Three-dimensional reconstruction method and system based on polarization fringe projection structured light fusion

Abstract

The present disclosure relates to a three-dimensional reconstruction method and system based on polarized stripe projection structured light fusion, including: obtaining a light intensity image and a Gray code image, solving the solution phase of the light intensity image, and obtaining the absolute phase of the object; according to the mapping relationship between the absolute phase and the three-dimensional space point cloud, calculating the three-dimensional coordinates of each pixel point of the object to obtain the point cloud, selecting adjacent points in the point cloud for fitting calculation to obtain the normal; calculating a full white image according to the light intensity image and the Gray code image, and splitting the full white image into four images with different polarization angles; solving the azimuth and zenith angle through the four images with different polarization angles, and calculating the target normal according to the azimuth and zenith angle; calculating the angle between the normal and the target normal, and removing inaccurate points in the point cloud by setting the angle threshold to generate a three-dimensional reconstruction result. The present disclosure can generate a three-dimensional reconstruction result with low data redundancy, and can effectively improve the accuracy and robustness of the three-dimensional reconstruction.

Description

Technical Field

[0001] The present disclosure relates to the field of visual sensor measurement technology, and in particular to a three-dimensional reconstruction method and system based on polarization stripe projection structured light fusion. Background Art

[0002] With the rapid development of robot vision, augmented reality and other fields, accurate and efficient 3D reconstruction technologies are becoming increasingly abundant and mature. Among the many 3D reconstruction technologies, polarization 3D imaging and fringe projection structured light are two representative methods. Polarization 3D imaging technology uses the polarization characteristics of light to restore the normal information of the object surface and obtain more accurate surface normals. The advantage of this method is that it can provide rich surface details and is insensitive to changes in ambient lighting. Fringe projection structured light technology projects a set of fringe patterns onto the surface of the object, captures the deformed fringe patterns with a camera, and calculates the 3D coordinates of the object by dephasing and matching. The advantages of this method are fast measurement speed, simple system structure and relatively low cost.

[0003] They each have unique limitations. The fringe projection method is easily disturbed when dealing with highly reflective or texture-poor surfaces, and its measurement accuracy and stability remain a challenge for fast-moving objects. The integration of the polarization normal and the restoration of the three-dimensional coordinates are difficult to achieve directly and require additional constraints. Therefore, it is necessary to study a three-dimensional reconstruction method that combines polarization three-dimensional imaging with fringe projection structured light technology to give full play to their respective advantages. Summary of the invention

[0004] In order to solve the problem that the traditional method has large data deviation in the 3D reconstruction process due to the occlusion and highlight surface of the object itself, and cannot add constraint information, the present disclosure proposes a 3D reconstruction method based on polarized stripe projection structured light fusion to solve the above problem.

[0005] According to one aspect of the present disclosure, a three-dimensional reconstruction method based on polarization fringe projection structured light fusion is provided, comprising:

[0006] S10, acquiring a light intensity image and a Gray code image, wherein the light intensity image and the Gray code image are acquired by projecting phase shift fringes and Gray code fringes onto an object respectively and collecting them through a camera;

[0007] S20, solving the phase solution of the light intensity image to obtain the absolute phase of the object;

[0008] S30, calculating the three-dimensional coordinates of each pixel of the object to obtain a point cloud according to the mapping relationship between the absolute phase and the three-dimensional space point cloud, and selecting adjacent points in the point cloud for fitting calculation to obtain a normal line;

[0009] S40, calculating a full white image according to the light intensity image and the Gray code image, and splitting the full white image into four images with different polarization angles;

[0010] S50, solving the azimuth and zenith angles through four images with different polarization angles, and calculating the target normal according to the azimuth and zenith angles;

[0011] S60, calculating the angle between the normal line and the target normal line, removing inaccurate points in the point cloud by setting an angle threshold, and generating a three-dimensional reconstruction result.

