A method for three-dimensional measurement based on three-ocular system under inclined stripe coding

Abstract

This invention discloses a three-dimensional measurement method based on tilted fringe coding in a trinocular system. First, the distance from the phase plane to the optical axis coordinates is calculated based on a linear imaging model. Then, the vector expression passing through the optical center within the phase plane is obtained through the principle of vector superposition. This expression is then cross-multiplied with the phase line vector to obtain the phase plane normal vector. Next, the plane equation of the phase plane is established based on the phase value. Then, new constraint equations are established based on the relationship between the three-dimensional coordinate points and the phase plane. Finally, the three-dimensional coordinates of the object are calculated by combining the two constraint equations of the pinhole camera imaging model. This invention first derives the constraint equations between the three-dimensional coordinates and the phase plane based on a linear abstract model, and then solves for the three-dimensional coordinates through the constraint conditions of the pinhole camera imaging model. This significantly improves the measurement accuracy of the system when the camera's viewing angle, the central projection optical path, and the baseline direction are not the same in a trinocular system.

Description

Technical Field

[0001] This invention belongs to the field of structured light three-dimensional measurement, specifically a tilted stripe coding three-dimensional measurement method based on a trinocular system. Background Technology

[0002] In the field of structured light 3D measurement, the integrity of the measured sample is a crucial indicator. To improve this integrity, traditional methods typically employ a turntable or robotic arm to perform a 360-degree scan around the sample. However, these methods invariably require stitching together the calculated 3D point cloud data, which is time-consuming. Furthermore, they necessitate complex mechanical structures to move the turntable or measuring device during scanning, demanding a high level of expertise. Therefore, finding a structured light 3D measurement technology that is simple in structure, fast, and significantly improves the integrity of the measured sample is of paramount importance. Summary of the Invention

[0003] In order to improve the measurement accuracy of samples while maximizing the integrity of sample measurements and reducing computational complexity, this invention provides a three-dimensional measurement method based on tilted stripe coding under a trinocular system.

[0004] The technical solution adopted in this invention is as follows: A three-dimensional measurement method based on tilted fringe coding under a trinocular system, comprising the following steps:

[0005] Step 1: Arrange three cameras evenly around the projector and project a tilted, coded structured light pattern.

[0006] Step 2: Calculate the distance vector from the principal point of the projector target surface to the equiphase line. Calculate the distance vector from the principal point of the projector target surface to the equiphase line based on the tilt angle of the projected pattern.

[0007] Step 3: Add the vector from the point where the optical center points to the point where the projector target surface intersects with the optical axis to the vector from the point where the projector target surface intersects with the optical axis to the equiphase surface, and obtain the vector expression for the phase surface passing through the optical center and pointing to the projector target surface.

[0008] Step 4: Cross-multiply the vector expression of the phase plane passing through the optical center with the equiphase line on the projector target surface to obtain the normal vector of the equiphase plane and the plane equation of the equiphase plane.

[0009] Step 5: Combine the two imaging equations under the pinhole model with the plane equation of the equiphase surface to form three equations to solve for the three-dimensional coordinates.

[0010] Preferably, the calculation of the distance vector from the principal point of the projector target surface to the equiphase line in step two is as follows:

[0011] The distance vector is represented as:

[0012]

[0013] Where α is the tilt angle of the projected pattern stripes. These represent the phases of the equiphase surface and the principal point position, respectively.

[0014] The spatial distance occupied by a unit phase is calculated using the following formula:

[0015]

[0016] L represents the total length of the target surface in the vertical stripe direction, and N represents the number of stripe periods in the projected pattern.

[0017] Preferably, the solution for the vector passing through the optical center and perpendicularly pointing to the DMD target surface in step three, in the case of equal phase, is as follows:

[0018] Add the vector passing through the optical center and pointing to the principal point of the DMD target surface to the distance vector obtained in step one to obtain the vector in phase that passes through the optical center and is perpendicular to the DMD target surface:

[0019]

[0020]

[0021] Among them, f p This is the actual focal length of the projector lens.

[0022] Preferably, the solution to the plane equation of the isophase surface in step four is as follows:

[0023] The normal vector of the equiphase plane is obtained by cross product of two mutually perpendicular vectors within the equiphase plane:

[0024]

[0025] Among them, l p The vector containing the equiphase lines on the projector target surface can be represented as:

[0026] l p =(sinα,cosα,0) (6)

[0027] This yields the plane equation for the equiphase surface:

[0028] k a x+k b y+k c z = 0 (7).

