A single-camera multi-projection structured light measurement system and three-dimensional reconstruction method

Abstract

The present invention discloses a single-camera multi-projection structured light measurement system and a three-dimensional reconstruction method, which relate to the field of structured light measurement technology. The method includes: a structured light projector emits structured light, projects structured light stripes with equal width within the full frame on the surface of the object to be measured, and controls the structured light stripes to move at a uniform speed; a high-speed acquisition camera acquires structured light stripe images on the surface of the object to be measured; reconstructs multiple height maps of the object to be measured based on the acquired structured light stripe images; identifies occluded noise areas and unobstructed true height areas in the multiple height maps, and removes the occluded noise areas from the multiple height maps; fuses the unobstructed true height areas in the multiple height maps to obtain a complete height map, and realizes three-dimensional reconstruction of the object to be measured based on the obtained height map. The present invention uses a single structured light projector to achieve the effect of multiple projections, which not only reduces costs but also improves the consistency of multiple projections.

Description

Technical Field

[0001] The present invention relates to the field of structured light measurement technology, and in particular to a single-camera multi-projection structured light measurement system and a three-dimensional reconstruction method. Background Art

[0002] Structured light 3D measurement technology is also known as active triangulation. Its measurement principle is to use a projector to project a coded pattern onto the surface of the measured object, and the acquisition camera obtains the deformed stripe image modulated by the object's morphology. The modulated image contains the object's height information, thereby obtaining the object's 3D morphology. It is usually composed of multiple projectors and multiple cameras, which can capture images of the object from different angles, and use the images captured from multiple perspectives and structured light patterns to reconstruct the object's 3D model.

[0003] Traditional measurement methods for structured light 3D reconstruction often rely on a single camera with multiple projections, or a single projection with multiple cameras. Due to the angular relationship between the projections and the camera and the object, a single projection and camera cannot measure every position of an object, resulting in the loss of certain height information. Conventional 3D reconstruction methods can only add pseudo-color to the object based on the height information after measuring the 3D information. To address these technical challenges, researchers in this field urgently need to develop a low-cost, blind-sight-free, and more accurate 3D reconstruction device and method. Summary of the Invention

[0004] The invention provides a single-camera multi-projection structured light measurement system, comprising: a structured light projector (1), a movable reflective structure group (2) and a structured light acquisition group (3); the movable reflective structure group (2) comprises a linear motor (21), a reflective mirror group (22) and a reflective prism group (23); the structured light acquisition group (3) comprises a high-speed acquisition camera (31) and a remote reflective prism (32); a plurality of reflective mirrors in the reflective mirror group (22) and a plurality of reflective prisms in the reflective prism group (23) are uniformly distributed along the circumferential direction; the linear motor (21) is connected to the reflective prism group (23) to control the position of the reflective prism group (23); the linear motor (21) moves the reflective prism group (23) and the reflective prism group (23) is uniformly distributed along the circumferential direction; The prism group (23) moves to the bottom of the structured light projector (1), the reflection prism group (23) reflects the structured light emitted by the structured light projector (1), and reflects the structured light to the corresponding reflection mirror group (22). The reflection mirror group (22) projects the structured light onto the object to be measured. The structured light converges on the surface of the object to be measured and projects to form full-width phase-coded stripes. The far-end reflection prism (32) is set directly above the object to be measured. The telecentric lens (311) on the high-speed acquisition camera (31) captures the coded stripe image on the surface of the object to be measured through the far-end reflection prism (32). The coded stripe image is subjected to structured light three-dimensional reconstruction to obtain a three-dimensional point cloud of the object to be measured.

[0005] A single-camera multi-projection structured light measurement system as described above, wherein the reflector group (22) includes a plurality of reflectors, and the corresponding reflective prism group (23) also includes a plurality of reflective prisms, the number of reflectors and reflective prisms is the same, and both are evenly distributed along the circumference, the structured light reflected from the reflective prisms can be transmitted to the corresponding reflectors, and all the reflectors can converge the structured light onto the object to be measured.

[0006] A single-camera multi-projection structured light measurement system as described above, wherein the reflector group (22) includes four reflectors, specifically including a first reflector (221), a second reflector (222), a third reflector (223) and a fourth reflector (224), the four reflectors are evenly distributed at 90 degrees along the circumference, and are respectively used to converge structured light reflected from four different directions onto the object to be measured; correspondingly, the reflective prism group (23) also includes four reflective prisms, specifically including a first reflective prism (231), a second reflective prism (232), a third reflective prism (233) and a fourth reflective prism (234), the four reflective prisms are also evenly distributed at 90 degrees along the circumference, and are respectively used to reflect structured light in four different directions.