[0012] Preferably, the phase solution of the fringe image is obtained, which is expressed as:

[0013] ,

[0014] In the formula, is the phase solution of the fringe image, I cn ( x,y ) is the light intensity of the reflected pattern projected onto the object by the collected phase-shifted fringes, N is the number of phase shift steps, n is not less than 1 and not greater than N A positive integer.

[0015] Preferably, calculating the three-dimensional coordinates of each pixel point of the object to obtain a point cloud includes: mapping the image coordinate system to the world coordinate system to generate a point cloud, which is expressed as:

[0016] ,

[0017] In the formula, ( xc , yc ) is the image pixel coordinate, xp is the horizontal coordinate of the projector, ( x , yw , z ) is the world coordinate of the object, and A is the mapping matrix.

[0018] Preferably, adjacent points in the point cloud are selected for fitting calculation to obtain the normal line, including: for each point in the point cloud, a certain number of adjacent points around it are selected, these points are fitted into a plane, and the normal vector of the plane is further obtained, and the normal vector is the normal line of the point.

[0019] Preferably, the full white image is calculated based on the light intensity image and the Gray code image. It is expressed as:

[0020] ,

[0021] In the formula, I cn ( x,y ) is the light intensity of the reflected pattern projected onto the object by the collected phase-shifted fringes, Nis the number of phase shift steps, n is not less than 1 and not greater than N A positive integer.

[0022] Preferably, solving the azimuth angle and the zenith angle through four images with different polarization angles includes:

[0023] ,

[0024] or ,

[0025] In the formula, is the refractive index, is the degree of linear polarization, is the Stokes matrix, is the azimuth, is the zenith angle.

[0026] Preferably, calculating the angle between the normal line and the target normal line includes:

[0027] The angle between the normal line and the target normal line is obtained by calculating the dot product of the normalized vectors corresponding to the normal line and the target normal line, calculating the cosine value of the angle between the normal line and the target normal line according to the dot product, and solving the inverse cosine function from the cosine value.

[0028] According to one aspect of the present disclosure, a three-dimensional reconstruction system based on polarization fringe projection structured light fusion is provided, comprising:

[0029] A light intensity image and Gray code image acquisition module is used to acquire a light intensity image and a Gray code image. The light intensity image and the Gray code image are acquired by projecting phase shift fringes and Gray code fringes onto an object and then capturing the image with a camera.

[0030] An absolute phase calculation module is used to solve the phase of the light intensity image to obtain the absolute phase of the object;

[0031] The normal calculation module calculates the three-dimensional coordinates of each pixel of the object to obtain the point cloud based on the mapping relationship between the absolute phase and the three-dimensional space point cloud, and selects the adjacent points in the point cloud for fitting calculation to obtain the normal line;

[0032] A full white image calculation module is used to calculate a full white image according to the light intensity image and the Gray code image, and split the full white image into four images with different polarization angles;

[0033] The target normal calculation module solves the azimuth and zenith angles through four images with different polarization angles, and calculates the target normal according to the azimuth and zenith angles;

[0034] The three-dimensional reconstruction module calculates the angle between the normal line and the target normal line, removes inaccurate points in the point cloud by setting an angle threshold, and generates a three-dimensional reconstruction result.

[0035] According to one aspect of the present disclosure, an electronic device is provided, comprising: a processor; and a memory for storing instructions executable by the processor; wherein the processor is configured to: execute the above-mentioned three-dimensional reconstruction method based on polarization stripe projection structured light fusion.

[0036] According to one aspect of the present disclosure, a computer-readable storage medium is provided, on which computer program instructions are stored. When the computer program instructions are executed by a processor, the three-dimensional reconstruction method based on polarization stripe projection structured light fusion is implemented.

[0037] Compared with the prior art, the beneficial effects of the present disclosure are:

[0038] 1) The present invention uses a new point cloud reconstruction method in data fusion, which improves the accuracy of the multi-step phase shift method, significantly reduces the influence of surface noise on the measured object, and improves the accuracy of the data.

