impede movement of the first stage (41) relative to the frame (40); The second stage (42) prevents the movement of the first stage (41) and has an index position in which it is isostatically connected to the frame (40).

6. The deflectometer (1) according to claim 5, configured to retract between

8. The deflectometer (1) according to claim 6 or 7, wherein the isostatic connection is a Boys connection.

9. 9. The deflectometer (1) according to claim 8, wherein the Boys connection is realized by three spherical partial pins (43) positioned in a triangle on the underside of the second stage (42) opposite the frame (40), each of the three spherical partial pins (43) being configured to abut two internal faces of a V-shape (45) positioned on the opposite side of the upper side of the frame (40).

10. The deflectometer (1) according to claim 5, wherein the display (10) and / or the camera (20) are configured to move along a linear axis.

11. The movement of the display (10) and / or the camera (20) may include: a motor-driven horizontal slide connection ensuring the movement of the first stage (41) relative to the frame (40); a motor-driven vertical slide connection for retracting the second stage (42) between the free position and the indexed position, the vertical slide connection having a clearance; The deflectometer (1) according to claim 7, secured by

12. The deflectometer (1) of claim 11, wherein motors of the horizontal and vertical slide connections are connected to the computer (30), and the program is configured to automatically control the motors to move the display (10) and / or the camera (20) between the free position and the index position.