[0029] Preferably, the solution for the three-dimensional coordinates in step five is as follows:

[0030] Substituting the three-dimensional coordinates of the object on the equiphase surface into the equation of the equiphase surface yields a constraint condition:

[0031] k a X P +k b Y P +k c Z P =0 (8)

[0032] Starting from the pinhole camera imaging model, two constraints can be obtained for solving the three-dimensional coordinates:

[0033]

[0034] Among them, (X) c ,Y c Z c f represents the object's 3D coordinates in the camera coordinate system. x_c with f y_c u is the focal length of the camera lens. c With v c These are the camera's pixel coordinates, u 0_c With v 0_c These are the principal points on the camera target surface;

[0035] Based on the camera calibration results, convert the object coordinates in the projector coordinate system to the object coordinates in the camera coordinate system:

[0036]

[0037] Among them, T c and These are the extrinsic parameter matrices for the camera and the projector, respectively. Represents the pseudo-inverse of a matrix;

[0038] By combining the two constraints in the pinhole camera imaging model, the three-dimensional coordinates of the object can be obtained.

[0039] The present invention has the following advantages: the present invention adopts a camera arrangement of three cameras around the projector, which greatly improves the measurement integrity of the system; the present invention uses tilted stripe coding to modulate the sample, which improves the measurement accuracy of the system. Attached Figure Description

[0040] Figure 1 This is a system block diagram of an embodiment of the present invention.

[0041] Figure 2 This is a schematic diagram of a model according to an embodiment of the present invention.

[0042] Figure 3 These are the structured light projection patterns of the three cameras in this embodiment of the invention.

[0043] Figure 4This is a diagram of the tilted coded imaging model in an embodiment of the present invention. Detailed Implementation

[0044] The present invention will be further described below with reference to the embodiments and accompanying drawings.

[0045] The process steps in this embodiment are as follows: Figure 1 As shown, three cameras are first evenly arranged around the projector to project a tilted, coded structured light pattern. The distance vector from the principal point on the projector's target surface to the equiphase line is then calculated. And calculate the distance vector from the principal point of the projector target surface to the equiphase line based on the tilt angle of the projected pattern. Add the vector pointing from the optical center to the point where the projector target surface intersects with the optical axis to the vector pointing from the point where the projector target surface intersects with the optical axis to the equiphase surface, and you get the vector expression for the phase surface passing through the optical center and pointing to the projector target surface. By cross-multiplying the vector expression of the phase plane passing through the optical center with the equiphase line on the projector target surface, we obtain the normal vector of the equiphase plane and the plane equation k of the equiphase plane. a x+k b y+k c z = 0. The two imaging equations under the combined pinhole model and the plane equation of the equiphase surface are combined into three equations to solve for the three-dimensional coordinates (X). c ,Y c Z c ).

[0046] The specific steps of the embodiment are as follows:

[0047] Step 1, follow as follows Figure 2 The camera and projector are arranged as shown. The three cameras employ... Figure 3 The structured light pattern shown is coded with a tilted shadow.

[0048] Step two: Calculate the distance vector from the principal point on the projector target surface to the equiphase line. The distance vector can be expressed as:

[0049]

[0050] Where α is the tilt angle of the projected pattern stripes. These represent the phases of the equiphase surface and the principal point position, respectively.

[0051] The spatial distance occupied by a unit phase can be calculated using the following formula:

[0052]

[0053] L represents the total length of the target surface in the vertical stripe direction, and N represents the number of stripe periods in the projected pattern.

[0054] Step 3: Solve for the vector passing through the optical center and perpendicularly pointing to the DMD target surface in the equiphase surface.

[0055] Add the vector passing through the optical center and pointing to the principal point of the DMD target surface to the distance vector obtained in step one to obtain the vector in phase that passes through the optical center and is perpendicular to the DMD target surface:

[0056]

[0057] in:

[0058]

[0059] f p This is the actual focal length of the projector lens.

[0060] Step four: Solving the equation of the equiphase surface. The normal vector of the equiphase surface is obtained by cross product of two mutually perpendicular vectors within the equiphase surface:

[0061]

[0062] Among them l p Let be the vector containing the equiphase lines on the projector target surface. This can be represented as:

[0063] l p =(sinα,cosα,0) (6)

[0064] This yields the plane equation for the equiphase surface:

[0065] k a x+k b y+k c z = 0 (7)

[0066] Step 5, Solving for the 3D coordinates. Substituting the 3D coordinates of the object on the equiphase surface into the equation of the equiphase surface yields a constraint condition:

[0067] k a X P +k b Y P +k c Z P =0 (8)

