Phase deflection measurement method for common-reference double-mirror-surface object

Abstract

The invention discloses a common-reference double-mirror-surface object phase deflection measurement method. The method comprises the following steps: 1) building a common-reference double-mirror-surface object phase deflection measurement system; 2) replacing a double-mirror object with a single-mirror calibration plate to obtain common reference calibration parameters of the system; 3) obtaining system parameters needing to be calibrated; 4) generating a stripe image; 5) acquiring a zero-deformation stripe image; 6) acquiring a deformed stripe image; 7) stripe phase calculation; 8) respectively solving the three-dimensional shape data of the A-side mirror surface and the B-side mirror surface of the double-mirror-surface object under the corresponding coordinate system; and 9) fusing the three-dimensional shape data of the double-mirror object. According to the invention, two single-screen moving direct phase deflection measurement subsystems are combined together, the two subsystems measure three-dimensional coordinates of two surfaces of a double-mirror object in a reference coordinate system and a coordinate system to be calibrated respectively, and then three-dimensional shape data measured by the two subsystems are converted to the same reference coordinate system. And obtaining a complete three-dimensional shape of the double-mirror object.

Description

Technical Field

[0001] The invention relates to the field of optical three-dimensional shape measurement, and in particular to a phase deflection measurement method for a common reference double-mirror object. Background Art

[0002] Phase deflectometry is an optical three-dimensional shape measurement technology suitable for mirror surface shape. Based on the principle of fringe reflection, the computer generates standard sinusoidal fringes for transmission and display on the display screen. The mirror to be measured and the mirror calibration plate are imaged through reflection in succession. The standard sinusoidal fringes displayed on the display screen are deformed by the modulation of the mirror surface shape, and then the virtual image of the deformed fringes is collected from another direction by the CCD camera and transmitted back to the computer for demodulation and calculation. By calculating and comparing the phase information in the sinusoidal fringes reflected by the mirror calibration plate and the phase information in the sinusoidal fringes reflected by the mirror to be measured, the depth of each point on the measured mirror relative to the reference position of the mirror calibration plate can be solved, and then the three-dimensional shape of the mirror can be restored.

[0003] Existing phase deflectometry can only measure a single mirror in principle, but there are many important dual-mirror components in military weapons and equipment, astronomical telescopes, and aerospace equipment. Since such components need to simultaneously measure the shape of aspheric or free-form surfaces and the position and posture between multiple working surfaces, it poses a huge challenge to existing precision measurement technology. There is a lack of effective deflectometry theory and methods for the simultaneous measurement of the shape and posture of the two surfaces of such dual-mirror components. The existing measurement technologies are as follows:

[0004] In the document "Li Ang, et al. Three-coordinate measurement data processing of high-order aspheric surfaces in optical processing [J]. Chinese Optics, 2020, 13(2): 302-312.", the three-coordinate machine uses a mechanical probe to directly contact the surface to be measured, scans along a certain scanning path in the two-dimensional direction to measure the vector height of discrete points on the surface to be measured point by point, and obtains the surface data by fitting. This method is slow and costly, and will cause certain damage to surfaces with high precision requirements. In addition, there is a large error when fitting the surface shape using scattered point data.

[0005] In the paper "Jiang S, He Q, Xing Y, et al. Multi-view stitching phase measuring deflectometry for freeform specular surface metrology [J]. Optics Express, 2023, 31 (22), 36557.", multiple cameras at different positions are used to capture the deformation fringes generated by reflections from different areas of the measured mirror surface, and stereo deflectometry is used in the intersection area of ​​the two camera lines of sight to determine their absolute depth. The gradient of each area is calculated in the traditional phase deflectometry system composed of each camera and the screen, and then the gradient is integrated to obtain the absolute depth of each area. The appearance of all areas is accurately stitched together to reconstruct the overall three-dimensional morphology of the surface. This method effectively solves the challenges encountered by the traditional phase measurement deflectometry method when measuring high-slope free-form surfaces, but the system structure is complex, and the integration results need to be continuously iterated to obtain the most accurate measurement data. The calculation process is complex and the integration process is prone to error accumulation.

[0006] The document with application number 202010655098.X discloses a method and device for measuring the three-dimensional shape of a mirror object based on a dual-screen transmission display. The semi-transparent and semi-reflective mirror and a display screen are replaced with a transparent screen, making the system structure clear. The two displays generate sinusoidal fringes on two parallel planes, respectively. The relationship between the phase and the depth of the object to be measured is also established for measurement, which can achieve the measurement of large-gradient and non-continuous mirror objects. However, the display characteristics of the transparent screen are different from those of the display screen, and the refraction effect caused by the transparent screen in the measurement system introduces errors to the measurement results.

[0007] From the above literature, it can be seen that although the contact-type shape measurement method can obtain the overall three-dimensional shape of the measured object, due to the characteristics of the contact type, it has disadvantages in terms of speed, accuracy and cost. Direct phase deflectometry based on fringe reflection is a technology recently proposed for mirror surface measurement, which can achieve the measurement of large gradient and non-continuous mirror objects. However, the range of the reconstructed three-dimensional shape is limited by the display screen and the camera angle. It can only measure the three-dimensional shape of the part of the mirror object illuminated by the fringe image within a single camera market, and cannot complete the measurement of the overall three-dimensional shape of the two surfaces of the dual-mirror element. For the simultaneous measurement of the two surface shapes and their relative postures of the dual-mirror element, an effective phase deflectometry theory and method is needed. Summary of the invention

[0008] In view of the deficiencies in the prior art, the technical problem to be solved by the present invention is to provide a method for measuring phase deflection of a common reference dual-mirror object.

[0009] The technical solution of the present invention to solve the technical problem is to provide a method for measuring phase deflection of a common reference double-mirror object, characterized in that the method comprises the following steps:

[0010] 1) Build a common reference dual-mirror object phase deflection measurement system:

[0011] The system includes a double mirror object, a computer, a first CCD camera, a second CCD camera, a first high-precision horizontal motion guide rail, a second high-precision horizontal motion guide rail, a third high-precision horizontal motion guide rail, a first display screen, and a second display screen;

[0012] The first high-precision horizontal moving guide rail is fixed on the optical platform, and the first display screen is slidably mounted on the first high-precision horizontal moving guide rail, is perpendicular to the optical platform, and can be linearly moved along the first high-precision horizontal moving guide rail to the first display screen initial position and the first display screen moving position;

[0013] The second high-precision horizontal moving guide rail is fixed on the optical platform, and the second display screen is slidably mounted on the second high-precision horizontal moving guide rail, is perpendicular to the optical platform, and can be linearly moved along the second high-precision horizontal moving guide rail to the second display screen initial position and the second display screen moving position;