[0039] 2) The present invention can simultaneously obtain high-precision normal information and precise three-dimensional coordinates of the object surface, thereby achieving more accurate and more robust three-dimensional reconstruction.

[0040] 3) The present disclosure can expand the application scope of 3D reconstruction, making it applicable to a wider range of industrial and scientific research fields.

[0041] It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.

[0042] Further features and aspects of the present disclosure will become apparent from the following detailed description of exemplary embodiments with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The drawings herein are incorporated into the specification and constitute a part of the specification. These drawings illustrate embodiments consistent with the present disclosure and are used to illustrate the technical solutions of the present disclosure together with the specification.

[0044] Figure 1 A flow chart of a three-dimensional reconstruction method based on polarized fringe projection structured light fusion is shown;

[0045] Figure 2 A flow chart of the fusion of polarized three-dimensional and fringe projection structured light in an embodiment of the present disclosure is shown;

[0046] Figure 3 A schematic diagram of a three-dimensional reconstruction system model in an embodiment of the present disclosure is shown;

[0047] Figure 4 A flow chart of step S60 in an embodiment of the present disclosure is shown;

[0048] Figure 5The structural block diagram of the three-dimensional reconstruction system based on polarization stripe projection structured light fusion is shown. DETAILED DESCRIPTION

[0049] Various exemplary embodiments, features and aspects of the present disclosure will be described in detail below with reference to the accompanying drawings. The same reference numerals in the accompanying drawings represent elements with the same or similar functions. Although various aspects of the embodiments are shown in the accompanying drawings, the drawings are not necessarily drawn to scale unless otherwise specified.

[0050] The word “exemplary” is used exclusively herein to mean “serving as an example, example, or illustration.” Any embodiment described herein as “exemplary” is not necessarily to be construed as preferred or advantageous over other embodiments.

[0051] The term "and / or" herein is only a description of the association relationship of the associated objects, indicating that there may be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the term "at least one" herein represents any combination of at least two of any one or more of a plurality of. For example, including at least one of A, B, and C can represent including any one or more elements selected from the set consisting of A, B, and C.

[0052] In addition, in order to better illustrate the present disclosure, numerous specific details are given in the following specific embodiments. It should be understood by those skilled in the art that the present disclosure can also be implemented without certain specific details. In some examples, methods, means, components and circuits well known to those skilled in the art are not described in detail in order to highlight the subject matter of the present disclosure.

[0053] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0054] Example 1

[0055] This embodiment proposes a three-dimensional reconstruction method based on polarization stripe projection structured light fusion. Figure 1 A flow chart of a three-dimensional reconstruction method based on polarized stripe projection structured light fusion is shown. The method comprises:

[0056] S10, acquiring a light intensity image and a Gray code image, wherein the light intensity image and the Gray code image are acquired by projecting phase shift fringes and Gray code fringes onto the object 400 and capturing them with a camera;

[0057] S20, solving the phase solution of the light intensity image to obtain the absolute phase of the object 400;

[0058] S30, according to the mapping relationship between the absolute phase and the three-dimensional space point cloud, the three-dimensional coordinates of each pixel point of the object 400 are calculated to obtain the point cloud, and the adjacent points in the point cloud are selected for fitting calculation to obtain the normal line;

[0059] S40, calculating a full white image according to the light intensity image and the Gray code image, and splitting the full white image into four images with different polarization angles;

[0060] S50, solving the azimuth and zenith angles through four images with different polarization angles, and calculating the target normal according to the azimuth and zenith angles;

[0061] S60, calculating the angle between the normal line and the target normal line, removing inaccurate points in the point cloud by setting an angle threshold, and generating a three-dimensional reconstruction result.

