A three-dimensional structure construction method and system based on two-dimensional information and a storage medium
By analyzing mirror symmetry and wavefront reconstruction techniques, calculating wavefront differences and performing error compensation, the problem of 3D structure errors caused by object occlusion in 2D images was solved, and more accurate 3D structure construction was achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- HARBIN INST OF TECH
- Filing Date
- 2026-05-07
- Publication Date
- 2026-06-12
AI Technical Summary
In existing technologies, when objects in two-dimensional images self-occlude, the mirror symmetry plane exhibits fundamental errors, leading to significant ambiguity in the construction of three-dimensional structures and affecting the practicality and accuracy of the models.
By analyzing the mirror symmetry of the target object, predicting the plane of symmetry, calculating the wavefront difference using wavefront reconstruction technology, solving the polynomial coefficient equation for error compensation, and generating a more accurate three-dimensional structure.
It improves the accuracy of 3D reconstruction and the practicality of the model, reduces the uncertainty caused by occlusion or viewpoint limitations, and enhances the reliability of the model.
Smart Images
Figure CN122199874A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of three-dimensional reconstruction technology, and more specifically, to a method, system, and storage medium for constructing three-dimensional structures based on two-dimensional information. Background Technology
[0002] In the field of 3D reconstruction, recovering the 3D structure of an object from its 2D information plays a vital role in operations such as object recognition, robot grasping, and object pose estimation. Therefore, the mirror symmetry method is usually used for absolute measurement to infer the spatial structure or geometry of the corresponding scene or object. Specifically, the mirror symmetry method is used to generate the mirror symmetry plane of the 2D object, and then the mirror symmetry plane is used as a constraint to guide the object to generate a 3D structure in a self-symmetrical manner.
[0003] In existing technologies, if objects in a single 2D image are self-occluded, it will cause a fundamental error in the mirror symmetry plane, making the inferred 3D structure ambiguous. Consequently, it cannot accurately represent the shape or structure of the actual object, leading to an increase in the difference between the model and the real-world object, affecting the model's practicality and accuracy, and limiting its application in various fields. Summary of the Invention
[0004] The problem addressed by this invention is how to improve the accuracy of constructing three-dimensional structures based on two-dimensional images.
[0005] To address the aforementioned problems, this invention provides a method, system, and storage medium for constructing three-dimensional structures based on two-dimensional information.
[0006] In a first aspect, the present invention provides a method for constructing a three-dimensional structure based on two-dimensional information, comprising: Based on the target object in the two-dimensional image, obtain the mirror symmetry plane of the target object; Wavefront reconstruction is performed based on the mirror symmetry plane of the target object to obtain the wavefront difference of the mirror symmetry plane; Solve the polynomial coefficient equation based on the wavefront difference to obtain the polynomial coefficients; The principle error of the mirror symmetry plane is compensated according to the polynomial coefficients, and the three-dimensional structure of the target object is generated according to the mirror symmetry plane after error compensation.
[0007] Optionally, obtaining the mirror symmetry plane of the target object based on the target object in the two-dimensional image includes: Feature extraction is performed on the target object in the two-dimensional image to obtain the geometric features of the target object; Based on the geometric features of the target object, the mirror symmetry plane of the target object is obtained.
[0008] Optionally, obtaining the mirror symmetry plane of the target object based on its geometric features includes: Based on the geometric features of the target object, determine the symmetry standard of the target object; By using a probabilistic prediction model, the geometric features are probabilistically predicted according to the symmetry standard, thereby generating the mirror symmetry plane of the geometric features of the target object under the symmetry standard.
[0009] Optionally, determining the symmetry standard of the target object based on its geometric features includes: Based on the geometric features of the target object, the region and shape of the target object in the two-dimensional image are obtained; Based on the region and shape of the target object in the two-dimensional image, select the axis of symmetry of the target object; Based on the axis of symmetry and the geometric features of the target object, determine the direction of symmetry of the target object; The symmetry standard is determined based on the axis of symmetry and the direction of symmetry.