[0014] The third high-precision horizontal moving guide rail is fixed on the optical platform, and the double-mirror object is slidably mounted on the third high-precision horizontal moving guide rail and can rotate, so that the virtual image of the first display screen reflected by the mirror surface on the A side of the double-mirror object is within the depth of field of the first CCD camera, and the virtual image of the second display screen reflected by the mirror surface on the B side of the double-mirror object is within the depth of field of the second CCD camera; the first CCD camera, the first display screen and the mirror surface on the A side are in a triangulated measurement relationship in space; the second CCD camera, the second display screen and the mirror surface on the B side are in a triangulated measurement relationship in space;

[0015] The computer, the first CCD camera, the first high-precision horizontal moving guide rail, the first display screen and the A-side mirror constitute the A-side subsystem; the computer, the second CCD camera, the second high-precision horizontal moving guide rail, the second display screen and the B-side mirror constitute the B-side subsystem;

[0016] The computer is respectively connected to the first CCD camera, the second CCD camera, the first high-precision horizontal moving guide rail, the second high-precision horizontal moving guide rail, the third high-precision horizontal moving guide rail, the first display screen, and the second display screen in communication;

[0017] 2) Remove the double-mirror object from the third high-precision horizontal guide rail, and then install the single-mirror calibration plate on the third high-precision horizontal guide rail to obtain the common reference calibration parameters of the system:

[0018] When the single mirror calibration plate is used to calibrate the A-side subsystem, it is set as the first-position mirror calibration plate. The virtual image of the first display screen reflected by the first-position mirror calibration plate and the first-position mirror calibration plate are within the depth of field of the first CCD camera. The first CCD camera, the first display screen and the first-position mirror calibration plate are in a triangular measurement relationship in space. The plane where the circular ring array on the surface of the first-position mirror calibration plate is located is set as the A-side reference plane, and the coordinate system determined by the circular ring array on the surface of the first-position mirror calibration plate is set as the reference coordinate system O. A -X A Y A Z A ;

[0019] When the single mirror calibration plate is used to calibrate the B-side subsystem, the second-pose mirror calibration plate is used. The virtual image of the second display screen reflected by the second-pose mirror calibration plate and the second-pose mirror calibration plate are within the depth of field of the second CCD camera. The second CCD camera, the second display screen and the second-pose mirror calibration plate are in a triangular measurement relationship in space. The plane where the circular ring array on the surface of the second-pose mirror calibration plate is located is set as the B-side reference plane, and the coordinate system determined by the circular ring array on the surface of the second-pose mirror calibration plate is set as the coordinate system to be calibrated. B -X B Y B Z B ;

[0020] The pose transformation method between the first pose mirror calibration plate and the second pose mirror calibration plate is known, and the reference coordinate system O is calculated using the rigid body transformation method. A -X A Y A Z A With the coordinate system to be calibrated O B -X B Y B Z B The transformation matrix between

[0021] 3) Obtain the system parameters that need to be calibrated:

[0022] For the A-side subsystem, the distance d between the initial position of the first display screen and the A-side reference plane is calculated. A ; Use the first high-precision horizontal moving guide rail to move the first display screen from the first display screen initial position to the first display screen moving position, and the moving distance is Δd A ;

[0023] Move the first posture mirror calibration plate along the third high-precision horizontal moving guide rail within the depth of field of the first CCD camera. A Positions, n A≥6, the first CCD camera at each position collects the circular array on the surface of the first pose mirror calibration plate, and records the center of each circular ring in the circular array in the reference coordinate system O A -X A Y A Z A The coordinates (X Ab ,Y Ab ,Z Ab ), and then according to the pixel coordinates of the center of each circle in the circle array extracted by the first CCD camera, the pixel coordinates (u A ,v A ) and X Ab and Y Ab The coefficient a in the corresponding relationship A0 、a A1 、a A2 , b A0 , b A1 , b A2 , complete the calibration of the X and Y coordinates of the A-side subsystem;

[0024]

[0025] In formula (1), a A0 、a A1 、a A2 , b A0 , b A1 , b A2 is the coefficient of the relationship;

[0026] For the B-side subsystem, the distance d between the initial position of the second display screen and the B-side reference plane is calculated. B ; Use the second high-precision horizontal moving guide rail to move the second display screen from the second display screen initial position to the second display screen moving position, and the moving distance is Δd B ;

[0027] Move the second pose mirror calibration plate along the third high-precision horizontal moving guide rail within the depth of field of the second CCD camera. B Positions, n B ≥6, the second CCD camera at each position collects the circular array on the surface of the second pose mirror calibration plate, and records the center of each circular ring in the circular array in the coordinate system to be calibrated O B -X B Y B Z B The coordinates (X Bb ,Y Bb ,Z Bb), and then according to the pixel coordinates of the center of each circle in the circle array extracted by the second CCD camera, the pixel coordinates (u B ,v B ) and X Bb and Y Bb The coefficient a in the corresponding relationship B0 、a B1 、a B2 , b B0 , b B1 , b B2 , complete the calibration of the X and Y coordinates of the B-side subsystem;

[0028]

[0029] In formula (2), a B0 、a B1 、a B2 , b B0 , b B1 , b B2 is the coefficient of the relationship;

[0030] 4) Generate sinusoidal straight fringe images;

[0031] 5) Collect zero-deformation fringe images:

[0032] For the A-side subsystem, the sinusoidal straight stripe images are respectively displayed at the initial position of the first display screen and the first display screen moving position, and then are not deformed after being reflected by the A-side reference plane. The first CCD camera collects the zero-deformation stripe images of the sinusoidal straight stripe images at the initial position of the first display screen and the first display screen moving position after being reflected by the A-side reference plane, and stores them in the computer;

[0033] For the B-side subsystem, the sinusoidal straight stripe images are respectively displayed at the initial position of the second display screen and the moving position of the second display screen, and then are not deformed after being reflected by the B-side reference plane. The second CCD camera collects the zero-deformation stripe images of the sinusoidal straight stripe images at the initial position of the second display screen and the moving position of the second display screen after being reflected by the B-side reference plane, and stores them in the computer;

[0034] 6) Collect deformed fringe images:

[0035] Remove the single-mirror calibration plate from the third high-precision horizontal guide rail, and then install the double-mirror object on the third high-precision horizontal guide rail, ensuring that the double-mirror object is within the depth of field of the first CCD camera and the second CCD camera, and the posture conforms to the triangulation relationship;

[0036] For the A-side subsystem, the sinusoidal straight stripe images are respectively displayed at the initial position of the first display screen and the first display screen moving position, and then modulated and deformed by the A-side mirror. The first CCD camera collects the deformed stripe images of the sinusoidal straight stripe images at the initial position of the first display screen and the first display screen moving position reflected by the A-side mirror, and stores them in the computer;