[0062] The disclosed embodiment provides a 3D reconstruction method based on the fusion of polarized fringe projection structured light, based on the process of fusion of polarized 3D and fringe projection structured light, such as Figure 2 As shown, the following steps are included:

[0063] S10, acquiring a light intensity image and a Gray code image, wherein the light intensity image and the Gray code image are acquired by respectively projecting phase shift fringes and Gray code fringes onto the object 400 and capturing them with a camera.

[0064] In this embodiment, a set of preset phase-shifted stripes are projected by a projector, and then a light intensity image modulated by the object 400 is collected. The phase information is obtained by phase-decoding the multiple collected images, and then the height information is solved by the phase information. Generally, the projected stripes are sinusoidal or Gaussian periodic stripes containing a certain frequency. When the stripes are phase-decoded, the order of the stripes will be blurred and deviated, resulting in height distortion. In addition, a set of Gray code stripes with a code value equal to the number of periodic stripes is projected to assist in solving the stripe order. The number of periods and the number of Gray codes can be designed according to the field of view of the device and present a periodic characteristic.

[0065] S20 , solving the phase solution of the light intensity image to obtain the absolute phase of the object 400 .

[0066] In this embodiment, a calibrated camera is used to collect phase-shifted fringes and Gray code fringes projected onto the object 400 by a projector to obtain a light intensity image and a Gray code image. The light intensity image of the projected sinusoidal fringes is generated using a fringe projection structured light method, which is expressed as:

[0067] ,

[0068] In the formula, is the projected light intensity image, is the average light intensity of the fringes, To modulate the light intensity, is the measured phase, N is the phase shift step number, n The value is 1, 2, ..., N .

[0069] Furthermore, by solving the phase solution of the intensity image, the absolute phase of the reconstructed object 400 is obtained. The stripes divide the projection space area into different phases, and each collected reflected light image will record the phase of the space area. The phase solution is obtained based on the intensity image of the collected phase-shifted stripe reflected image, which is expressed as:

[0070] ,

[0071] In the formula, is the phase solution of the fringe image, I cn ( x,y ) is the light intensity of the reflected pattern projected onto the object 400 by the collected phase-shifted fringes, N is the phase shift step number, n The value is 1, 2, ..., N ;

[0072] Among them, the phase solution formula of four-step phase shift is:

[0073] ,

[0074] In the formula, is the first phase shift pattern collected, is the second phase shift pattern collected, is the third phase shift pattern collected, This is the fourth phase shift pattern collected.

[0075] S30, according to the mapping relationship between the absolute phase and the three-dimensional space point cloud, the three-dimensional coordinates of each pixel point of the object 400 are calculated to obtain the point cloud, and the adjacent points in the point cloud are selected for fitting calculation to obtain the normal line.

[0076] In this embodiment, the underwater 3D reconstruction technology based on the double N-step orthogonal polarization state phase shift strategy is described by taking a structured light vision sensor as an example. The 3D reconstruction system model includes: a host computer 100, a camera 200, a projector 300 and an object 400, such as Figure 3 As shown, the image pixel coordinates are ( xc , yc ), the horizontal coordinate of the projector is xp , the world coordinates of object 400 are ( x , yw , z). If there is no special requirement, the upper left corner of the calibration plate is the origin of the world coordinate system, and the plane of the calibration plate is the origin of the world coordinate system. xy The point cloud is obtained by calculating the three-dimensional coordinates of each pixel point of the object 400, including: mapping the image coordinate system to the world coordinate system to generate a point cloud, which is expressed as:

[0077] ,

[0078] In the formula, ( xc , yc ) is the image pixel coordinate, xp is the horizontal coordinate of the projector, ( x , yw , z ) is the world coordinate of the object, and A is the mapping matrix.

[0079] Furthermore, assuming that each point in the point cloud can be fitted by a local plane, adjacent points in the point cloud are selected for fitting calculation to obtain the normal line, including: for each point in the point cloud, a certain number of neighboring points (such as k nearest neighbors) around it are selected, and these points are fitted into a plane, and further the normal vector of the plane is obtained, which is the normal line of the point.