[0010] Optionally, the step of reconstructing the wavefront based on the mirror symmetry plane of the target object to obtain the wavefront difference of the mirror symmetry plane includes: Obtain the wavefront data of the mirror-symmetric plane; By rotating the mirror-symmetric plane by a preset angle, the wavefront data of the rotated mirror-symmetric plane is obtained; The wavefront difference value of the mirror-symmetric plane is determined based on the phase difference between the wavefront data before rotation and the wavefront data after rotation.
[0011] Optionally, the step of solving the polynomial coefficient equation based on the wavefront difference to obtain the polynomial coefficients includes: Based on the wavefront difference, the Zernike polynomial is obtained; The polynomial coefficient equation is established based on the Zernike polynomial, wherein the unknowns of the polynomial coefficient equation are the polynomial coefficients in the Zernike polynomial. The polynomial coefficient equation is solved numerically to obtain the polynomial coefficients of each order and weight in the Zernike polynomial.
[0012] Optionally, the step of compensating for the fundamental error of the mirror-symmetric plane based on the polynomial coefficients includes: The compensation order and compensation weight are determined based on the preset accuracy. The polynomial coefficients in the fundamental error are determined based on the compensation order and the compensation weight. The wavefront data before rotation is compensated based on the polynomial coefficients to obtain the final wavefront data of the mirror-symmetric plane. Based on the final wavefront data, the final mirror-symmetric plane is obtained.
[0013] Optionally, generating the three-dimensional structure of the target object based on the error-compensated mirror symmetry plane includes: Using the final mirror symmetry plane as a constraint, and based on the two-dimensional image of the target object, symmetry processing is performed according to the symmetry standard to generate the three-dimensional structure of the target object.
[0014] Secondly, the present invention provides a three-dimensional structure construction system based on two-dimensional information, comprising: A mirror symmetry unit is used to obtain the mirror symmetry plane of the target object based on the target object in the two-dimensional image; A wavefront reconstruction unit is used to reconstruct the wavefront based on the mirror symmetry plane of the target object to obtain the wavefront difference of the mirror symmetry plane. The equation calculation unit is used to solve the polynomial coefficient equation based on the wavefront difference to obtain the polynomial coefficients. A three-dimensional construction unit is used to compensate for the fundamental error of the mirror symmetry plane based on the polynomial coefficients, and to generate the three-dimensional structure of the target object based on the error-compensated mirror symmetry plane.
[0015] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, it implements the above-described method for constructing a three-dimensional structure based on two-dimensional information.
[0016] This invention discloses a method, system, and storage medium for constructing three-dimensional structures based on two-dimensional information. By analyzing the mirror symmetry of a target object, the system predicts the object's plane of symmetry, reducing uncertainties caused by occlusion or viewpoint limitations. Next, wavefront reconstruction technology is used to simulate the phase changes of the wavefront on the plane of symmetry, thereby calculating the wavefront difference. The wavefront difference is used to understand wavefront distortion, which is often the source of error in the two-dimensional to three-dimensional conversion. Furthermore, by solving polynomial coefficient equations, distortion is captured and quantified. Finally, these polynomial coefficients are applied to compensate for fundamental errors, correcting the wavefront data and generating a more accurate three-dimensional structure on the mirror symmetry plane. This not only improves the accuracy of three-dimensional reconstruction but also enhances the model's practicality and reliability by reducing errors and uncertainties. Attached Figure Description
[0017] Figure 1 This is a flowchart of a three-dimensional structure construction method based on two-dimensional information according to an embodiment of the present invention; Figure 2 This is a two-dimensional image of the target object before mirror symmetry in an embodiment of the present invention; Figure 3 The mirror symmetry plane of the target object after mirror symmetry in the embodiments of the present invention; Figure 4 This is a schematic diagram of the three-dimensional structure construction system based on two-dimensional information according to an embodiment of the present invention. Detailed Implementation
[0018] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.
[0019] Combination Figure 1 As shown, the present invention provides a method for constructing a three-dimensional structure based on two-dimensional information, comprising: Based on the target object in the two-dimensional image, obtain the mirror symmetry plane of the target object.