[0037] For the B-side subsystem, the sinusoidal straight stripe images are displayed at the initial position of the second display screen and the moving position of the second display screen respectively, and then modulated and deformed by the B-side mirror. The second CCD camera collects the deformed stripe images of the sinusoidal straight stripe images at the initial position of the second display screen and the moving position of the second display screen reflected by the B-side mirror, and stores them in the computer;

[0038] 7) Fringe phase calculation:

[0039] For the A-side subsystem, the zero-deformation fringe image displayed on the first display screen obtained in step 5) and reflected by the A-side reference plane is calculated to obtain an absolute phase unwrapped image, and then the absolute phase value of the initial position of the first display screen reflected by the A-side reference plane is obtained. and the absolute phase value of the first display screen moving position reflected by the reference plane on side A For the deformed fringe image obtained by step 6) displayed on the first display screen and reflected by the mirror on the side A, the absolute phase unwrapping diagram is calculated, and then the absolute phase value of the initial position of the first display screen reflected by the mirror on the side A is obtained. and the absolute phase value of the first display screen moving position reflected by the mirror on side A

[0040] For the B-side subsystem, the same method as the A-side subsystem is used to obtain the absolute phase value of the initial position of the second display screen reflected by the B-side reference plane. The absolute phase value of the moving position of the second display screen reflected by the reference plane on side B The absolute phase value of the initial position of the second display screen reflected by the mirror on the B side The absolute phase value of the moving position of the second display screen reflected by the mirror on the B side

[0041] 8) Solve the three-dimensional shape data of the A-side mirror surface and the B-side mirror surface of the double-mirror object in the corresponding coordinate system:

[0042] For the A-side subsystem, the relationship between the phase and the depth information of the A-side mirror is established as shown in equation (3), and the depth information h of the A-side mirror relative to the A-side reference plane is obtained: A :

[0043]

[0044] Then, according to the Z value of the mirror on the A side at each pixel of the first CCD camera, A The coefficient a of the relationship between coordinates and lateral calibration A0 、a A1 、a A2 、b A0 、b A1 、b A2 , calculate each pixel point in the reference coordinate system O A -X A Y A Z A The X and Y coordinates of the mirror surface on the A side in the reference coordinate system O are obtained. A -X A Y A Z A The three-dimensional coordinates (X A ,Y A ,Z A );

[0045] For the B-side subsystem, the same method as the A-side subsystem is used to establish the relationship between the phase and the depth information of the B-side mirror as shown in equation (4), and the depth information h of the B-side mirror relative to the B-side reference plane is obtained: B :

[0046]

[0047] Then, according to the Z value of each pixel of the second CCD camera, the mirror on the B side B The coefficient a of the relationship between coordinates and lateral calibration B0 、a B1 、a B2 、b B0 、b B1 、b B2 , calculate each pixel point in the coordinate system to be calibrated O B -X B Y B Z B The X and Y coordinates of the mirror surface on the B side in the coordinate system to be calibrated O are obtained. B -X B Y B Z B The three-dimensional coordinates (X B ,Y B ,Z B );

[0048] 9) 3D shape data fusion of dual mirror objects:

[0049] According to the reference coordinate system O obtained in step 2) A -X A Y A Z AWith the coordinate system to be calibrated O B -X B Y B Z B The transformation matrix between By using formula (5), a point on the mirror surface on the B side is located in the coordinate system to be calibrated O B -X B Y B Z B The three-dimensional coordinates (X B ,Y B ,Z B ) is converted to the reference coordinate system O A -X A Y A Z A The coordinates below

[0050]

[0051] At this time, the three-dimensional coordinates (X A ,Y A ,Z A ) and the three-dimensional coordinates of the mirror surface on the B side Both are in the reference coordinate system O A -X A Y A Z A The overall three-dimensional morphology data of the double-mirror object is obtained.

[0052] Compared with the prior art, the present invention has the following beneficial effects:

[0053] 1) The measurement system of the present invention is simple and ingenious, and can measure the surface shape and posture of both sides of a dual-mirror object: The present invention combines two single-screen mobile direct phase deflection measurement subsystems to overcome the shortcoming of a small measurable range of a direct phase deflection measurement system with only a single camera. The two subsystems respectively measure the three-dimensional coordinates of the two surfaces of the dual-mirror object in the reference coordinate system and the coordinate system to be calibrated, and then obtain the transformation matrix between the reference coordinate system and the coordinate system to be calibrated through the known posture transformation method between the first posture mirror calibration plate and the second posture mirror calibration plate, thereby realizing the common reference calibration of the two subsystems, thereby converting the three-dimensional morphology data measured by the two subsystems to the same reference coordinate system, and accurately and quickly obtaining the complete three-dimensional morphology of the dual-mirror object.

[0054] 2) High measurement accuracy: A direct phase deflection measurement system based on single-screen movement is used as a subsystem. The accuracy of the display screen movement distance is guaranteed by a high-precision horizontal moving guide rail with an accuracy of 1 micron. There is no light refraction error introduced by a transparent screen and field of view limitation introduced by a semi-transparent and semi-reflective mirror, which improves the measurement accuracy of the system.

[0055] 3) Nondestructive testing: This method is an optical three-dimensional topography measurement method. It does not contact the surface being measured and will not damage the surface being measured.

[0056] 4) The common reference dual-mirror object phase deflection measurement method and system of the present invention are universal and easy to promote. In practical applications, they play an important role in the precise measurement of the surface shape and position of important dual-mirror components in military weapons and equipment, astronomical telescopes, and aerospace equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 It is a structural diagram of the measurement system of the present invention;

[0058] Figure 2 A schematic diagram of the positions of the first display screen and the second display screen of the present invention;

[0059] Figure 3 It is a schematic diagram of the calibration structure of the measurement system of the present invention;

[0060] Figure 4 It is a schematic diagram of the measurement principle of the present invention;

[0061] Figure 5 The phase diagram of the sinusoidal fringe expansion obtained by the present invention after being reflected by the mirror surface on the A side of the double mirror object;

[0062] Figure 6 The phase diagram of the sinusoidal fringe expansion obtained by the present invention after being reflected by the B-side mirror of the double-mirror object;

[0063] Figure 7 The three-dimensional shape of the mirror surface on the A side of the double-mirror object obtained by the present invention in the reference coordinate system;

[0064] Figure 8 The three-dimensional shape of the B-side mirror surface of the double-mirror object obtained by the present invention in the coordinate system to be calibrated;

[0065] Fig. 9 The three-dimensional morphologies of the two surfaces of the double-mirror object obtained by the present invention in the reference coordinate system.