[0007] A single-camera multi-projection structured light measurement system as described above, wherein, during measurement, the structured light projector (1) is first turned on to project coded structured light, and the coded structured light presents a striped image when projected onto an object. Subsequently, a linear motor (21) is used to sequentially move each reflective prism to the bottom of the structured light projector (1), and a high-speed acquisition camera (31) is used to respectively acquire structured light stripe images in different directions, and the structured light stripe images serve as the data basis for structured light three-dimensional reconstruction.

[0008] The present invention further provides a single-camera multi-projection three-dimensional reconstruction method, which is applied to any of the structured light measurement systems described above. The three-dimensional reconstruction method includes:

[0009] Step 1: The structured light projector emits structured light, which is reflected by prisms and mirrors in different directions and projects structured light stripes with equal width within the full frame on the surface of the object being measured. The structured light stripes are controlled to move at a uniform speed.

[0010] Step 2: A high-speed acquisition camera collects the structured light stripe images of the surface of the object being measured after being reflected by the prisms and mirrors in different directions;

[0011] Step 3: Reconstruct multiple height maps of the measured object based on the collected structured light fringe images;

[0012] Step 4: Identify the occluded noise areas and unoccluded true height areas in multiple height maps, and remove the occluded noise areas from the multiple height maps;

[0013] Step 5: Fuse the unobstructed real height areas in multiple height maps to obtain a complete height map, and perform 3D reconstruction of the measured object based on the obtained height map.

[0014] The single-camera multi-projection 3D reconstruction method described above, wherein a structured light projector emits structured light, and based on a structured light fringe generation algorithm, the period and phase offset of the fringe are precisely controlled when generating the fringe, specifically includes the following sub-steps:

[0015] Create an optimization function for fringe period and phase offset;

[0016] The optimal solution of fringe period and phase offset is obtained based on the optimization function;

[0017] Structured light fringes are generated based on the optimal solution of fringe period and phase offset.

[0018] The single-camera multi-projection 3D reconstruction method described above, wherein multiple height maps of the object to be measured are reconstructed based on the collected structured light fringe images, specifically includes the following sub-steps:

[0019] A phase value is created based on the grayscale values ​​of pixels in the structured light fringe image to solve the equation group;

[0020] Solve the equations in the system of equations by simultaneous phase value calculation to obtain the phase value of each point of the object being measured;

[0021] The height of the object surface is calculated based on the phase value of each point on the measured object, and four height maps are generated.

[0022] The single-camera multi-projection 3D reconstruction method described above includes identifying occluded noise regions and unobstructed true height regions in multiple height maps, and removing the occluded noise regions from the multiple height maps, specifically comprising the following sub-steps:

[0023] Extract the gradient features of each pixel in the structured light stripe image;

[0024] Identify the label of the region to which the pixel belongs based on the gradient characteristics of each pixel;

[0025] In the height map, remove the pixels corresponding to the occluded noise area.

[0026] The single-camera multi-projection 3D reconstruction method described above, wherein the unobstructed true height regions in multiple height maps are fused to obtain a complete height map, specifically includes the following sub-steps:

[0027] Transform the four height maps into a common reference coordinate system;

[0028] Remove the duplicate points from the four heightmaps and concatenate the remaining unique points into one heightmap.

[0029] The present invention achieves the following beneficial effects: It utilizes a single structured light projector and a single high-speed acquisition camera, using a linear motor and a reflective optical path, to implement a single-camera, multi-projection measurement system. This single-camera, multi-projection measurement system projects structured light onto the surface of the object being measured through multiple reflections of the structured light emitted by the single structured light projector. This enables comprehensive 3D reconstruction and measurement of the object, resulting in more accurate, non-obstructed, and dark-angled measurement results with higher and more precise measurements. The present invention achieves this multi-projection effect using a single structured light projector, reducing costs and improving the consistency of multiple projections. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments described in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0031] Figure 1 and Figure 2 This is a schematic diagram of a single-camera multi-projection structured light measurement system provided in Example 1 of the present application;

[0032] Figure 3 This is a flow chart of a single-camera multi-projection three-dimensional reconstruction method provided in Example 2 of the present application.