[0020] Specifically, the three-dimensional symmetry of a target object is identified and inferred from a two-dimensional image using a vision system or algorithm. In a preferred embodiment of the invention, the inherent symmetry structures of natural and man-made objects can be utilized to determine the mirror symmetry plane for the reconstructed object through probabilistic prediction. Figure 2 The two-dimensional image of the target object before mirror symmetry is shown. Figure 2 After mirroring the objects in the image, we get... Figure 3 The mirror symmetric plane in, and according to Figure 2 and Figure 3 The comparison shows that the density of the mesh remains unchanged after mirror symmetry, and the material is not affected. Thus, the material structure consistency of the mirror symmetry plane is utilized, saving the time of remodeling.
[0021] Wavefront reconstruction is performed based on the mirror symmetry plane of the target object to obtain the wavefront difference of the mirror symmetry plane.
[0022] Specifically, wavefront reconstruction is performed based on the mirror symmetry plane of the target object to obtain the wavefront difference of the mirror symmetry plane. Wavefront reconstruction usually refers to simulating or predicting the changes of the wavefront based on known wavefront data in an optical system. In this invention, wavefront reconstruction is used to describe the process of modeling the device structure in TCAD process simulation, that is, based on a known symmetry plane, the state of the wavefront on the plane is reconstructed through simulation or experiment, and the difference between it and the ideal state is calculated.
[0023] The polynomial coefficients are obtained by solving the polynomial coefficient equation based on the wavefront difference.
[0024] Specifically, in a preferred embodiment of the present invention, the fundamental error in the absolute measurement of the mirror symmetry method can be compensated by solving the coefficients of the Zernike polynomial, thereby accurately quantifying the wavefront difference and providing the necessary parameters for subsequent error compensation.
[0025] The principle error of the mirror symmetry plane is compensated according to the polynomial coefficients, and the three-dimensional structure of the target object is generated according to the mirror symmetry plane after error compensation.
[0026] Specifically, error compensation is achieved by adding rotations at different angles and utilizing the rotation invariance of Zernike polynomials to correct errors in wavefront reconstruction and improve the accuracy of 3D structure inference. Based on this, the error-compensated data, combined with symmetry features, is used in a 3D reconstruction algorithm to generate an accurate 3D model of the target object. In a preferred embodiment of the invention, a Linux system can be used as the operating environment, with a 2D image of the target object as input, processed through the steps of the invention to obtain a 3D structure file of the target object.
[0027] This invention presents a 3D structure construction method based on 2D information. By analyzing the mirror symmetry of the target object, the method predicts the object's symmetry plane, reducing uncertainties caused by occlusion or viewpoint limitations. Next, wavefront reconstruction technology is used to simulate the phase changes of the wavefront on the symmetry plane, thereby calculating the wavefront difference. The wavefront difference is used to understand wavefront distortion, which is often the source of error in 2D-to-3D conversion. Furthermore, by solving polynomial coefficient equations, distortion is captured and quantified. Finally, these polynomial coefficients are applied to compensate for fundamental errors, correcting the wavefront data and generating a more accurate 3D structure on the mirror symmetry plane. This not only improves the accuracy of 3D reconstruction but also enhances the model's practicality and reliability by reducing errors and uncertainties.
[0028] Optionally, obtaining the mirror symmetry plane of the target object based on the target object in the two-dimensional image includes: Feature extraction is performed on the target object in the two-dimensional image to obtain the geometric features of the target object; Based on the geometric features of the target object, the mirror symmetry plane of the target object is obtained.
[0029] Specifically, firstly, feature extraction is performed on the target object in the 2D image. This involves using image processing and computer vision techniques to identify and quantify the object's geometric features, such as edges, corners, and lines. These features are key information about the object's shape and structure. After feature extraction, the object's mirror symmetry plane is predicted based on these geometric features. Geometric principles and symmetry detection algorithms are applied to identify the object's potential axes of symmetry or centers of symmetry. In a preferred embodiment of the invention, the mirror symmetry plane is determined using probabilistic prediction techniques inspired by the inherent symmetry structures of natural and man-made objects. This not only improves the accuracy of 3D reconstruction but also effectively addresses problems caused by occlusion or viewpoint limitations.