[0066] In the figure, there are a first pose mirror calibration plate 1, a second pose mirror calibration plate 2, a double mirror object 3, a computer 4, a first CCD camera 5, a second CCD camera 6, a first display screen initial position 7, a second display screen initial position 8, a first display screen moving position 9, a second display screen moving position 10, a first high-precision horizontal moving guide rail 11, a second high-precision horizontal moving guide rail 12, a third high-precision horizontal moving guide rail 13, a side mirror 14 of the double mirror object 3, a side mirror 15 of the double mirror object 3, an A side reference plane 16, a B side reference plane 17, a first display screen 18, and a second display screen 19. DETAILED DESCRIPTION

[0067] The specific embodiments of the present invention are given below. The specific embodiments are only used to further illustrate the present invention in detail and do not limit the protection scope of the present invention.

[0068] The present invention provides a common reference double mirror object phase deflection measurement method (hereinafter referred to as method), characterized in that the method comprises the following steps:

[0069] 1) Build a common reference dual-mirror object phase deflection measurement system (referred to as the system):

[0070] The system includes a double mirror object 3, a computer 4, a first CCD camera 5, a second CCD camera 6, a first high-precision horizontal moving guide rail 11, a second high-precision horizontal moving guide rail 12, a third high-precision horizontal moving guide rail 13, a first display screen 18 and a second display screen 19;

[0071] The first high-precision horizontal moving guide rail 11 is fixed on the optical platform, and the first display screen 18 is slidably mounted on the first high-precision horizontal moving guide rail 11, perpendicular to the optical platform and capable of linearly moving along the first high-precision horizontal moving guide rail 11 to the first display screen initial position 7 and the first display screen moving position 9;

[0072] The second high-precision horizontal moving guide rail 12 is fixed on the optical platform, and the second display screen 19 is slidably mounted on the second high-precision horizontal moving guide rail 12, perpendicular to the optical platform and capable of linearly moving along the second high-precision horizontal moving guide rail 12 to the second display screen initial position 8 and the second display screen moving position 10;

[0073] The third high-precision horizontal moving guide rail 13 is fixed on the optical platform, and the double-mirror object 3 is slidably installed on the third high-precision horizontal moving guide rail 13 and the double-mirror object 3 can rotate, so that the virtual image of the first display screen 18 reflected by the A-side mirror 14 (referred to as the A-side mirror 14) of the double-mirror object 3 is within the depth of field of the first CCD camera 5, and the virtual image of the second display screen 19 reflected by the B-side mirror 15 (referred to as the B-side mirror 15) of the double-mirror object 3 is within the depth of field of the second CCD camera 6; the first CCD camera 5, the first display screen 18 and the A-side mirror 14 are in a triangulated measurement relationship in space; the second CCD camera 6, the second display screen 19 and the B-side mirror 15 are in a triangulated measurement relationship in space;

[0074] The computer 4, the first CCD camera 5, the first high-precision horizontal moving guide rail 11, the first display screen 18 and the A-side mirror 14 constitute the A-side subsystem (i.e., the A-side phase deflection measurement subsystem); the computer 4, the second CCD camera 6, the second high-precision horizontal moving guide rail 12, the second display screen 19 and the B-side mirror 15 constitute the B-side subsystem (i.e., the B-side phase deflection measurement subsystem); the structures on both sides of the double-mirror object are arranged in mirror symmetry through the cross section thereof, and the structures on both sides share a computer;

[0075] The computer 4 is connected to the first display screen 18 and the second display screen 19 through the HDMI interface, and controls the first display screen 18 and the second display screen 19 to display three groups of sinusoidal stripes that meet the best three-stripe selection method; the computer 4 is connected to the first CCD camera 5 and the second CCD camera 6 through the USB3.0 interface, and controls the first CCD camera 5 and the second CCD camera 6 to collect stripe images; the computer 4 is connected to the first high-precision horizontal moving guide rail 11, the second high-precision horizontal moving guide rail 12 and the third high-precision horizontal moving guide rail 13, and controls the precise movement of the first display screen 18, the second display screen 19 and the single-mirror calibration plate or the double-mirror object 3 thereon;

[0076] 2) Remove the double-mirror object 3 from the third high-precision horizontal guide rail 13, and then install the single-mirror calibration plate on the third high-precision horizontal guide rail 13 to obtain the common reference calibration parameters of the system:

[0077] When the single mirror calibration plate is calibrated for the A-side subsystem, the first-position mirror calibration plate 1 is set; the virtual image of the first display screen 18 reflected by the first-position mirror calibration plate 1 and the first-position mirror calibration plate 1 are within the depth of field of the first CCD camera 5; the first CCD camera 5, the first display screen 18 and the first-position mirror calibration plate 1 are in a triangular measurement relationship in space; the plane where the circular ring array on the surface of the first-position mirror calibration plate 1 is located is set as the A-side reference plane 16, and the coordinate system determined by the circular ring array on the surface of the first-position mirror calibration plate 1 is set as the reference coordinate system O A -X A Y A Z A ;

[0078] When the single mirror calibration plate is calibrated for the B-side subsystem, the second posture mirror calibration plate 2 is set, the virtual image of the second display screen 19 reflected by the second posture mirror calibration plate 2 and the second posture mirror calibration plate 2 are within the depth of field of the second CCD camera 6, and the second CCD camera 6, the second display screen 19 and the second posture mirror calibration plate 2 are in a triangular measurement relationship in space; the plane where the circular ring array on the surface of the second posture mirror calibration plate 2 is located is set as the B-side reference plane 17, and the coordinate system determined by the circular ring array on the surface of the second posture mirror calibration plate 2 is set as the coordinate system to be calibrated. B -XB Y B Z B ;

[0079] The pose transformation between the first pose mirror calibration plate 1 and the second pose mirror calibration plate 2 is known, and the reference coordinate system O is calculated using the rigid body transformation method. A -X A Y A Z A With the coordinate system to be calibrated O B -X B Y B Z B The transformation matrix between That is, common reference calibration parameters;

[0080] 3) Obtain the system parameters that need to be calibrated:

[0081] For the A-side subsystem, a circular array with known spatial distance and radius is displayed on the first display screen 18 located at the initial position 7 of the first display screen through software. The first CCD camera 5 collects the circular array of the first-position mirror calibration plate 1 and the circular array of the first display screen 18 reflected by the first-position mirror calibration plate 1, respectively, extracts the boundary of the circular array, and determines the center of the circular array by ellipse fitting. The distance d between the initial position 7 of the first display screen and the A-side reference plane 16 is calculated using Zhang Zhengyou's camera-based machine vision calibration method. A The first high-precision horizontal movable guide rail 11 is used to move the first display screen 18 from the first display screen initial position 7 to the first display screen moving position 9, and the moving distance is Δd A ;

