WO2020237660A1

WO2020237660A1 - Structured light-based three-dimensional reconstruction method, terminal device, and storage medium

Info

Publication number: WO2020237660A1
Application number: PCT/CN2019/089621
Authority: WO
Inventors: 宋展
Original assignee: 中国科学院深圳先进技术研究院
Priority date: 2019-05-31
Filing date: 2019-05-31
Publication date: 2020-12-03

Abstract

The present application is applicable to the technical field of computers, and provides a structured light-based three-dimensional reconstruction method, a terminal device, and a storage medium. The method comprises: obtaining a target coding image formed by superposing a target projection image with surface information of a measured object after the target projection image is projected to the surface of the measured object; detecting each main coding point in the target coding image according to a preset detection algorithm; determining all speckle images around each main coding point, and determining a first coding value of each main coding point according to the types of all the speckle images around each main coding point; determining coding areas of all the speckle images in the target projection image, and respectively performing speckle matching decoding on each coding area to obtain a second coding value of each speckle image; and performing three-dimensional image reconstruction on the measured object according to pre-obtained calibration parameters of a three-dimensional image reconstruction apparatus, the first coding value, and the second coding value to obtain a three-dimensional image of the measured object. The accuracy in three-dimensional reconstruction of an object is improved.

Description

Three-dimensional reconstruction method, terminal equipment and storage medium based on structured light

Technical field

This application belongs to the field of computer technology, and in particular relates to a three-dimensional reconstruction method, device, terminal device and storage medium based on structured light.

Background technique

In the structured light 3D reconstruction technology, for the 3D reconstruction of dynamic targets and scenes, the coding method usually used is the spatial coding method. This is because the spatial coding method only needs to project a coding pattern, and its coding information is generated by spatial coding features or different permutations and combinations. The coding and decoding processes can be completed in a single image, with fast measurement speed and real-time performance. Good advantages. However, the existing spatial coding method, because the density of its coding elements is much lower than that of the time-coded structured light, causes the problem of a small number of point clouds that can be reconstructed and insufficient spatial resolution. At present, in order to improve the spatial resolution of the point cloud that can be reconstructed by spatial coding, a method of increasing the number of coding elements or increasing the coding window is usually adopted.

However, increasing the number of coding elements will increase the difficulty of coding recognition, and increasing the coding window will result in a decrease in the decoding success rate, resulting in low accuracy of the three-dimensional reconstruction of the object. Therefore, how to increase the coding density of spatial coding and the spatial resolution of reconstructed point clouds is one of the problems to be solved in the current spatial coding structured light-based 3D reconstruction technology based on structured light.

technical problem

The embodiments of the present application provide a structured light-based three-dimensional reconstruction method, terminal equipment, and storage medium to solve the problem that in the prior art, the coding density of structured light spatial coding is low and the spatial resolution of reconstructed point clouds is not high, resulting in The problem of inaccurate 3D reconstruction based on structured light.

Technical solutions

The first aspect of the embodiments of the present application provides a three-dimensional reconstruction method based on structured light, including:

Obtain a target coded image formed by projecting the target projection image onto the surface of the measured object and superimposed on the surface information of the measured object; the target coded image includes a first preset number of main code points and a second preset Number of speckle images;

Detect each main coding point in the target coded image according to a preset detection algorithm;

Determining all speckle images around each main coding point, and determining the first code value of each main coding point according to the types of all speckle images around each main coding point;

Determine the coding regions of all speckle images around each main coding point in the target projection image, and perform speckle matching and decoding on each of the coding regions to obtain the second speckle image of each Code value

According to the pre-acquired calibration parameters of the three-dimensional image reconstruction device, the first code value and the second code value, perform three-dimensional image reconstruction on the measured object to obtain a three-dimensional image of the measured object.