[0033] Reference numerals:

[0034] 1. Structured light projector; 2. Movable reflective structure group; 3. Structured light acquisition group; 21. Linear motor; 22. Reflector group; 23. Reflective prism group; 31. High-speed acquisition camera; 32. Remote reflective prism; 311. Telecentric lens; 221. First reflector; 222. Second reflector; 223. Third reflector; 224. Fourth reflector; 231. First reflector; 232. Second reflector; 233. Third reflector; 234. Fourth reflector. DETAILED DESCRIPTION

[0035] The following is a clear and complete description of the technical solutions in the embodiments of the present invention, in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0036] Example 1

[0037] like Figure 1 As shown, the first embodiment of the present application provides a single-camera multi-projection structured light measurement system, comprising: a structured light projector 1, a movable reflective structure group 2, and a structured light acquisition group 3. The movable reflective structure group 2 includes a linear motor 21, a reflective mirror group 22, and a reflective prism group 23. The structured light acquisition group 3 includes a high-speed acquisition camera 31 and a remote reflective prism 32. The multiple reflectors in the reflective mirror group 22 and the multiple reflective prisms in the reflective prism group 23 are uniformly distributed along the circumference. The linear motor 21 is connected to the reflecting prism group 23 to control the position of the reflecting prism group 23. The linear motor 21 moves the reflecting prism group 23 to the bottom of the structured light projector 1. The reflecting prism group 23 reflects the structured light emitted by the structured light projector 1 and reflects the structured light to the corresponding reflecting mirror group 22. The reflecting mirror group 22 projects the structured light onto the object to be measured. After the structured light converges on the surface of the object to be measured, it is projected to form full-width phase-coded stripes. The far-end reflecting prism 32 is set directly above the object to be measured. The telecentric lens 311 on the high-speed acquisition camera 31 captures the coded stripe image on the surface of the object to be measured through the far-end reflecting prism 32, and performs structured light three-dimensional reconstruction on the coded stripe image to obtain a three-dimensional point cloud of the object to be measured.

[0038] In the embodiment of the present application, the reflector group 22 includes a plurality of reflectors, and the corresponding reflective prism group 23 also includes a plurality of reflective prisms. The number of reflectors and reflective prisms is the same, and both are evenly distributed along the circumference. It should be noted that the number of reflector groups 22 and reflective prism groups 23 can be set arbitrarily, and it is necessary to ensure that the structured light reflected from the reflective prisms can be transmitted to the corresponding reflectors, and that all reflectors can converge the structured light onto the object being measured.

[0039] Figure 1 Take four reflectors as an example, see Figure 2 The reflector assembly 22 includes a first reflector 221, a second reflector 222, a third reflector 223, and a fourth reflector 224. The four reflectors are evenly spaced at 90° along the circumference and are respectively used to converge structured light reflected from four different directions onto the object being measured. Correspondingly, four reflective prisms are provided, including a first reflective prism 231, a second reflective prism 232, a third reflective prism 233, and a fourth reflective prism 234. The four reflective prisms are also evenly spaced at 90° along the circumference and are respectively used to reflect structured light in four different directions.

[0040] During measurement, the structured light projector 1 is first turned on to project the encoded structured light. When the encoded structured light is projected on the object, a striped image is presented. Then, the linear motor 21 is used to move each reflective prism to the bottom of the structured light projector 1 in sequence. For example, the first reflective prism 231 is first moved to the bottom of the structured light projector 1. The structured light emitted by the structured light projector 1 passes through the first reflective prism 231 to the first reflective mirror 221, and is reflected to the surface of the object to be measured by the first reflective mirror 221. Then, the high-speed acquisition camera 31 is used to collect the structured light through the telecentric lens 311. The highly modulated structured light stripe image is collected with the remote reflection prism 32. At this point, the single structured light stripe image of one direction of the object to be measured is collected. Then, the linear motor 21 is used to adjust the positions of the second reflection prism 232, the third reflection prism 233, and the fourth reflection prism 234 in sequence to project the structured light in the other three directions onto the surface of the object to be measured. The high-speed collection camera 31 then synchronously collects the structured light stripe images in the other three directions, thereby realizing the single-camera multi-projection structured light stripe image collection. These images provide the data basis for the structured light 3D reconstruction method.