[0082] The first posture mirror calibration plate 1 is moved along the third high-precision horizontal moving guide rail 13 within the depth of field of the first CCD camera 5. A Positions, n A ≥6 (in this embodiment, n A =11), the first CCD camera 5 at each position collects the circular array on the surface of the first pose mirror calibration plate 1, and records the center of each circular ring in the circular array in the reference coordinate system O A -X A Y A Z A The coordinates (X Ab ,Y Ab ,Z Ab ), and then according to the pixel coordinates of the center of each circle in the circle array extracted by the first CCD camera 5, the pixel coordinates (u A ,v A ) and X Ab and Y AbThe coefficient a in the corresponding relationship A0 、a A1 、a A2 、b A0 、b A1 、b A2 , complete the calibration of the X and Y coordinates of the A-side subsystem, i.e., lateral calibration;

[0083]

[0084] In formula (1), a A0 、a A1 、a A2 、b A0 、b A1 、b A2 is the coefficient of the relationship;

[0085] For the B-side subsystem, a circular array with known spatial distance and radius is displayed on the second display screen 19 located at the initial position 8 of the second display screen through software. The second CCD camera 6 collects the circular array of the second posture mirror calibration plate 2 and the circular array of the second display screen 19 reflected by the second posture mirror calibration plate 2, respectively, extracts the boundary of the circular array, and determines the center of the circular array by ellipse fitting. The distance d between the initial position 8 of the second display screen and the B-side reference plane 17 is calculated using Zhang Zhengyou's camera-based machine vision calibration method. B The second high-precision horizontal movable guide rail 12 is used to move the second display screen 19 from the second display screen initial position 8 to the second display screen moving position 10, and the moving distance is Δd B ;

[0086] The second posture mirror calibration plate 2 is moved along the third high-precision horizontal moving guide rail 13 within the depth of field of the second CCD camera 6 by n. B Positions, n B ≥6 (in this embodiment, n B =11), the second CCD camera 6 at each position collects the circular array on the surface of the second posture mirror calibration plate 2, and records the center of each circular ring in the circular array in the coordinate system to be calibrated O B -X B Y B Z B The coordinates (X Bb ,Y Bb ,Z Bb ), and then according to the pixel coordinates of the center of each circle in the circle array extracted by the second CCD camera 6, the pixel coordinates (u B ,v B ) and X Bb and Y Bb The coefficient a in the corresponding relationshipB0 、a B1 、a B2 、b B0 、b B1 、b B2 , complete the calibration of the X and Y coordinates of the B-side subsystem, i.e., lateral calibration;

[0087]

[0088] In formula (2), a B0 、a B1 、a B2 、b B0 、b B1 、b B2 is the coefficient of the relationship;

[0089] 4) generating fringe images: using a computer 4 and an existing fringe generation method, three groups of sinusoidal straight stripes to be displayed in the measurement field are selected according to the size of the measurement field and the requirements of the measurement accuracy, and the number of stripes in the three groups of sinusoidal straight stripes meets the optimal three-fringe selection method; each group of sinusoidal straight stripes includes four sinusoidal straight stripe images with a phase shift of 90 degrees between each other (referred to as sinusoidal straight stripe images);

[0090] 5) Collect zero-deformation fringe images:

[0091] For the A-side subsystem, the sinusoidal straight stripe images are respectively displayed at the first display screen initial position 7 and the first display screen moving position 9, and then are not deformed after being reflected by the A-side reference plane 16. The first CCD camera 5 collects the zero-deformation stripe images of the sinusoidal straight stripe images at the first display screen initial position 7 and the first display screen moving position 9 after being reflected by the A-side reference plane 16, and stores them in the computer 4 for subsequent processing;

[0092] For the B-side subsystem, the sinusoidal straight stripe images are respectively displayed at the second display screen initial position 8 and the second display screen moving position 10, and then are not deformed after being reflected by the B-side reference plane 17. The second CCD camera 6 collects the zero-deformation stripe images of the sinusoidal straight stripe images at the second display screen initial position 8 and the second display screen moving position 10 after being reflected by the B-side reference plane 17, and stores them in the computer 4 for subsequent processing;

[0093] 6) Collect deformed fringe images:

[0094] The single-mirror calibration plate is removed from the third high-precision horizontal guide rail 13, and then the double-mirror object 3 is installed on the third high-precision horizontal guide rail 13, so that the virtual image of the first display screen 18 reflected by the A-side mirror 14 is within the depth of field of the first CCD camera 5, and the virtual image of the second display screen 19 reflected by the B-side mirror 15 is within the depth of field of the second CCD camera 6; the first CCD camera 5, the first display screen 18 and the A-side mirror 14 are in a triangular measurement relationship in space; the second CCD camera 6, the second display screen 19 and the B-side mirror 15 are in a triangular measurement relationship in space;

[0095] For the A-side subsystem, the sinusoidal straight stripe images are displayed at the first display screen initial position 7 and the first display screen moving position 9 respectively, and then modulated and deformed by the A-side mirror 14. The first CCD camera 5 collects the deformed stripe images of the sinusoidal straight stripe images at the first display screen initial position 7 and the first display screen moving position 9 reflected by the A-side mirror 14, and stores them in the computer 4 for subsequent processing;

[0096] For the B-side subsystem, the sinusoidal straight stripe images are respectively displayed at the second display screen initial position 8 and the second display screen moving position 10, and then modulated and deformed by the B-side mirror 15. The second CCD camera 6 collects the deformed stripe images of the sinusoidal straight stripe images at the second display screen initial position 8 and the second display screen moving position 10 reflected by the B-side mirror 15, and stores them in the computer 4 for subsequent processing;

[0097] 7) Fringe phase calculation:

[0098] For the A-side subsystem, the zero-deformation fringe image displayed on the first display screen 18 obtained in step 5) and reflected by the A-side reference plane 16 is calculated using the best three-fringe selection method to obtain an absolute phase unfolding diagram, and then the absolute phase value of the initial position 7 of the first display screen reflected by the A-side reference plane 16 is obtained. and the absolute phase value of the first display screen moving position 9 reflected by the reference plane 16 on the A side The deformed fringe image displayed on the first display screen 18 obtained in step 6) and reflected by the mirror surface 14 on the A side is used to calculate the absolute phase unfolding diagram using the best three-fringe selection method, and then the absolute phase value of the initial position 7 of the first display screen reflected by the mirror surface 14 on the A side is obtained. and the absolute phase value of the first display screen moving position 9 reflected by the mirror surface 14 on the A side (See Figure 5 );