[0030] In this optional embodiment, feature extraction accurately captures the geometric features of the object, providing a basis for determining the plane of symmetry. This helps reduce ambiguity in the 3D reconstruction process, especially when parts of the object are occluded. Secondly, the method of predicting the mirror symmetry plane utilizes the object's inherent symmetry, significantly improving the stability and accuracy of the reconstruction process.
[0031] Optionally, obtaining the mirror symmetry plane of the target object based on its geometric features includes: Based on the geometric features of the target object, determine the symmetry standard of the target object; By using a probabilistic prediction model, the geometric features are probabilistically predicted according to the symmetry standard, thereby generating the mirror symmetry plane of the geometric features of the target object under the symmetry standard.
[0032] Specifically, the step of obtaining the mirror symmetry plane of a target object begins with an in-depth analysis of the object's geometric features to identify basic geometric elements such as edges, corners, and lines, and to determine the object's symmetry standard based on these elements. The symmetry standard refers to the property of an object that can be reflected geometrically along its axis of symmetry or center of symmetry. Once the symmetry standard is determined, a probabilistic prediction model can be used for probabilistic prediction. This model is usually based on statistical learning or machine learning algorithms to make probabilistic predictions on the object's geometric features, so as to reconstruct the mirror symmetry plane of each object through probabilistic prediction.
[0033] In this optional embodiment, by accurately identifying and utilizing the geometric features of objects, the accuracy of 3D reconstruction can be significantly improved, and a probabilistic prediction model can be used to handle various complex and irregular object shapes, making the reconstruction process more flexible and adaptable.
[0034] Optionally, determining the symmetry standard of the target object based on its geometric features includes: Based on the geometric features of the target object, the region and shape of the target object in the two-dimensional image are obtained; Based on the region and shape of the target object in the two-dimensional image, select the axis of symmetry of the target object; Based on the axis of symmetry and the geometric features of the target object, determine the direction of symmetry of the target object; The symmetry standard is determined based on the axis of symmetry and the direction of symmetry.
[0035] Specifically, firstly, by analyzing the geometric features of the target object in the two-dimensional image, the region and shape information of the object in the image are extracted. Then, based on the extracted region and shape, the object's axis of symmetry is selected. For example, for common symmetrical objects, the axis of symmetry is a straight line passing through the center of the object and dividing it into two identical parts. Next, based on the selected axis of symmetry and the object's geometric features, the object's direction of symmetry is determined. This direction of symmetry is typically aligned with the object's natural orientation or main features. Finally, the axis of symmetry and the direction of symmetry are used as a symmetry standard.
[0036] In this optional embodiment, the process of selecting the axis of symmetry and determining the direction of symmetry enables the reconstruction algorithm to adapt to the geometric characteristics of different objects, improving the algorithm's versatility and flexibility. Establishing a symmetry standard not only helps reduce ambiguity in 3D reconstruction but also improves efficiency when dealing with objects with complex geometries.
[0037] Optionally, the step of reconstructing the wavefront based on the mirror symmetry plane of the target object to obtain the wavefront difference of the mirror symmetry plane includes: Obtain the wavefront data of the mirror-symmetric plane; By rotating the mirror-symmetric plane by a preset angle, the wavefront data of the rotated mirror-symmetric plane is obtained; The wavefront difference value of the mirror-symmetric plane is determined based on the phase difference between the wavefront data before rotation and the wavefront data after rotation.
[0038] Specifically, acquiring the initial wavefront data of the mirror-symmetric plane is typically achieved using high-precision optical measurement equipment, such as an interferometer, which can capture the phase distribution of the wavefront on an object's surface or in an optical system. Next, the wavefront data is acquired again by rotating the mirror-symmetric plane by a preset angle. This step simulates the operations that might be performed on the object or optical element during actual measurement, facilitating the analysis of wavefront changes with rotation. Finally, by comparing the wavefront data before and after rotation, the phase difference between them is calculated, thereby determining the wavefront difference values, which reflect the changes in the wavefront under rotation. In a preferred embodiment of the invention, the wavefront difference values before and after rotation can be obtained by adding a rotation of a different angle, based on the rotation invariance of the Zernike polynomial in polar coordinates.