[0080] S40, calculating a full white image according to the light intensity image and the Gray code image, and splitting the full white image into four images with different polarization angles.

[0081] In this embodiment, a full white image is calculated based on the light intensity image and the Gray code image, which is expressed as:

[0082] ,

[0083] In the formula, I cn ( x,y ) is the light intensity of the reflected pattern projected onto the object by the collected phase-shifted fringes, N is the phase shift step number, n The value is 1, 2, ..., N .

[0084] S50, solving the azimuth angle and the zenith angle through four images with different polarization angles, and calculating the target normal according to the azimuth angle and the zenith angle.

[0085] In this embodiment, the azimuth angle and the zenith angle are solved by using four images with different polarization angles, including:

[0086] ,

[0087] or ,

[0088] In the formula, is the refractive index, is the degree of linear polarization, is the Stokes matrix, is the azimuth, is the zenith angle.

[0089] The polarization characteristics of the light measured by the camera are represented by the Stokes matrix, which is expressed as:

[0090] ,

[0091] The camera collects light intensity at different polarization angles. I (0°), I (45°), I (90°), I (135°) are polarization images at four different angles;

[0092] The degree of linear polarization can also be derived from the Stokes vector:

[0093] ,

[0094] In the formula, is the degree of linear polarization, is the first element in the Stokes matrix s0, is the second element in the Stokes matrix s0, is the third element in the Stokes matrix s0.

[0095] Furthermore, the target normal is calculated based on the azimuth and zenith angles. Each pixel is an image of a microelement on the target surface. When the microelement is divided into very small pieces, each microelement can be simplified into a tiny plane. The assumed normal is in the incident plane formed by the incident light and the reflected light (the light entering the camera), and the angle relationship follows the basic reflection theorem. The light entering the camera can be regarded as a specular reflection from the measured object and a diffuse reflection from the surface of the measured object. Part of the diffuse reflection is polarized light, and part is non-polarized light. This just reflects the advantage of polarized 3D imaging, which can be measured without being greatly affected by ambient light.

[0096] S60, calculating the angle between the normal line and the target normal line, removing inaccurate points in the point cloud by setting an angle threshold, and generating a three-dimensional reconstruction result.

[0097] In this embodiment, calculating the angle between the normal line and the target normal line includes:

[0098] By calculating the dot product of the normalized vector corresponding to the normal and the target normal, B is expressed as:

[0099] B= ,

[0100] In the formula, , are the direction vectors corresponding to the normal and target normal respectively, , All have been normalized and the vector length is 1.

[0101] The cosine of the angle between the normal and the target normal is calculated based on the dot product of the direction vector, expressed as:

[0102] cos(θ)=B,

[0103] The arccosine function is solved by the cosine value to calculate the angle between the two direction vectors, that is, the angle between the normal line and the target normal line is obtained, which is expressed as:

[0104] θ =arccos(cos( θ )),

[0105] Set up a θ A threshold is set to remove points with significant deviations in the normal direction, retaining only points with high reliability in the point cloud, and thereby generating a high-precision 3D model.

[0106] The specific flow chart based on the above steps is as follows Figure 4 As shown, the angle between the normal and the target normal is calculated by taking the point-by-point normal and the target normal calculated by the azimuth and zenith angle as input. θ ,set up θ The threshold is 20°. If the angle between the normal and the target normal is θ If the angle is less than 20°, the point cloud and normal are removed; otherwise, the point cloud and normal are retained and a 3D point cloud result is generated based on them.

[0107] This embodiment combines polarization 3D imaging with fringe projection structured light technology by performing normal and point cloud fusion calculation on the reconstructed object 400, and simultaneously obtains high-precision normal information and accurate 3D coordinates of the surface of the object 400, thereby generating a high-precision 3D reconstruction result, and effectively improving the accuracy and robustness of the 3D reconstruction.