[0099] For the B-side subsystem, the same method as the A-side subsystem is used to obtain the absolute phase value of the second display screen initial position 8 reflected by the B-side reference plane 17. The absolute phase value of the second display screen moving position 10 reflected by the reference plane 17 on the B side The absolute phase value of the second display screen initial position 8 reflected by the mirror 15 on the B side and the absolute phase value of the second display screen moving position 10 reflected by the B side mirror 15 (See Figure 6 );

[0100] 8) Solve the three-dimensional shape data of the A-side mirror surface 14 and the B-side mirror surface 15 of the double-mirror object 3 in the corresponding coordinate system:

[0101] For the A-side subsystem, the relationship between the phase and the depth information of the A-side mirror 14 is established as shown in equation (3), and the depth information h of the A-side mirror 14 relative to the A-side reference plane 16 is obtained: A (That is, in the reference coordinate system O A -X A Y A Z A Z A coordinate):

[0102]

[0103] Then, according to the Z of each pixel of the first CCD camera 5, the mirror surface 14 on the A side A The coefficient a of the relationship between coordinates and lateral calibration A0 、a A1 、a A2 , b A0 , b A1 , b A2 , calculate each pixel point in the reference coordinate system O A -X A Y A Z A The X and Y coordinates of the mirror surface 14 on the A side in the reference coordinate system O are obtained. A -X A Y A Z A The three-dimensional coordinates (X A ,Y A ,Z A )(See Figure 7 );

[0104] For the B-side subsystem, the same method as the A-side subsystem is used to establish the relationship between the phase and the depth information of the B-side mirror 15 as shown in equation (4), and the depth information h of the B-side mirror 15 relative to the B-side reference plane 17 is obtained: B (That is, in the coordinate system to be calibrated O B -X B Y B Z B Z B coordinate):

[0105]

[0106] Then, according to the Z of each pixel of the second CCD camera 6, the B-side mirror 15 B The coefficient a of the relationship between coordinates and lateral calibration B0 、a B1 、a B2 、b B0 、b B1 、b B2 , calculate each pixel point in the coordinate system to be calibrated O B -X B Y B Z B The X and Y coordinates of the mirror surface 15 on the B side in the coordinate system to be calibrated O are obtained. B -X B Y B Z B The three-dimensional coordinates (X B ,Y B ,Z B )(See Figure 8 );

[0107] 9) 3D shape data fusion of dual mirror objects:

[0108] According to the reference coordinate system O obtained in step 2) A -X A Y A Z A With the coordinate system to be calibrated O B -X B Y B Z B The transformation matrix between By using formula (5), a point on the mirror surface 15 on the B side is located in the coordinate system to be calibrated O B -X B Y B Z B The three-dimensional coordinates (X B ,Y B ,Z B ) is converted to the reference coordinate system O A -X A Y A Z A The coordinates below

[0109]

[0110] At this time, the three-dimensional coordinates (X A ,Y A ,Z A ) and the three-dimensional coordinates of the B side mirror surface 15 Both are in the reference coordinate system O A -X A Y A Z A Thus, the overall three-dimensional shape data of the double-mirror object 3 is obtained (see Fig. 9 ).

[0111] Example

[0112] In step 1), the computer 4 is installed with software for generating fringe images and image processing (MATLAB software in this embodiment), as well as software for controlling the camera to collect and store images. These software are well-known technologies and can be commercially obtained.

[0113] The first CCD camera 5 and the second CCD camera 6 are both black-and-white industrial area array cameras produced by Daheng Image, model MER2-501-79U3M, with a resolution of 2448×2048 and a pixel size of 3.45μm×3.45μm.

[0114] The first display screen 18 and the second display screen 19 both adopt Zhixianda G1316, with a resolution of 1920×1080 and a single pixel physical size of 153.16μm×153.70μm.

[0115] The first high-precision horizontal moving guide rail 11, the second high-precision horizontal moving guide rail 12, and the third high-precision horizontal moving guide rail 13 are all GCD203300M produced by Daheng Optoelectronics Company, with a total stroke of 300mm, a single pulse resolution of 0.001mm, and a repeated positioning error of no more than 0.005mm.

[0116] In step 2), the reference coordinate system O is calculated using the rigid body transformation method. A -X A Y A Z A With the coordinate system to be calibrated O B -X B Y B Z B The transformation matrix between That is, common reference calibration parameters;

[0117] The rigid body transformation method is a well-known method, and the reference is "LaValle S M. Planning algorithms [M]. Cambridge university press, 2006".

[0118] In step 3), Zhang Zhengyou's camera-based machine vision calibration method is a well-known method, and the reference is "Zhang Z. A flexible new technique for camera calibration [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2000, 22 (11): 1330-1334".

[0119] In step 4), the numbers of the three groups of sinusoidal straight stripes generated are 81, 80, and 72 respectively;

[0120] In step 7), the absolute phase unwrapping image is calculated using an optimal three-fringe selection method; the optimal three-fringe selection method is a well-known method in the art, and the reference is "Zhang Z. Time efficient color fringe projection system for simultaneous 3D shape and color using optimum 3-frequency selection. Optics Express, 2006, 14(14): 6444-6455".

[0121] A-side subsystem: The first display screen 18 is used to display the sinusoidal straight stripes generated by the software. Since the A-side mirror 14 has a mirror reflection characteristic, these sinusoidal straight stripes are deformed by the modulation of the gradient and depth of the A-side mirror 14. The first CCD camera 5 collects the reflected deformed stripes for phase demodulation, and uses the system parameters of the A-side subsystem to solve the three-dimensional morphology of the A-side mirror 14 of the dual-mirror object 5.

[0122] B-side subsystem: The second display screen 19 is used to display the sinusoidal straight stripes generated by the software. Since the B-side mirror 15 has the mirror reflection characteristic, these sinusoidal straight stripes are deformed by the modulation of the gradient and depth of the B-side mirror 15. The second CCD camera 6 collects the reflected deformed stripes for phase demodulation, and uses the system parameters of the B-side subsystem to solve the three-dimensional morphology of the B-side mirror 15 of the dual-mirror object 5.

[0123] Any matters not described in the present invention are applicable to the prior art.