A second aspect of the embodiments of the present application provides a terminal device, including a memory, a processor, and computer instructions stored in the memory and running on the processor. When the processor executes the computer instructions, Implement the following steps:

A third aspect of the embodiments of the present application provides a computer-readable storage medium, the computer-readable storage medium stores computer instructions, and when the computer instructions are executed by a processor, the following steps are implemented:

Compared with the prior art, the embodiment of the present application has the following beneficial effects: firstly, each main code point in the target coded image is detected by a preset detection algorithm, and then all speckle images around each main code point are determined. And determine the first code value of each main code point according to all speckle images around each main code point, and then determine the code area of all speckle images around each main code point in the target projection image, and respectively Perform speckle matching and decoding on each code area to obtain the second code value of each speckle image. Finally, according to the calibration parameters of the 3D image reconstruction device, the first code value and the second code value, the measured object Perform 3D image reconstruction. The problems of low coding density of existing spatial coding and low spatial resolution of reconstructed point clouds are solved, and the accuracy of 3D image reconstruction of objects is improved.

Description of the drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the accompanying drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only of the present application. For some embodiments, for those of ordinary skill in the art, other drawings can be obtained from these drawings without creative labor.

FIG. 1 is an implementation flowchart of a structured light-based 3D reconstruction method provided by an embodiment of the present application;

Figure 2 is a flow chart of the specific implementation of S102 in Figure 1;

Figure 3 is a flow chart of the specific implementation of S103 in Figure 1;

Figure 4 is a flowchart of the specific implementation of S1033 in Figure 3;

Figure 5 is a flowchart of the specific implementation of S104 in Figure 1;

6 is a schematic diagram of the structure of the structured light-based 3D reconstruction device of the present invention;

Fig. 7 is a schematic diagram of a terminal device provided by an embodiment of the present invention.

Embodiments of the invention

In the following description, for the purpose of illustration rather than limitation, specific details such as a specific device structure and technology are proposed for a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known devices, devices, circuits, and methods are omitted to avoid unnecessary details from obstructing the description of this application.

In order to illustrate the technical solutions described in the present application, specific embodiments are used for description below.

As shown in FIG. 1, it is an implementation flowchart of a structured light-based 3D reconstruction method provided by an embodiment of the present application. It can be seen from FIG. 1 that, in the structured light-based 3D reconstruction method provided by this application, the execution subject of this embodiment is a terminal device, including:

S101. Obtain a target encoded image that is superimposed on the surface information of the measured object after the target projection image is projected onto the surface of the measured object; the target encoded image includes a first preset number of main encoding points and a second A preset number of speckle images.

In this embodiment, the target projection image is an image formed by embedding different local speckle images of preset types (for example, 8 types) into a preset coding grid according to the coding principle. After the target projection image is projected onto the surface of the measured object by means of laser and DOE diffraction grating, it is superimposed with the surface information such as the color and texture of the surface of the measured object to form a target encoded image. Specifically, the target coded image includes a first preset number of main code points and a second preset number of speckle images. Each of the main coding points contains different local speckle images. Understandably, the target coded image is coded by using the local speckle image around each of the main coding points as coding elements.

S102: Detect each main coding point in the target coded image according to a preset detection algorithm.

Specifically, the preset detection algorithm includes: determining a local symmetry detection operator for detecting each target main coding point in the target coding image according to the geometric characteristics of the coding grid in the target coding image, and The local symmetry detection operator detects each target main coding point in the target coded image. The coding grid in the target coding image has geometric characteristics, and the main coding point is the intersection of grid lines of the coding grid with geometric characteristics.

Further, in order to explain S102 more clearly, as shown in FIG. 2, it is a flowchart of specific implementation of S102 in FIG. 1. It should be noted that, in this embodiment, the coding grid is a grid pattern. Specifically, it can be seen from FIG. 2 that S102 includes:

S1021: Extract all grid lines intersection points of the coding grid.

In this embodiment, the coding grid has significant geometric features, such as a square coding grid (a common grid) or a triangular coding grid. Based on the geometric characteristics of the coding grid, the coding grid is extracted The intersection of the grid lines of the grid, for example, the intersection of the horizontal grid lines and the vertical grid lines of a grid coding grid. It is understandable that, according to the geometric characteristics of the coding grid, each coding grid usually contains a plurality of intersections of grid lines. For example, in this embodiment, taking the grid as the coding grid as an example, each grid includes four intersections of horizontal grid lines and vertical grid lines.