[0039] In this optional embodiment, by acquiring and comparing wavefront data before and after rotation, minute changes in the wavefront can be accurately captured. This helps to identify and compensate for errors that may affect reconstruction accuracy. Furthermore, by quantifying the wavefront difference, necessary input data is provided for subsequent polynomial coefficient solving, thereby achieving effective compensation for fundamental errors.
[0040] Optionally, the step of solving the polynomial coefficient equation based on the wavefront difference to obtain the polynomial coefficients includes: Based on the wavefront difference, the Zernike polynomial is obtained; The polynomial coefficient equation is established based on the Zernike polynomial, wherein the unknowns of the polynomial coefficient equation are the polynomial coefficients in the Zernike polynomial. The polynomial coefficient equation is solved numerically to obtain the polynomial coefficients of each order and weight in the Zernike polynomial.
[0041] Specifically, in some measurement techniques, limitations inherent in the measurement method itself, such as the number of rotations, prevent the complete capture of all wavefront information, resulting in fundamental errors. These errors manifest as a lack of high-frequency components in wavefront reconstruction, particularly in higher-order terms of the Zernike polynomials. Therefore, in this invention, firstly, the wavefront differences are approximated using Zernike polynomials, a widely used mathematical tool in optics for describing the phase distribution of wavefronts, especially when dealing with wavefront distortion on spheres, due to their orthogonality and rotational symmetry within the unit circle. Next, a polynomial coefficient equation is established, where the unknowns are the coefficients of the Zernike polynomials, representing the projections of the wavefront differences onto the polynomial basis. Finally, this system of equations is solved numerically, such as using least squares, to obtain the polynomial coefficients corresponding to each order and weight.
[0042] In this optional embodiment, Zernike polynomials are used to approximate the wavefront difference, describing the phase error of the wavefront in a systematic and standardized manner, thereby simplifying the complexity of the problem. This significantly improves the accuracy of wavefront reconstruction and reduces inaccuracies in reconstruction due to fundamental errors.
[0043] Optionally, the step of compensating for the fundamental error of the mirror-symmetric plane based on the polynomial coefficients includes: The compensation order and compensation weight are determined based on the preset accuracy. The polynomial coefficients in the Zernike polynomial are determined based on the compensation order and the compensation weight. The wavefront data before rotation is compensated based on the polynomial coefficients to obtain the final wavefront data of the mirror-symmetric plane. Based on the final wavefront data, the final mirror-symmetric plane is obtained.
[0044] Specifically, the error compensation process first requires determining the order and weights of the compensation based on preset accuracy requirements. This is because the order and weights determine the level of detail in the compensation and the computational resources required. The selection of the order and weights is based on a trade-off between the required accuracy and computational cost. Next, using these compensation orders and weights, specific coefficients for compensating for fundamental errors are determined from the polynomial coefficients obtained previously. Then, these coefficients are applied to the unrotated wavefront data to perform error compensation, thereby obtaining the corrected wavefront data. This compensation process involves adjusting the original wavefront data to eliminate errors caused by measurement limitations or system imperfections. Finally, a mirror-symmetric plane is generated based on the compensated wavefront data.
[0045] In this optional embodiment, by determining the compensation order and weights according to a preset precision, it is ensured that the compensation process meets the required precision without causing unnecessary computational burden. Secondly, selecting appropriate polynomial coefficients for error compensation can effectively reduce fundamental errors in wavefront reconstruction and improve the accuracy of three-dimensional structure recovery.
[0046] Optionally, generating the three-dimensional structure of the target object based on the error-compensated mirror symmetry plane includes: Using the final mirror symmetry plane as a constraint, and based on the two-dimensional image of the target object, symmetry processing is performed according to the symmetry standard to generate the three-dimensional structure of the target object.