[0041] Example 2

[0042] like Figure 3 As shown, the second embodiment of the present application provides a single-camera multi-projection 3D reconstruction method, including:

[0043] Step S310: The structured light projector emits structured light, projects structured light stripes of equal width within the full frame on the surface of the object being measured after being reflected by reflection prisms and mirrors in different directions, and controls the structured light stripes to move at a uniform speed;

[0044] To ensure that the width of the structured light stripes is equal within the full frame, the structured light projector emits structured light. Based on the structured light stripe generation algorithm, the period and phase offset of the stripes are precisely controlled when generating the stripes. Specifically:

[0045] Step 311: creating an optimization function for fringe period and phase offset;

[0046] The optimization function is expressed as: Where T is the fringe period to be solved, φ is the phase offset to be solved, α and β are the weight coefficients of decoding efficiency and decoding error, respectively, which are used to balance the weight between decoding accuracy and coding efficiency, k1 and k2 are constant terms used when the system generates structured light fringes, k3 is the error amplitude, and γ is the exponent of the error growth rate.

[0047] Step 312: Obtaining the optimal solution of fringe period and phase offset based on the optimization function;

[0048] Continuously adjust the fringe period and phase offset in the optimization function, and observe the changing curve of the optimization function return value. The fringe period and phase offset corresponding to the minimum return value are the optimal solution.

[0049] Step 313: Generate structured light fringes based on the optimal solution of fringe period and phase offset;

[0050] Use the system's preset structured light stripe generation tool class to generate and save structured light stripes based on the optimal solution of stripe period and phase offset.

[0051] Step S320: The high-speed acquisition camera acquires the structured light fringe images of the surface of the object being measured after being reflected by the reflection prisms and the reflector in different directions;

[0052] The single-camera, multi-projector structured light measurement system can capture deformed images of structured light stripes after being modulated by the object's height from four directions. When capturing structured light stripe images in a single direction, the structured light projector controls the projected stripes to move at a constant speed, capturing images every quarter of a cycle. Therefore, four structured light stripe images are captured in each direction, for a total of four directions and sixteen images. Different structured light stripe images are marked with the capture direction and capture order for easy distinction.

[0053] Step S330: reconstructing multiple height maps of the measured object based on the collected structured light fringe images;

[0054] Based on the collected structured light fringe image, the depth value corresponding to each pixel can be calculated. This depth value is the height of the object surface. Specifically:

[0055] Step 331: creating a phase value solving equation group according to the grayscale values ​​of pixels in the structured light fringe image;

[0056] The grayscale value corresponding to the object point (x, y) in the structured light fringe image is expressed as I(x, y):

[0057]

[0058] Among them, A(x,y) is the intensity of the ambient light, B(x,y) is the light wave amplitude value related to the projection light intensity, is the phase value corresponding to the object point (x, y), and θ is the phase shift of the phase shift fringe.

[0059] When collecting structured light stripe images, each time the structured light stripe moves cycles, which means the phase shift each time is Then the phase value solution equation group is expressed as:

[0060]

[0061] Among them, I1, I2, I3, and I4 respectively represent the grayscale values ​​corresponding to the object point (x, y) in the four structured light fringe images in a single acquisition direction.

[0062] Step 332: Solve the equations in the system of equations by simultaneous phase value calculation to obtain the phase value of each point of the object under test;

[0063] Using the formula: Determine the phase value corresponding to each point (x, y) of the object, where I1, I2, I3, and I4 represent the grayscale value corresponding to the object point (x, y) in the four structured light fringe images in a single acquisition direction.

[0064] Since the inverse tangent function is used in the phase calculation formula, the obtained phase value range is [-π, π]. This phase is called a wrapped phase or truncated phase. In order to reconstruct a continuous phase distribution, it is necessary to unfold the wrapped phase. By comparing the truncated phase values ​​of two adjacent pixels, the continuous phase is restored by adding or subtracting 2kπ.

[0065] Step 333: Calculate the height of the object surface according to the phase value of each point of the measured object and generate four height maps;

[0066] Using the formula To calculate the height h of each point of the measured object, AC represents the distance between the calculated point A and the adjacent point C, β is the angle between point C and the light projected by the structured light projector, and λ is the fringe interval. is the phase difference between point A and point C.

[0067] The four structured light stripe images in each direction can generate a corresponding height map, and different height maps are marked with different acquisition directions.

[0068] Step S340: Identify the obstructed noise areas and the unobstructed true height areas in the multiple height maps, and remove the obstructed noise areas from the multiple height maps;

[0069] The collected structured light stripe images are contrast enhanced by using filters and histogram adjustments, and then the obstructed noise areas and unobstructed true height areas are identified. Specifically:

[0070] Step 341: extracting the gradient features of each pixel in the structured light fringe image;

[0071] Using the formula: Extract the gradient features of the pixel points and store them as a feature set P, where G(u,v) is the gradient feature of the pixel point (u,v), grad X (u,v) represents the gradient of the pixel (u,v) on the X axis, gradY (u,v) represents the gradient of pixel (u,v) on the Y axis.