Claims

1. A method for measuring phase deflection of a common reference double-mirror object, characterized in that: The method comprises the following steps: 1) Build a common reference dual-mirror object phase deflection measurement system: The system comprises a double mirror object (3), a computer (4), a first CCD camera (5), a second CCD camera (6), a first high-precision horizontal moving guide rail (11), a second high-precision horizontal moving guide rail (12), a third high-precision horizontal moving guide rail (13), a first display screen (18) and a second display screen (19); The first high-precision horizontal moving guide rail (11) is fixed on the optical platform, and the first display screen (18) is slidably mounted on the first high-precision horizontal moving guide rail (11), is perpendicular to the optical platform, and can be linearly moved along the first high-precision horizontal moving guide rail (11) to the first display screen initial position (7) and the first display screen moving position (9); The second high-precision horizontal moving guide rail (12) is fixed on the optical platform, and the second display screen (19) is slidably mounted on the second high-precision horizontal moving guide rail (12), is perpendicular to the optical platform, and can be linearly moved along the second high-precision horizontal moving guide rail (12) to the second display screen initial position (8) and the second display screen moving position (10); The third high-precision horizontal moving guide rail (13) is fixed on the optical platform, and the double-mirror object (3) is slidably mounted on the third high-precision horizontal moving guide rail (13) and the double-mirror object (3) can rotate, so that the virtual image of the first display screen (18) reflected by the A-side mirror surface (14) of the double-mirror object (3) is within the depth of field of the first CCD camera (5), and the virtual image of the second display screen 19 reflected by the B-side mirror surface (15) of the double-mirror object (3) is within the depth of field of the second CCD camera (6); the first CCD camera (5), the first display screen (18) and the A-side mirror surface (14) are in a triangulated measurement relationship in space; the second CCD camera (6), the second display screen (19) and the B-side mirror surface (15) are in a triangulated measurement relationship in space; The computer (4), the first CCD camera (5), the first high-precision horizontal moving guide rail (11), the first display screen (18) and the A-side mirror (14) constitute the A-side subsystem; the computer (4), the second CCD camera (6), the second high-precision horizontal moving guide rail (12), the second display screen 19 and the B-side mirror (15) constitute the B-side subsystem; The computer (4) is respectively connected to the first CCD camera (5), the second CCD camera (6), the first high-precision horizontal moving guide rail (11), the second high-precision horizontal moving guide rail (12), the third high-precision horizontal moving guide rail (13), the first display screen (18) and the second display screen (19); 2) Remove the double mirror object (3) from the third high-precision horizontal guide rail (13), and then install the single mirror calibration plate on the third high-precision horizontal guide rail (13), and obtain the common reference calibration parameters of the system: When the single mirror calibration plate is used to calibrate the A-side subsystem, the first-position mirror calibration plate (1) is set; a virtual image of the first display screen (18) reflected by the first-position mirror calibration plate (1) and the first-position mirror calibration plate (1) are within the depth of field of the first CCD camera (5); the first CCD camera (5), the first display screen (18) and the first-position mirror calibration plate (1) are in a triangular measurement relationship in space; the plane where the circular ring array on the surface of the first-position mirror calibration plate (1) is located is set as the A-side reference plane (16); and a coordinate system determined by the circular ring array on the surface of the first-position mirror calibration plate (1) is set as a reference coordinate system O A -X A Y A Z A ; When the single mirror calibration plate is used to calibrate the B-side subsystem, the second posture mirror calibration plate (2) is set; the virtual image of the second display screen (19) reflected by the second posture mirror calibration plate (2) and the second posture mirror calibration plate (2) are within the depth of field of the second CCD camera (6); the second CCD camera (6), the second display screen (19) and the second posture mirror calibration plate (2) are in a triangular measurement relationship in space; the plane where the circular ring array on the surface of the second posture mirror calibration plate (2) is located is set as the B-side reference plane (17); and the coordinate system determined by the circular ring array on the surface of the second posture mirror calibration plate (2) is set as the coordinate system to be calibrated. B -X B Y B Z B ; The posture transformation mode between the first posture mirror calibration plate (1) and the second posture mirror calibration plate (2) is known, and the reference coordinate system O is calculated by using the rigid body transformation method. A -X A Y A Z A With the coordinate system to be calibrated O B -X B Y B Z B The transformation matrix between 3) Obtain the system parameters that need to be calibrated: For the A-side subsystem, the distance d between the initial position (7) of the first display screen and the A-side reference plane (16) is calculated. A ; Using the first high-precision horizontal moving guide rail (11), the first display screen (18) is moved from the first display screen initial position (7) to the first display screen moving position (9), and the moving distance is Δd A ; The first posture mirror calibration plate (1) is moved n times along the third high-precision horizontal moving guide rail (13) within the depth of field of the first CCD camera (5) A Positions, n A ≥6, the first CCD camera (5) at each position collects the circular array on the surface of the first pose mirror calibration plate (1), and records the center of each circular ring in the circular array in the reference coordinate system O A -X A Y A Z A The coordinates (X Ab ,Y Ab ,Z Ab ), and then according to the pixel coordinates of the center of each circular ring in the circular ring array extracted by the first CCD camera (5), the pixel coordinates (u A ,v A ) and X Ab and Y Ab The coefficient a in the corresponding relationship A0 、a A1 、a A2 、b A0 、b A1 、b A2 , complete the calibration of the X and Y coordinates of the A-side subsystem; In formula (1), a A0 、a A1 、a A2 、b A0 、b A1 、b A2 is the coefficient of the relationship; For the B-side subsystem, the distance d between the initial position (8) of the second display screen and the B-side reference plane (17) is calculated. B ; Using the second high-precision horizontal moving guide rail (12), the second display screen (19) is moved from the second display screen initial position (8) to the second display screen moving position (10), and the moving distance is Δd B ; The second posture mirror calibration plate (2) is moved n times along the third high-precision horizontal moving guide rail (13) within the depth of field of the second CCD camera (6). B Positions, n B ≥6, the second CCD camera (6) at each position collects the circular array on the surface of the second posture mirror calibration plate (2), and records the center of each circular ring in the circular array in the coordinate system to be calibrated O B -X B Y B Z B The coordinates (X Bb ,Y Bb ,Z Bb ), and then according to the pixel coordinates of the center of each circular ring in the circular array extracted by the second CCD camera (6), the pixel coordinates (u B ,v B ) and X Bb and Y Bb The coefficient a in the corresponding relationship B0 、a B1 、a B2 、b B0 、b B1 、b B2 , complete the calibration of the X and Y coordinates of the B-side subsystem; In formula (2), a B0 、a B1 、a B2 、b B0 、b B1 、b B2 is the coefficient of the relationship; 4) Generate sinusoidal straight fringe images; 5) Collect zero-deformation fringe images: For the A-side subsystem, the sinusoidal straight stripe images are respectively displayed at the first display screen initial position (7) and the first display screen moving position (9), and then are not deformed after being reflected by the A-side reference plane (16). The first CCD camera (5) collects zero-deformation stripe images of the sinusoidal straight stripe images at the first display screen initial position (7) and the first display screen moving position (9) after being reflected by the A-side reference plane (16), and stores them in the computer (4); For the B-side subsystem, the sinusoidal straight stripe images are respectively displayed at the second display screen initial position (8) and the second display screen moving position (10), and then are not deformed after being reflected by the B-side reference plane (17). The second CCD camera (6) collects the zero-deformation stripe images of the sinusoidal straight stripe images at the second display screen initial position (8) and the second display screen moving position (10) after being reflected by the B-side reference plane (17), and stores them in the computer (4); 6) Collect deformed fringe images: The single-mirror calibration plate is removed from the third high-precision horizontal guide rail (13), and the double-mirror object (3) is then mounted on the third high-precision horizontal guide rail (13), ensuring that the double-mirror object (3) is within the depth of field of the first CCD camera (5) and the second CCD camera (6), and that the position and posture conform to the triangulation relationship; For the A-side subsystem, the sinusoidal straight stripe images are respectively displayed at the first display screen initial position (7) and the first display screen moving position (9), and then modulated and deformed by the A-side mirror (14). The first CCD camera (5) collects the deformed stripe images of the sinusoidal straight stripe images at the first display screen initial position (7) and the first display screen moving position (9) reflected by the A-side mirror (14), and stores them in the computer (4); For the B-side subsystem, the sinusoidal straight stripe images are respectively displayed at the second display screen initial position (8) and the second display screen moving position (10), and then modulated and deformed by the B-side mirror (15). The second CCD camera (6) collects the deformed stripe images of the sinusoidal straight stripe images at the second display screen initial position (8) and the second display screen moving position (10) reflected by the B-side mirror (15), and stores them in the computer (4); 7) Fringe phase calculation: For the A-side subsystem, the zero-deformation fringe image displayed on the first display screen (18) obtained in step 5) and reflected by the A-side reference plane (16) is calculated to obtain an absolute phase unfolding diagram, and then the absolute phase value of the initial position (7) of the first display screen reflected by the A-side reference plane (16) is obtained. and the absolute phase value of the first display screen moving position (9) reflected by the A side reference plane (16) For the deformed fringe image displayed on the first display screen (18) obtained in step 6) and reflected by the mirror surface (14) on the side A, an absolute phase unfolding diagram is calculated, and then the absolute phase value of the initial position (7) of the first display screen reflected by the mirror surface (14) on the side A is obtained. and the absolute phase value of the first display screen moving position (9) reflected by the mirror surface (14) on the A side For the B-side subsystem, the same method as the A-side subsystem is used to obtain the absolute phase value of the initial position (8) of the second display screen reflected by the B-side reference plane (17) The absolute phase value of the second display screen moving position (10) reflected by the reference plane (17) on the B side The absolute phase value of the initial position (8) of the second display screen reflected by the mirror surface (15) on the B side and the absolute phase value of the second display screen moving position (10) reflected by the B side mirror (15) 8) Solve the three-dimensional shape data of the A-side mirror surface (14) and the B-side mirror surface (15) of the double-mirror object (3) in the corresponding coordinate system: For the A-side subsystem, the relationship between the phase and the depth information of the A-side mirror (14) is established as shown in equation (3), and the depth information h of the A-side mirror (14) relative to the A-side reference plane (16) is obtained: A : Then, according to the Z value of each pixel point of the A side mirror (14) at the first CCD camera (5), A The coefficient a of the relationship between coordinates and lateral calibration A0 、a A1 、a A2 、b A0 、b A1 、b A2 , calculate each pixel point in the reference coordinate system O A -X A Y A Z A The X and Y coordinates of the mirror surface (14) on the A side in the reference coordinate system O are obtained. A -X A Y A Z A The three-dimensional coordinates (X A ,Y A ,Z A ); For the B-side subsystem, the same method as the A-side subsystem is used to establish the relationship between the phase and the depth information of the B-side mirror (15) as shown in equation (4), and the depth information h of the B-side mirror (15) relative to the B-side reference plane (17) is obtained: B : Then, according to the Z of each pixel of the second CCD camera (6) at the B side mirror (15), B The coefficient a of the relationship between coordinates and lateral calibration B0 、a B1 、a B2 、b B0 、b B1 、b B2 , calculate each pixel point in the coordinate system to be calibrated O B -X B Y B Z B The X and Y coordinates of the mirror surface (15) on the B side in the coordinate system to be calibrated O are obtained. B -X B Y B Z B The three-dimensional coordinates (X B ,Y B ,Z B ); 9) 3D shape data fusion of dual mirror objects: According to the reference coordinate system O obtained in step 2) A -X A Y A Z A With the coordinate system to be calibrated O B -X B Y B Z B The transformation matrix between By using equation (5), a point on the mirror surface (15) on the B side is located in the coordinate system to be calibrated O B -X B Y B Z B The three-dimensional coordinates (X B ,Y B ,Z B ) is converted to the reference coordinate system O A -X A Y A Z A The coordinates below At this time, the three-dimensional coordinates (X A ,Y A ,Z A ) and the three-dimensional coordinates of the mirror surface (15) on the B side Both are in the reference coordinate system O A -X A Y A Z A Thus, the overall three-dimensional morphological data of the double-mirror object (3) is obtained.