[0047] Specifically, a mirror-symmetry plane is determined using error-compensated wavefront data. Then, combined with a two-dimensional image of the target object, the previously determined symmetry standard is applied to symmetry the object. This involves reflecting half of the object along the axis of symmetry in the computational model to generate a complete three-dimensional structure. In a preferred embodiment of the invention, this process can utilize computer graphics and computational geometry techniques to achieve symmetric extension of the object through mathematical operations.
[0048] In this optional embodiment, the error-compensated mirror symmetry plane is used as a constraint, which significantly improves the accuracy of 3D structure reconstruction. The compensation process corrects errors that may affect the accuracy of reconstruction. Secondly, the symmetry processing utilizes the inherent symmetry of the object, reducing the amount of computation while avoiding complex calculations on the entire object, and also improving the reconstruction speed.
[0049] Combination Figure 4 As shown, the present invention also provides a three-dimensional structure construction system based on two-dimensional information, comprising: A mirror symmetry unit is used to obtain the mirror symmetry plane of the target object based on the target object in the two-dimensional image; A wavefront reconstruction unit is used to reconstruct the wavefront based on the mirror symmetry plane of the target object to obtain the wavefront difference of the mirror symmetry plane. The equation calculation unit is used to solve the polynomial coefficient equation based on the wavefront difference to obtain the polynomial coefficients. A three-dimensional construction unit is used to compensate for the fundamental error of the mirror symmetry plane based on the polynomial coefficients, and to generate the three-dimensional structure of the target object based on the error-compensated mirror symmetry plane.
[0050] This invention presents a 3D structure construction system based on 2D information. By analyzing the mirror symmetry of a target object, it predicts the object's plane of symmetry, reducing uncertainties caused by occlusion or viewpoint limitations. Next, using wavefront reconstruction technology, it simulates the phase changes of the wavefront on the plane of symmetry and calculates the wavefront difference. The wavefront difference is used to understand wavefront distortion, which is often the source of error in 2D-to-3D conversion. Furthermore, by solving polynomial coefficient equations, distortion is captured and quantified. Finally, these polynomial coefficients are applied to compensate for fundamental errors, correcting the wavefront data and generating a more accurate 3D structure on the mirror symmetry plane. This not only improves the accuracy of 3D reconstruction but also enhances the model's practicality and reliability by reducing errors and uncertainties.
[0051] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, it implements the above-described method for constructing a three-dimensional structure based on two-dimensional information.
[0052] The computer-readable storage medium of this invention predicts the plane of symmetry of a target object by analyzing its mirror symmetry, reducing uncertainties caused by occlusion or viewpoint limitations. Next, wavefront reconstruction technology is used to simulate the phase changes of the wavefront on the plane of symmetry, thereby calculating the wavefront difference. The wavefront difference is used to understand wavefront distortion, which is often the source of error in 2D-to-3D conversion. Furthermore, by solving polynomial coefficient equations, distortion is captured and quantified. Finally, these polynomial coefficients are applied to compensate for fundamental errors, correcting the wavefront data and generating a more accurate 3D structure on the mirror symmetry plane. This not only improves the accuracy of 3D reconstruction but also enhances the practicality and reliability of the model by reducing errors and uncertainties.
[0053] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided by this invention can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.
[0054] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.
[0055] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.
Claims
1. A method for constructing a three-dimensional structure based on two-dimensional information, characterized in that, include: Based on the target object in the two-dimensional image, obtain the mirror symmetry plane of the target object; Wavefront reconstruction is performed based on the mirror symmetry plane of the target object to obtain the wavefront difference of the mirror symmetry plane; Solve the polynomial coefficient equation based on the wavefront difference to obtain the polynomial coefficients; The principle error of the mirror symmetry plane is compensated according to the polynomial coefficients, and the three-dimensional structure of the target object is generated according to the mirror symmetry plane after error compensation.