[0072] Step 342: Identify the label of the region to which the pixel belongs based on the gradient feature of each pixel;

[0073] The maximum and minimum values ​​of the pixel gradient features in the known occluded noise areas in each historical height map are taken to form the first recognition domain. The maximum and minimum values ​​of the pixel gradient features in the known unobstructed true height areas in each historical height map are then taken to form the second recognition domain. Then, it is determined to which recognition domain the gradient features of each pixel in the feature set P belong. If it belongs to the first recognition domain, the pixel is labeled as an occluded noise area. If it belongs to the second recognition domain, the pixel is labeled as an unobstructed true height area.

[0074] Step 343: Remove the points corresponding to the pixels labeled as the occluded noise area in the height map.

[0075] Return to see Figure 3 , step S350: fusing the unobstructed real height areas in the multiple height maps to obtain a complete height map, and realizing three-dimensional reconstruction of the measured object based on the obtained height map.

[0076] Each of the four height maps contains part of the object, so the four height maps need to be merged into a complete height map. Specifically:

[0077] Step 351: transform the four height maps into a common reference coordinate system;

[0078] Based on the image processing algorithm in the OpenCV library, the key points in the four height maps are detected and matched. Based on these matching points, the transformation matrix between the images is calculated using RANSAC or other robust methods. The calculated transformation matrix is ​​then applied to transform all images into a common reference coordinate system.

[0079] Step 352: Remove the duplicate points in the four height maps, and combine the remaining unique points into one map;

[0080] Although the four height maps are generated based on different acquisition directions, there are inevitably duplicate points. When removing duplicate points, you can first determine which acquisition direction the point is closer to. Then retain the point in the height map of the closer acquisition direction as the unique point, remove other duplicate points, and finally splice all the unique points together to get a complete height map.

[0081] Determining the height map is the most important step in 3D reconstruction. 3D reconstruction of the object under test is achieved based on the obtained height map, specifically including:

[0082] After determining the height map, the spatial coordinates (X, Y, Z) of each sampling point encoded in the measured object are obtained, and the resulting set of points is called a point cloud. The point cloud is further simplified into a mesh form. The mesh consists of a collection of vertices and polygons, usually triangles, quadrilaterals or other simple convex polygons. The point cloud data is converted into a polyhedron shape, and then the image is covered on the surface of the polyhedron to achieve texture mapping, which increases the details and realism of the model. The final three-dimensional model is the result of the combination of the polyhedron and the mapping or rendering.

[0083] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent replacements, improvements, etc. made on the basis of the technical solution of the present invention should be included in the scope of protection of the present invention.

Claims

1. A single-camera multi-projection 3D reconstruction method, characterized in that: include: Step 1: The structured light projector emits structured light, which is reflected by prisms and mirrors in different directions. The structured light stripes are then projected onto the surface of the object. The stripes are then moved at a constant speed. Based on the structured light stripe generation algorithm, the period and phase offset of the stripes are precisely controlled during stripe generation. This process includes the following sub-steps: Create an optimization function for fringe period and phase offset; The optimization function is expressed as: , where T is the fringe period to be solved, is the phase offset to be solved, are the weight coefficients of decoding efficiency and decoding error, respectively, used to balance the weight between decoding accuracy and coding efficiency. The constant term used by the system to generate structured light stripes, is the margin of error, is the exponential of the error growth rate; The optimal solution of fringe period and phase offset is obtained based on the optimization function; Generate structured light stripes based on the optimal solution of stripe period and phase offset; Step 2: A high-speed acquisition camera collects the structured light stripe images of the surface of the object being measured after being reflected by the prisms and mirrors in different directions; Step 3: Reconstruct multiple height maps of the measured object based on the collected structured light fringe images; Step 4: Identify the occluded noise areas and unoccluded true height areas in multiple height maps, and remove the occluded noise areas from the multiple height maps; Step 5: Fuse the unobstructed real height areas in multiple height maps to obtain a complete height map, and perform 3D reconstruction of the measured object based on the obtained height map.