2. The method for measuring phase deflection of a common reference dual-mirror object according to claim 1, characterized in that: In step 3), the distance d between the initial position (7) of the first display screen and the reference plane (16) on the A side is calculated. A Specifically, a circular array with known spatial distance and radius is displayed on a first display screen (18) located at an initial position (7) of the first display screen, a first CCD camera (5) collects the circular array of the first position mirror calibration plate (1) and the circular array of the first display screen (18) reflected by the first position mirror calibration plate (1), extracts the boundary of the circular array, determines the center of the circular array by ellipse fitting, and calculates the distance d between the initial position (7) of the first display screen and the reference plane (16) on the A side by using Zhang Zhengyou's camera-based machine vision calibration method. A .

3. The method for measuring phase deflection of a common reference dual-mirror object according to claim 1, characterized in that: In step 3), the distance d between the initial position (8) of the second display screen and the reference plane (17) on the side B is calculated. B Specifically, a circular ring array with known spatial distance and radius is displayed on a second display screen (19) located at an initial position (8) of the second display screen, a second CCD camera (6) collects the circular ring array of the second posture mirror calibration plate (2) and the circular ring array of the second display screen (19) reflected by the second posture mirror calibration plate (2), extracts the boundary of the circular ring, determines the center of the circular ring by ellipse fitting, and uses Zhang Zhengyou's camera-based machine vision calibration method to calculate the distance d between the initial position (8) of the second display screen and the B-side reference plane (17). B .

4. The method for measuring phase deflection of a common reference dual-mirror object according to claim 1, characterized in that: Step 4) specifically comprises: using a computer (4) to select three groups of sinusoidal straight stripes to be displayed in the measurement field according to the size of the measurement field and the measurement accuracy requirements, and the number of stripes in the three groups of sinusoidal straight stripes meets the optimal three-stripe selection method; Each set of sinusoidal straight stripes contains four sinusoidal straight stripe images with a phase shift of 90° between each other.

Images