[0108] Example 2

[0109] As another aspect of the embodiment of the present disclosure, a three-dimensional reconstruction system 10 based on polarization stripe projection structured light fusion is also provided. Figure 5 As shown, including:

[0110] A light intensity image and Gray code image acquisition module 1 is used to acquire a light intensity image and a Gray code image. The light intensity image and the Gray code image are acquired by projecting phase shift fringes and Gray code fringes onto an object and then capturing the image with a camera.

[0111] An absolute phase calculation module 2 is used to solve the phase of the light intensity image to obtain the absolute phase of the object;

[0112] Normal calculation module 3, based on the mapping relationship between the absolute phase and the three-dimensional space point cloud, calculates the three-dimensional coordinates of each pixel of the object to obtain the point cloud, and selects the adjacent points in the point cloud for fitting calculation to obtain the normal line;

[0113] A full white image calculation module 4 is configured to calculate a full white image according to the light intensity image and the Gray code image, and split the full white image into four images with different polarization angles;

[0114] The target normal calculation module 5 solves the azimuth and zenith angles through four images with different polarization angles, and calculates the target normal according to the azimuth and zenith angles;

[0115] The three-dimensional reconstruction module 6 calculates the angle between the normal line and the target normal line, removes inaccurate points in the point cloud by setting an angle threshold, and generates a three-dimensional reconstruction result.

[0116] In the absence of any contradiction, the above modules in the system of the embodiment of the present disclosure can implement any implementation of the above method.

[0117] Based on the description of the above embodiments, it can be seen that the embodiments of the present disclosure can achieve the following technical effects:

[0118] 1) The present invention uses a new point cloud reconstruction method in data fusion, which improves the accuracy of the multi-step phase shift method, significantly reduces the influence of surface noise on the measured object, and improves the accuracy of the data.

[0119] 2) The present invention can simultaneously obtain high-precision normal information and precise three-dimensional coordinates of the object surface, thereby achieving more accurate and more robust three-dimensional reconstruction.

[0120] 3) The present disclosure can expand the application scope of 3D reconstruction, making it applicable to a wider range of industrial and scientific research fields.

[0121] The disclosed embodiment also provides an electronic device, comprising: a processor; a memory for storing instructions executable by the processor; wherein the processor is configured to implement the above-mentioned three-dimensional reconstruction method based on polarized stripe projection structured light fusion. The electronic device can be provided as a terminal, a server, or other forms of equipment.

[0122] The present disclosure also provides a computer-readable storage medium on which computer program instructions are stored, and when the computer program instructions are executed by a processor, the three-dimensional reconstruction method based on polarization stripe projection structured light fusion is implemented. The computer-readable storage medium may be a non-volatile computer-readable storage medium.

[0123] Those skilled in the art will understand that in the above-mentioned three-dimensional reconstruction method and system based on polarization stripe projection structured light fusion in the specific implementation mode, the writing order of each step does not mean a strict execution order and does not constitute any limitation on the implementation process. The specific execution order of each step should be determined by its function and possible internal logic.

[0124] The flow chart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to multiple embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a part of a module, program segment or instruction, and a part of the module, program segment or instruction includes one or more executable instructions for realizing the specified logical function. In some alternative implementations, the function marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous square boxes can actually be executed substantially in parallel, and they can sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs the specified function or action, or can be implemented with a combination of special hardware and computer instructions.