S1022: Generate a local symmetry detection operator for detecting main coding points contained in the target coded image according to the geometric characteristics of the coding grid.

In this embodiment, the coding grid is a grid, and the main coding point is the intersection of a horizontal grid line and a vertical grid line. Then, according to the geometric characteristics of the grid, the code for detecting the target is generated. The local symmetry detection operator of the main coding point contained in the image is:

Among them, I represents the target projection image, (i, j) is the pixel coordinate of the current detection point, w represents the length or width of the grid with (i, j) as the intersection point, and (α, β) represents the intersection point (i, The offset of each pixel in all grids of j) relative to the intersection point (i, j), so α, β ∈ [-w, w]. Let l=w/3,

Is the pixel offset of all areas outside the grid containing the intersection (i, j),

S1023: Detect each main coding point in the target coded image based on the local symmetry detection operator.

Specifically, in this embodiment, the above-mentioned local symmetry detection operator T may be used to detect each main coding point from the target coded image. Furthermore, the incorrectly selected main coding point can be further detected according to the local rotational symmetry characteristics of the main coding point, so as to improve the accuracy of the detection of the main coding point.

S103: Determine all speckle images around each main coding point, and determine a first code value of each main coding point according to the types of all speckle images around each main coding point.

Specifically, according to the geometric characteristics of the coding grid in the target coded image, it can be known that there are multiple adjacent speckle images around each main coding point, for example, in the target coded image with the grid as the coding grid , There are 4 adjacent speckle images around each main code point. According to the imaging principle of speckle images, each speckle image is different. Different speckle images are defined as different speckle images Types of. Therefore, all speckle images around each main coding point can be determined, and the first code value of each main coding point can be determined according to the types of all speckle images around each main coding point.

Further, as shown in FIG. 3, it is a flowchart of specific implementation of S103 in FIG. When the intersection of the grid lines of the coding grid has rotational symmetry and the speckle image is filled in the coding grid, S103 may include:

S1031: Determine the topological coding grid of each main coding point in the target image according to the rotational symmetry of the intersection of the grid lines of the coding grid.

Specifically, in practical applications, the geometric characteristics of the coding grid are determined, so the topological coding grid of each main coding point can be determined according to the rotational symmetry of the intersection of the grid lines of the coding grid, For example, according to the rotational symmetry of the intersection of the horizontal and vertical grid lines of the grid, it can be determined that each main code point contains 4 code grids (grids), that is, the topology of each main code point The coding grid is 4 adjacent grids. It is understandable that, in this embodiment, the topological coding grid of each main coding point in the target image includes 4 adjacent grids. .

S1032: Obtain all speckle images filled in the topological coding grid of each main coding point.

Specifically, according to the generation principle of the speckle image, it can be determined that the speckle image filled in each coding grid is different and unique. In this embodiment, the topological coding grid of each main code includes 4 adjacent grids, and the speckle images in each of the adjacent 4 grids are obtained respectively.

S1033: Perform image type recognition on all speckle images around each main coding point, and determine the first image type of each main coding point based on the types of all speckle images around each main coding point An encoding value.

Specifically, because the types of speckle images around each main coding point are different, and the type of each speckle image is unique, it can be based on the types of all speckle images around each main coding point. In the example, there are 4 types of speckle images, and the first code value of each main code point is determined.

Further, as shown in FIG. 4, it is a flowchart of specific implementation of S1033 in FIG. 3. As shown in Figure 4, S1033 includes:

S1034: According to the pre-trained speckle image type recognition model, perform image type recognition on all speckle images around each main coding point, to obtain the types of all speckle images around each main coding point .

Specifically, the pre-trained speckle image type recognition model is a deep neural network model, and the training samples of the neural network model are a preset number of collected speckle image blocks of different types. For example, in this embodiment In, a preset number of 8 different speckle image blocks are collected as a training sample set, and a speckle image type recognition model used to identify the type of speckle image is trained. The speckle image is subjected to image type recognition, and the types of all speckle images around each main coding point are obtained.