2. The single-camera multi-projection 3D reconstruction method according to claim 1, characterized in that: Reconstructing multiple height maps of the measured object based on the collected structured light fringe images includes the following sub-steps: A phase value is created based on the grayscale values ​​of pixels in the structured light fringe image to solve the equation group; Solve the equations in the system of equations by simultaneous phase value calculation to obtain the phase value of each point of the object being measured; The height of the object surface is calculated based on the phase value of each point on the measured object, and four height maps are generated.

3. The single-camera multi-projection 3D reconstruction method according to claim 2, characterized in that: Identify the occluded noise areas and unobstructed true height areas in multiple height maps, and remove the occluded noise areas from the multiple height maps. This includes the following sub-steps: Extract the gradient features of each pixel in the structured light stripe image; Identify the label of the region to which the pixel belongs based on the gradient characteristics of each pixel; In the height map, remove the pixels corresponding to the occluded noise area.

4. The single-camera multi-projection 3D reconstruction method according to claim 3, characterized in that: The unobstructed true height areas in multiple height maps are fused to obtain a complete height map. This includes the following sub-steps: Transform the four height maps into a common reference coordinate system; Remove the duplicate points from the four heightmaps and concatenate the remaining unique points into one heightmap.

5. A single-camera multi-projection structured light measurement system, wherein the system executes the single-camera multi-projection 3D reconstruction method according to any one of claims 1 to 4, characterized in that: The system comprises: a structured light projector (1), a movable reflective structure group (2) and a structured light collection group (3); the movable reflective structure group (2) comprises a linear motor (21), a reflective mirror group (22) and a reflective prism group (23); the structured light collection group (3) comprises a high-speed collection camera (31) and a remote reflective prism (32); the multiple reflective mirrors in the reflective mirror group (22) and the multiple reflective prisms in the reflective prism group (23) are uniformly distributed along the circumference; the linear motor (21) is connected to the reflective prism group (23) to control the position of the reflective prism group (23); the linear motor (21) moves the reflective prism group (23) to the structured light projector (1), the movable reflective structure group (2) comprises a linear motor (21), a reflective mirror group (22) and a reflective prism group (23); the linear motor (21) moves the reflective prism group (23) to the structured light projector (1), the reflective mirror group (22) and the reflective prism group (23) to the structured light projector (1), the reflective mirror group (22) and the reflective prism group (23) to the structured light projector (1), the reflective prism group (23) moves ... Below the structured light projector (1), the reflecting prism group (23) reflects the structured light emitted by the structured light projector (1) and reflects the structured light onto the corresponding reflecting mirror group (22). The reflecting mirror group (22) projects the structured light onto the object to be measured. After the structured light converges on the surface of the object to be measured, it is projected to form structured light stripes with equal width within the full frame. The far-end reflecting prism (32) is set directly above the object to be measured. The telecentric lens (311) on the high-speed acquisition camera (31) captures the coded stripe image on the surface of the object to be measured through the far-end reflecting prism (32). The coded stripe image is subjected to structured light three-dimensional reconstruction to obtain a three-dimensional point cloud of the object to be measured.

6. The single-camera multi-projection structured light measurement system according to claim 5, characterized in that: The reflector group (22) includes a plurality of reflectors, and the corresponding reflector prism group (23) also includes a plurality of reflector prisms. The number of reflectors and reflector prisms is the same, and both are evenly distributed along the circumference. The structured light reflected from the reflector prism can be transmitted to the corresponding reflector, and all the reflectors can converge the structured light onto the object to be measured.

7. The single-camera multi-projection structured light measurement system according to claim 6, characterized in that: The reflector group (22) includes four reflectors, specifically including a first reflector (221), a second reflector (222), a third reflector (223) and a fourth reflector (224). The four reflectors are evenly distributed at 90 degrees along the circumference and are respectively used to converge structured light reflected in four different directions onto the object to be measured. Correspondingly, the reflective prism group (23) also includes four reflective prisms, specifically including a first reflective prism (231), a second reflective prism (232), a third reflective prism (233) and a fourth reflective prism (234). The four reflective prisms are also evenly distributed at 90 degrees along the circumference and are respectively used to reflect structured light in four different directions.

8. The single-camera multi-projection structured light measurement system according to claim 5, characterized in that: During measurement, the structured light projector (1) is first turned on to project the encoded structured light. When the encoded structured light is projected onto the object, a stripe image is presented. Subsequently, a linear motor (21) is used to move each reflective prism to the bottom of the structured light projector (1) in turn. The high-speed acquisition camera (31) collects structured light stripe images in different directions. The structured light stripe images serve as the data basis for structured light three-dimensional reconstruction.

Images