2. The method for constructing a three-dimensional structure based on two-dimensional information according to claim 1, characterized in that, Based on the target object in the two-dimensional image, obtain the mirror symmetry plane of the target object, including: Feature extraction is performed on the target object in the two-dimensional image to obtain the geometric features of the target object; Based on the geometric features of the target object, the mirror symmetry plane of the target object is obtained.
3. The method for constructing a three-dimensional structure based on two-dimensional information according to claim 2, characterized in that, The step of obtaining the mirror symmetry plane of the target object based on the geometric features of the target object includes: Based on the geometric features of the target object, determine the symmetry standard of the target object; By using a probabilistic prediction model, the geometric features are probabilistically predicted according to the symmetry standard, thereby generating the mirror symmetry plane of the geometric features of the target object under the symmetry standard.
4. The method for constructing a three-dimensional structure based on two-dimensional information according to claim 3, characterized in that, Determining the symmetry standard of the target object based on its geometric features includes: Based on the geometric features of the target object, the region and shape of the target object in the two-dimensional image are obtained; Based on the region and shape of the target object in the two-dimensional image, select the axis of symmetry of the target object; Based on the axis of symmetry and the geometric features of the target object, determine the direction of symmetry of the target object; The symmetry standard is determined based on the axis of symmetry and the direction of symmetry.
5. The method for constructing a three-dimensional structure based on two-dimensional information according to claim 4, characterized in that, The step of reconstructing the wavefront based on the mirror symmetry plane of the target object to obtain the wavefront difference of the mirror symmetry plane includes: Obtain the wavefront data of the mirror-symmetric plane; By rotating the mirror-symmetric plane by a preset angle, the wavefront data of the rotated mirror-symmetric plane is obtained; The wavefront difference value of the mirror-symmetric plane is determined based on the phase difference between the wavefront data before rotation and the wavefront data after rotation.
6. The method for constructing a three-dimensional structure based on two-dimensional information according to claim 5, characterized in that, The process of solving the polynomial coefficient equation based on the wavefront difference to obtain the polynomial coefficients includes: Based on the wavefront difference, the Zernike polynomial is obtained; The polynomial coefficient equation is established based on the Zernike polynomial, wherein the unknowns of the polynomial coefficient equation are the polynomial coefficients in the Zernike polynomial. The polynomial coefficient equation is solved numerically to obtain the polynomial coefficients of each order and weight in the Zernike polynomial.
7. The method for constructing a three-dimensional structure based on two-dimensional information according to claim 6, characterized in that, The step of compensating for the fundamental error of the mirror-symmetric plane based on the polynomial coefficients includes: The compensation order and compensation weight are determined based on the preset accuracy. The polynomial coefficients in the Zernike polynomial are determined based on the compensation order and the compensation weight. The wavefront data before rotation is compensated based on the polynomial coefficients to obtain the final wavefront data of the mirror-symmetric plane. Based on the final wavefront data, the final mirror-symmetric plane is obtained.
8. The method for constructing a three-dimensional structure based on two-dimensional information according to claim 7, characterized in that, The step of generating the three-dimensional structure of the target object based on the error-compensated mirror symmetry plane includes: Using the final mirror symmetry plane as a constraint, and based on the two-dimensional image of the target object, symmetry processing is performed according to the symmetry standard to generate the three-dimensional structure of the target object.
9. A three-dimensional structure construction system based on two-dimensional information, characterized in that, include: A mirror symmetry unit is used to obtain the mirror symmetry plane of the target object based on the target object in the two-dimensional image; A wavefront reconstruction unit is used to reconstruct the wavefront based on the mirror symmetry plane of the target object to obtain the wavefront difference of the mirror symmetry plane. The equation calculation unit is used to solve the polynomial coefficient equation based on the wavefront difference to obtain the polynomial coefficients. A three-dimensional construction unit is used to compensate for the fundamental error of the mirror symmetry plane based on the polynomial coefficients, and to generate the three-dimensional structure of the target object based on the error-compensated mirror symmetry plane.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the three-dimensional structure construction method based on two-dimensional information as described in any one of claims 1 to 8.