S1035: Determine the first code value of each main code point according to the preset mapping relationship between the first code value of the main code point and all speckle image types around the main code point.

Specifically, the mapping relationship between the first code value of the main code point and all speckle image types around the main code point is preset and stored, for example, assuming that there is a first code of the main code point in the mapping relationship The value is 1234, and the corresponding speckle image types are 1-2-3-4 in the clockwise direction. In this embodiment, it is assumed that the speckle image type recognition model has identified 4 speckle image types around the main code point A. The types of speckle images are (clockwise) 1-2-3-4, and then the encoding value of the main code point A can be determined to be 1234 according to the mapping relationship. By analogy, each main code can be obtained. The first code value of the point.

S104. Determine the coding regions of all speckle images around each main coding point in the target projection image, and perform speckle matching and decoding on each of the coding regions to obtain the speckle image The second code value.

Specifically, by using the coding coordinates of each main coding point in the target coded image and the corresponding first coding value, the projection coordinates of each main coding point in the target projection image can be determined, and the The projection coordinates of each main coding point are used as a reference to determine the coding regions of all speckle images in the target image. Further, by taking each coding region as a unit, performing speckle matching decoding on the speckle image of each coding region, to obtain the second coding value of each speckle image.

Specifically, as shown in FIG. 5, it is a flowchart of specific implementation of S104 in FIG. It can be seen from Figure 5 that S104 includes:

S1041: Determine the position of each main coding point in the target projected image according to the first coding value of each main coding point and the coding coordinates of each main coding point in the target coded image. Projection coordinates.

Specifically, according to the principle of structured light three-dimensional reconstruction, there is a coding mapping relationship between the coding coordinates in the target coding image and the projection coordinates in the target projection image. Therefore, the coding value of the main coding point is determined Then, the coding coordinates corresponding to each main coding point and 4 adjacent speckle image regions around each main coding point can be determined in the target projection image.

S1042: Determine, based on the projection coordinates of each main coding point, the coding area of the speckle image around each main coding point in the target projection image.

According to the principle of structured light, for the speckle in the segmented speckle area, the standard speckle corresponding to the speckle image area around each main code point can be determined in the target projection image based on the projection coordinates of each main code point Image area.

S1043, according to a preset local image matching and decoding algorithm, respectively perform matching and decoding on the coding region of each speckle image in the target projection image to obtain a second coding value of the coding region of each speckle image.

Specifically, the preset local image matching and decoding algorithm is a pixel-level matching algorithm. For example, the preset speckle image decoding algorithm includes, for example, the absolute value of pixel difference (SAD, Sum of Absolute Differences), and the corresponding image sequence Matching algorithms such as the sum of squares of pixel difference (SSD, Sum of Squared Differences), and image correlation (NCC, Normalized Cross Correlation), etc., are not specifically limited here.

S105: Perform 3D image reconstruction on the measured object according to the pre-acquired calibration parameters of the 3D image reconstruction device, the first coded value, and the second coded value, to obtain a three-dimensional image of the measured object.

Specifically, the calibration parameters of the 3D image reconstruction device include the internal parameters of the camera and the projection equipment in the structured light device, and the conversion relationship between the two (also referred to as external parameters). The internal parameters of the camera include focal length and image The ratio of element size, principal point and tilt angle, external parameters include rotation and translation matrix, distortion parameters, etc. It is understandable that in structured light reconstruction, a three-dimensional image (contour) of the object to be measured can be obtained according to the acquired calibration parameters, the first code value, and the second code value.

It can be seen from the above embodiments that the structured light-based 3D reconstruction method proposed in this application first detects each main coding point in the target coded image through a preset detection algorithm, and then determines all speckles around each main coding point Image, and determine the first code value of each main code point according to all speckle images around each main code point, and then determine the code area of all speckle images around each main code point in the target projection image, and Perform speckle matching and decoding on each coded region to obtain the second coded value of each speckle image. Finally, according to the calibration parameters of the three-dimensional image reconstruction device, the first coded value and the second coded value, Three-dimensional image reconstruction of the measured object. The problems of low coding density of existing spatial coding and low spatial resolution of reconstructed point clouds are solved, and the accuracy of 3D image reconstruction of objects is improved.