[0125] The embodiments of the present disclosure have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or technical improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A three-dimensional reconstruction method based on polarized fringe projection structured light fusion, characterized in that: The steps include: S10, acquiring a light intensity image and a Gray code image, wherein the light intensity image and the Gray code image are acquired by projecting phase shift fringes and Gray code fringes onto an object respectively and collecting them through a camera; S20, solving the phase solution of the light intensity image to obtain the absolute phase of the object; S30, calculating the three-dimensional coordinates of each pixel of the object to obtain a point cloud according to the mapping relationship between the absolute phase and the three-dimensional space point cloud, and selecting adjacent points in the point cloud for fitting calculation to obtain a normal line; S40, calculating a full white image according to the light intensity image and the Gray code image, and splitting the full white image into four images with different polarization angles; S50, solving the azimuth and zenith angles through four images with different polarization angles, and calculating the target normal according to the azimuth and zenith angles; S60, calculating the angle between the normal line and the target normal line, removing inaccurate points in the point cloud by setting an angle threshold, and generating a three-dimensional reconstruction result.

2. The method according to claim 1, characterized in that Solve the phase of the intensity image, expressed as: ， In the formula, is the phase solution of the fringe image, I cn ( x,y ) is the light intensity of the reflected pattern projected onto the object by the collected phase-shifted fringes, N is the number of phase shift steps, n is not less than 1 and not greater than N A positive integer.

3. The method according to claim 1, characterized in that Calculate the three-dimensional coordinates of each pixel of the object to obtain a point cloud, including: mapping the image coordinate system to the world coordinate system to generate a point cloud, expressed as: ， In the formula, ( xc , yc ) is the image pixel coordinate, xp is the horizontal coordinate of the projector, ( x , yw , z ) is the world coordinate of the object, and A is the mapping matrix.

4. The method according to claim 1, characterized in that Selecting adjacent points in the point cloud for fitting calculation to obtain the normal line includes: for each point in the point cloud, selecting a certain number of adjacent points around it, fitting these points into a plane, and further obtaining the normal vector of the plane, which is the normal line of the point.

5. The method according to claim 3, characterized in that: The full white image is calculated based on the light intensity image and the Gray code image It is expressed as: ， In the formula, I cn ( x,y ) is the light intensity of the reflected pattern projected onto the object by the collected phase-shifted fringes, N is the number of phase shift steps, n is not less than 1 and not greater than N A positive integer.

6. The method according to claim 4, characterized in that The azimuth and zenith angles are solved by four images with different polarization angles, including: ， or , In the formula, is the refractive index, is the degree of linear polarization, is the Stokes matrix, is the azimuth, is the zenith angle.

7. The method according to claim 1, characterized in that Calculate the angle between the normal and the target normal, including: The angle between the normal line and the target normal line is obtained by calculating the dot product of the normalized vectors corresponding to the normal line and the target normal line, calculating the cosine value of the angle between the normal line and the target normal line according to the dot product, and solving the inverse cosine function from the cosine value.

8. A three-dimensional reconstruction system based on polarized fringe projection structured light fusion, characterized in that: include: A light intensity image and Gray code image acquisition module is used to acquire a light intensity image and a Gray code image. The light intensity image and the Gray code image are acquired by projecting phase shift fringes and Gray code fringes onto an object and then capturing the image with a camera. An absolute phase calculation module is used to solve the phase of the light intensity image to obtain the absolute phase of the object; The normal calculation module calculates the three-dimensional coordinates of each pixel of the object to obtain the point cloud based on the mapping relationship between the absolute phase and the three-dimensional space point cloud, and selects the adjacent points in the point cloud for fitting calculation to obtain the normal line; A full white image calculation module is used to calculate a full white image according to the light intensity image and the Gray code image, and split the full white image into four images with different polarization angles; The target normal calculation module solves the azimuth and zenith angles through four images with different polarization angles, and calculates the target normal according to the azimuth and zenith angles; The three-dimensional reconstruction module calculates the angle between the normal line and the target normal line, removes inaccurate points in the point cloud by setting an angle threshold, and generates a three-dimensional reconstruction result.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the three-dimensional reconstruction method based on polarization stripe projection structured light fusion according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the three-dimensional reconstruction method based on polarization stripe projection structured light fusion as described in any one of claims 1 to 7 is implemented.

Images