[0068] Starting from the pinhole camera imaging model, two constraints can be obtained for solving the three-dimensional coordinates:

[0069]

[0070] Where (X) c ,Y c Z cf represents the object's three-dimensional coordinates in the camera coordinate system. x_c with f y_c This refers to the focal length of the camera lens. c With v c These are the pixel coordinates of the camera. 0_c With v 0_c These are the principal points of the camera target surface. Based on the camera calibration results, the object coordinates (X, Y) in the projector coordinate system are... P ,Y P Z P Convert to object coordinates in camera coordinate system (X) c ,Y c Z c ):

[0071]

[0072] Where T c and These are the extrinsic parameter matrices for the camera and the projector, respectively. This represents the pseudo-inverse of the matrix. Combining the two constraints in the pinhole camera imaging model, the three-dimensional coordinates of the object can be obtained.

[0073] Obviously, the above embodiments of the present invention are merely illustrative examples to illustrate the invention and are not intended to limit the implementation of the invention. Other obvious variations or modifications derived from the essential spirit of the invention still fall within the protection scope of the invention.

Claims

1. A three-dimensional measurement method based on a triocular system with tilted fringe coding, characterized in that... The steps are as follows: Step 1: Arrange three cameras evenly around the projector and project a tilted, coded structured light pattern. Step 2: Calculate the distance vector from the principal point of the projector target surface to the equiphase line. Calculate the distance vector from the principal point of the projector target surface to the equiphase line based on the tilt angle of the projected pattern. Step 3: Add the vector from the point where the optical center points to the point where the projector target surface intersects with the optical axis to the vector from the point where the projector target surface intersects with the optical axis to the equiphase surface, and obtain the vector expression for the phase surface passing through the optical center and pointing to the projector target surface. Step 4: Cross-multiply the vector expression of the phase plane passing through the optical center with the equiphase line on the projector target surface to obtain the normal vector of the equiphase plane and the plane equation of the equiphase plane. Step 5: Combine the two imaging equations under the pinhole camera imaging model with the plane equation of the equiphase surface to form three equations to solve for the three-dimensional coordinates.

2. The three-dimensional measurement method based on a triocular system with tilted fringe coding according to claim 1, characterized in that... The calculation of the distance vector from the principal point of the projector target surface to the equiphase line in step two is as follows: The distance vector is represented as: Equation (1) in, The tilt angle of the projected pattern stripes. , These represent the phases of the equiphase surface and the principal point position, respectively. The spatial distance occupied by a unit phase is calculated using the following formula: Equation (2) This refers to the total length of the target surface along the vertical stripe direction. The number of periods of the projected pattern stripes.

3. The three-dimensional measurement method based on a triocular system with tilted fringe coding according to claim 1, characterized in that... Step 3, solving for the vector passing through the optical center and perpendicularly pointing to the DMD target surface in the equal phase, specifically involves: Add the vector passing through the optical center and pointing to the principal point of the DMD target surface to the distance vector obtained in step one to obtain the vector in phase that passes through the optical center and is perpendicular to the DMD target surface: Equation (3) Equation (4) in, This is the actual focal length of the projector lens.

4. The three-dimensional measurement method based on a triocular system with tilted fringe coding according to claim 1, characterized in that... Step four involves solving the plane equations for the equal-phase surface as follows: The normal vector of the equiphase plane is obtained by cross product of two mutually perpendicular vectors within the equiphase plane: Equation (5) in, The vector containing the equiphase lines on the projector target surface can be represented as: Equation (6) This yields the plane equation for the equiphase surface: Equation (7).

5. The three-dimensional measurement method based on a triocular system with tilted fringe coding according to claim 4, characterized in that... The specific steps for solving the three-dimensional coordinates in step five are as follows: Substituting the three-dimensional coordinates of the object on the equiphase surface into the plane equation of the equiphase surface yields a constraint condition: Equation (8) Starting from the pinhole camera imaging model, we can obtain two imaging equations for solving the three-dimensional coordinates, and thus obtain two constraints: Equation (9) in, The object's three-dimensional coordinates in the camera coordinate system. and The focal length of the camera lens. and These are the pixel coordinates of the camera. and These are the principal points on the camera target surface; Based on the camera calibration results, convert the object coordinates in the projector coordinate system to the object coordinates in the camera coordinate system: Equation (10) in, and These are the extrinsic parameter matrices for the camera and the projector, respectively. " denotes the pseudo-inverse of a matrix; By combining the two constraints in the pinhole camera imaging model, the three-dimensional coordinates of the object can be obtained.

Images