It should be understood that the size of the sequence number of each step in the foregoing embodiment does not mean the order of execution. The execution sequence of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiment of the present application.

Fig. 6 is a schematic diagram of a terminal device provided by an embodiment of the present application. As shown in FIG. 6, the terminal device 6 of this embodiment includes: a processor 60, a memory 61, and computer instructions 62 stored in the memory 61 and running on the processor 60, such as a three-dimensional structured light-based Rebuild the program. When the processor 60 executes the computer instructions 62, the steps in the above-mentioned structured light-based three-dimensional reconstruction method embodiments, such as steps 101 to 105 shown in FIG. 1, are implemented. Alternatively, when the processor 60 executes the computer instruction 62, the functions of the modules/units in the foregoing device embodiments, for example, the functions of the modules 601 to 605 shown in FIG. 6 are realized.

Exemplarily, the computer instructions 62 may be divided into one or more modules/units, and the one or more modules/units are stored in the memory 61 and executed by the processor 60 to complete This application. The one or more modules/units may be a series of computer instruction instruction segments capable of completing specific functions, and the instruction segments are used to describe the execution process of the computer instructions 62 in the terminal device 6. For example, the computer instruction 62 may be divided into an acquisition module, a detection module, a determination module, a decoding module, and a reconstruction module (a module in a virtual device), and the specific functions of each module are as follows:

The acquisition module is used to acquire a target coded image that is superimposed on the surface information of the tested object after the target projection image is projected onto the surface of the tested object; the target coded image includes a first preset number of main code points And a second preset number of speckle images;

The detection module is used to detect each main coding point in the target coded image according to a preset detection algorithm;

The decoding module is used to determine all speckle images around each main code point, and determine the first code value of each main code point according to the types of all speckle images around each main code point ；

The obtaining module is used to determine the coding regions of all speckle images around each main coding point, and perform speckle matching and decoding on each of the coding regions to obtain the second coding value of each speckle image ；

The reconstruction module is used to reconstruct the three-dimensional image of the object under test according to the calibration parameters of the three-dimensional image reconstruction device, the first code value, and the second code value obtained in advance to obtain the three-dimensional image of the object under test. image.

The terminal device 6 may be a computing device such as a desktop computer, a notebook, a palmtop computer, and a cloud server. The terminal device may include, but is not limited to, a processor 60 and a memory 61. Those skilled in the art can understand that FIG. 6 is only an example of the terminal device 6 and does not constitute a limitation on the terminal device 6. It may include more or less components than shown in the figure, or a combination of certain components, or different components For example, the terminal device 6 may also include input and output devices, network access devices, buses, and the like.

The so-called processor 60 may be a central processing unit (Central Processing Unit, CPU), other general-purpose processors, digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), Ready-made programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor or the like.

The memory 61 may be an internal storage unit of the terminal device 6, such as a hard disk or memory of the terminal device 6. The memory 61 may also be an external storage device of the terminal device 6, for example, a plug-in hard disk equipped on the terminal device 6, a smart memory card (Smart Media Card, SMC), or a Secure Digital (SD) Card, Flash Card, etc. Further, the memory 61 may also include both an internal storage unit of the terminal device 6 and an external storage device. The memory 61 is used to store the computer instructions and other programs and data required by the terminal device. The memory 61 can also be used to temporarily store data that has been output or will be output.

Those skilled in the art can clearly understand that for the convenience and conciseness of description, only the division of the above-mentioned functional units and modules is used as an example. In practical applications, the above-mentioned functions can be allocated to different functional units, Module completion, that is, divide the internal structure of the device into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist alone physically, or two or more units can be integrated into one unit. The above-mentioned integrated units can be hardware-based Formal realization can also be realized in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the protection scope of the present application. For the specific working process of the units and modules in the above-mentioned device, reference may be made to the corresponding process in the foregoing method embodiment, which will not be repeated here.

In the above-mentioned embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail or recorded in an embodiment, reference may be made to related descriptions of other embodiments.

A person of ordinary skill in the art may be aware that the units and algorithm steps of the examples described in combination with the embodiments disclosed herein can be implemented by electronic hardware or a combination of computer software and electronic hardware. Whether these functions are executed by hardware or software depends on the specific application and design constraint conditions of the technical solution. Professionals and technicians can use different methods for each specific application to implement the described functions, but such implementation should not be considered beyond the scope of this application.

In the embodiments provided in this application, it should be understood that the disclosed apparatus/terminal device and method may be implemented in other ways. For example, the device/terminal device embodiments described above are only illustrative. For example, the division of the modules or units is only a logical function division, and there may be other divisions in actual implementation, such as multiple units. Or components can be combined or integrated into another device, or some features can be omitted or not implemented. In addition, the displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or units, and may be in electrical, mechanical or other forms.

The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, the functional units in each embodiment of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units may be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or software functional unit.

If the integrated module/unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on this understanding, this application implements all or part of the processes in the above-mentioned embodiments and methods, and can also be completed by instructing related hardware through computer instructions. The computer instructions can be stored in a computer-readable storage medium. When the instructions are executed by the processor, they can implement the steps of the foregoing method embodiments. . Wherein, the computer instruction includes computer instruction code, and the computer instruction code may be in the form of source code, object code, executable file, or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer instruction code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory) , Random Access Memory (RAM, Random Access Memory), electrical carrier signal, telecommunications signal, and software distribution media, etc. It should be noted that the content contained in the computer-readable medium can be appropriately added or deleted according to the requirements of the legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to the legislation and patent practice, the computer-readable medium Does not include electrical carrier signals and telecommunication signals.

The above-mentioned embodiments are only used to illustrate the technical solutions of the present application, not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that it can still implement the foregoing The technical solutions recorded in the examples are modified, or some of the technical features are equivalently replaced; these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the application, and should be included in Within the scope of protection of this application.

Claims

A three-dimensional reconstruction method based on structured light, which is characterized in that it includes:

Obtain a target coded image formed by projecting the target projection image onto the surface of the measured object and superimposed on the surface information of the measured object; the target coded image includes a first preset number of main code points and a second preset Number of speckle images;

Detect each main coding point in the target coded image according to a preset detection algorithm;

Determining all speckle images around each main coding point, and determining the first code value of each main coding point according to the types of all speckle images around each main coding point;

Determine the coding regions of all speckle images around each main coding point in the target projection image, and perform speckle matching and decoding on each of the coding regions to obtain the second speckle image of each Code value

According to the pre-acquired calibration parameters of the three-dimensional image reconstruction device, the first code value and the second code value, perform three-dimensional image reconstruction on the measured object to obtain a three-dimensional image of the measured object.
The 3D reconstruction method based on structured light according to claim 1, wherein the main coding point is the intersection of grid lines of a coding grid with geometric characteristics, and the detection algorithm detects the Each main code point in the target coded image includes:

Extracting all grid lines intersection points of the coding grid;

Generating a local symmetry detection operator for detecting the main coding points contained in the target coded image according to the geometric characteristics of the coding grid;

Each main coding point in the target coded image is detected based on the local symmetry detection operator.
3. The structured light-based 3D reconstruction method of claim 2, wherein the intersection of grid lines of the coding grid has rotational symmetry, and the speckle image is filled in the coding grid;

The determining all speckle images around each main coding point, and determining the first code value of each main coding point according to the types of all speckle images around each main coding point includes:

Determining the topological coding grid of each main coding point in the target image according to the rotational symmetry of the intersection of the grid lines of the coding grid;

Acquiring all the speckle images filled in the topological coding grid of each main coding point;

Perform image type recognition on all speckle images around each main code point, and determine the first code of each main code point based on the types of all speckle images around each main code point value.
The structured light-based three-dimensional reconstruction method according to claim 3, wherein the image type recognition is performed on all speckle images around each main coding point respectively, and based on each main coding point The types of all surrounding speckle images, and determining the first code value of each main code point, include:

According to the pre-trained speckle image type recognition model, perform image type recognition on all speckle images around each main coding point to obtain the types of all speckle images around each main coding point;

Determine the first code value of each main code point according to the preset mapping relationship between the first code value of the main code point and all speckle image types around the main code point.
The structured light-based three-dimensional reconstruction method according to claim 1, wherein said determining the coding region of all speckle images around each main coding point in the target projection image is for each Performing speckle matching and decoding on the coding region to obtain the second coding value of each speckle image includes:

Determine the projection coordinates of each main coding point in the target projection image according to the first coding value of each main coding point and the coding coordinates of each main coding point in the target coded image ；

Determining the coding region of the speckle image around each main coding point in the target projection image based on the projection coordinates of each main coding point;

According to a preset local image matching and decoding algorithm, the coding region of each speckle image in the target projection image is respectively matched and decoded to obtain the second coding value of the coding region of each speckle image.
The 3D reconstruction method based on structured light according to claim 1, wherein the coding grid is a grid pattern, and the main coding points are horizontal grid lines and vertical grids of the grid pattern. The intersection of the lines.
The structured light-based 3D reconstruction method according to claim 2, wherein the coding grid is a grid pattern, and the main coding points are horizontal grid lines and vertical grids of the grid pattern. The intersection of the lines.
The 3D reconstruction method based on structured light according to claim 3, wherein the coding grid is a grid pattern, and the main coding points are horizontal grid lines and vertical grids of the grid pattern. The intersection of the lines.
The 3D reconstruction method based on structured light according to claim 4, wherein the coding grid is a grid pattern, and the main coding point is a horizontal grid line and a vertical grid of the grid pattern. The intersection of the lines.
The 3D reconstruction method based on structured light according to claim 5, wherein the coding grid is a grid pattern, and the main coding points are horizontal grid lines and vertical grids of the grid pattern. The intersection of the lines.
A terminal device, comprising a memory, a processor, and computer instructions stored in the memory and running on the processor, characterized in that the processor implements the following steps when executing the computer instructions:

Obtain a target coded image formed by projecting the target projection image onto the surface of the measured object and superimposed on the surface information of the measured object; the target coded image includes a first preset number of main code points and a second preset Number of speckle images;

Detect each main coding point in the target coded image according to a preset detection algorithm;

Determining all speckle images around each main coding point, and determining the first code value of each main coding point according to the types of all speckle images around each main coding point;

Determine the coding regions of all speckle images around each main coding point in the target projection image, and perform speckle matching and decoding on each of the coding regions to obtain the second speckle image of each Code value

According to the pre-acquired calibration parameters of the three-dimensional image reconstruction device, the first code value and the second code value, perform three-dimensional image reconstruction on the measured object to obtain a three-dimensional image of the measured object.
The method of 3D reconstruction based on structured light according to claim 11, wherein the main coding point is the intersection of grid lines of a coding grid with geometric characteristics, and the detection algorithm detects the Each main code point in the target coded image includes:

Extracting all grid lines intersection points of the coding grid;

Generating a local symmetry detection operator for detecting the main coding points contained in the target coded image according to the geometric characteristics of the coding grid;

Each main coding point in the target coded image is detected based on the local symmetry detection operator.
The method of 3D reconstruction based on structured light according to claim 12, wherein the intersection of grid lines of the coding grid has rotational symmetry, and the speckle image is filled in the coding grid;

The determining all speckle images around each main coding point, and determining the first code value of each main coding point according to the types of all speckle images around each main coding point includes:

Determining the topological coding grid of each main coding point in the target image according to the rotational symmetry of the intersection of the grid lines of the coding grid;

Acquiring all the speckle images filled in the topological coding grid of each main coding point;

Perform image type recognition on all speckle images around each main code point, and determine the first code of each main code point based on the types of all speckle images around each main code point value.
The structured light-based three-dimensional reconstruction method according to claim 13, wherein the image type recognition is performed on all speckle images around each main coding point respectively, and based on each main coding point The types of all surrounding speckle images, and determining the first code value of each main code point, include:

According to the pre-trained speckle image type recognition model, perform image type recognition on all speckle images around each main coding point to obtain the types of all speckle images around each main coding point;

Determine the first code value of each main code point according to the preset mapping relationship between the first code value of the main code point and all speckle image types around the main code point.
The method of 3D reconstruction based on structured light according to claim 11, wherein said determining the coding area of all speckle images around each main coding point in said target projection image is for each Performing speckle matching and decoding on the coding region to obtain the second coding value of each speckle image includes:

Determine the projection coordinates of each main coding point in the target projection image according to the first coding value of each main coding point and the coding coordinates of each main coding point in the target coded image ；

Determining the coding region of the speckle image around each main coding point in the target projection image based on the projection coordinates of each main coding point;

According to a preset local image matching and decoding algorithm, the coding region of each speckle image in the target projection image is respectively matched and decoded to obtain the second coding value of the coding region of each speckle image.
A computer-readable storage medium storing computer instructions, wherein the computer instructions are executed by a processor to implement the following steps:

Obtain a target coded image formed by projecting the target projection image onto the surface of the measured object and superimposed on the surface information of the measured object; the target coded image includes a first preset number of main code points and a second preset Number of speckle images;

Detect each main coding point in the target coded image according to a preset detection algorithm;

Determining all speckle images around each main coding point, and determining the first code value of each main coding point according to the types of all speckle images around each main coding point;

Determine the coding regions of all speckle images around each main coding point in the target projection image, and perform speckle matching and decoding on each of the coding regions to obtain the second speckle image of each Code value

According to the pre-acquired calibration parameters of the three-dimensional image reconstruction device, the first code value and the second code value, perform three-dimensional image reconstruction on the measured object to obtain a three-dimensional image of the measured object.
The method of 3D reconstruction based on structured light according to claim 16, wherein the main coding point is the intersection of the grid lines of the coding grid with geometric characteristics, and the detection algorithm detects the Each main code point in the target coded image includes:

Extracting all grid lines intersection points of the coding grid;

Generating a local symmetry detection operator for detecting the main coding points contained in the target coded image according to the geometric characteristics of the coding grid;

Each main coding point in the target coded image is detected based on the local symmetry detection operator.
16. The structured light-based 3D reconstruction method of claim 16, wherein the intersection of grid lines of the coding grid has rotational symmetry, and the speckle image is filled in the coding grid;

The determining all speckle images around each main coding point, and determining the first code value of each main coding point according to the types of all speckle images around each main coding point includes:

Determining the topological coding grid of each main coding point in the target image according to the rotational symmetry of the intersection of the grid lines of the coding grid;

Acquiring all the speckle images filled in the topological coding grid of each main coding point;

Perform image type recognition on all speckle images around each main code point, and determine the first code of each main code point based on the types of all speckle images around each main code point value.
The structured light-based three-dimensional reconstruction method according to claim 18, wherein the image type recognition is performed on all speckle images around each main coding point respectively, and based on each main coding point The types of all surrounding speckle images, and determining the first code value of each main code point, include:

According to the pre-trained speckle image type recognition model, perform image type recognition on all speckle images around each main coding point to obtain the types of all speckle images around each main coding point;

Determine the first code value of each main code point according to the preset mapping relationship between the first code value of the main code point and all speckle image types around the main code point.
The method of 3D reconstruction based on structured light according to claim 16, wherein said determining the coding region of all speckle images around each main coding point in the target projection image is performed for each Performing speckle matching and decoding on the coding region to obtain the second coding value of each speckle image includes:

Determine the projection coordinates of each main coding point in the target projection image according to the first coding value of each main coding point and the coding coordinates of each main coding point in the target coded image ；

Determining the coding region of the speckle image around each main coding point in the target projection image based on the projection coordinates of each main coding point;

According to a preset local image matching and decoding algorithm, the coding region of each speckle image in the target projection image is respectively matched and decoded to obtain the second coding value of the coding region of each speckle image.