A method for three-dimensional measurement of surface structure light based on holographic interference

Abstract

The present application relates to the field of optical three-dimensional measurement, and discloses a kind of surface structure light three-dimensional measurement method based on holographic interference, comprising the following steps: extracting object light complex amplitude from the holographic interference data;Multiple candidate propagation distances are generated based on the propagation distance range;The object light complex amplitude is numerically propagated according to the candidate propagation distance;According to the residual complex transfer parameter corresponding to each candidate propagation distance, the phase shift reconstruction data in the phase shift base frequency complex data is corrected;Coordinate closure information is generated;Target propagation distance and target projection coordinates are determined;The three-dimensional surface data of the object to be measured is generated.Through the object light complex amplitude extraction of the surface structure light holographic interference data corresponding to different spatial frequencies and different phase shift states, and using the residual complex transfer parameter corresponding to the target propagation distance to correct the phase shift base frequency complex data, the local refocusing distance and the structure light phase are determined.

Description

Technical Field

[0001] This invention relates to the field of optical three-dimensional measurement technology, specifically to a surface structured light three-dimensional measurement method based on holographic interference. Background Technology

[0002] With the development of precision manufacturing, micro-nano fabrication, and complex surface inspection technologies, the demand for non-contact 3D measurement of the surface of objects under test is constantly increasing. Structured light 3D measurement projects coded fringes and acquires deformed images, then uses the joint calibration relationship between the camera and the projection unit to recover the object's 3D coordinates. Digital holographic measurement, on the other hand, records the holographic interferogram formed by the object light and the reference light, recovers the complex amplitude of the object light, and obtains the reconstructed light field at different propagation distances through numerical propagation.

[0003] Existing surface structured light 3D measurement typically obtains the phase through multi-phase-shifting fringe demodulation, and then combines it with multi-spatial frequency coding to determine the projection coordinates and generate 3D surface data. However, for the object to be measured with large depth variations, small steps, deep grooves or continuous curved surfaces, the object distances corresponding to different regions are different. A single imaging plane is difficult to make all regions simultaneously within the effective depth of field. Local defocusing can easily cause structured light phase mixing, which in turn affects the accurate determination of the target propagation distance and the structured light phase. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a holographic interference-based three-dimensional measurement method for surface structured light, which solves the problem that local defocusing in existing surface structured light three-dimensional measurements can easily cause phase mixing of structured light, thereby affecting the accurate determination of target propagation distance and structured light phase.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a three-dimensional measurement method based on holographic interference and structured light, comprising the following steps: Acquire surface structured light holographic interference data of the object under test corresponding to different spatial frequencies and different phase shift states, and extract the optical complex amplitude from the holographic interference data; The initial propagation distance and propagation distance range are determined based on the object light complex amplitude, and multiple candidate propagation distances are generated based on the propagation distance range; The object light complex amplitude is propagated numerically according to the candidate propagation distance to obtain phase shift reconstruction data corresponding to different candidate propagation distances; Phase shift coding consistency information is obtained from the phase shift reconstruction data, and the phase shift fundamental frequency complex data in the phase shift reconstruction data is corrected according to the residual complex transfer parameters corresponding to each candidate propagation distance to obtain the candidate correction phase; The common projection coordinates are determined based on the candidate correction phases corresponding to different spatial frequencies, and coordinate closure information is generated. Geometric depth information is determined based on the common projection coordinates, and a joint constraint relationship is constructed by combining the candidate propagation distance, the phase shift coding consistency information, and the coordinate closure information to determine the target propagation distance and target projection coordinates; Based on the target propagation distance, residual complex transfer correction is performed on the phase-shift reconstruction data to obtain the corrected structured light phase; Based on the corrected structured light phase, the target projection coordinates, and the joint calibration relationship between the camera and the projection unit, three-dimensional surface data of the object under test are generated.

[0006] By adopting the above technical solution, the complex amplitude of the object light is extracted from the holographic interferometric data of surface structured light corresponding to different spatial frequencies and different phase shift states. Numerical propagation is performed at multiple candidate propagation distances. The candidate propagation distances are jointly constrained by combining phase shift coding consistency information, common projection coordinates corresponding to multiple spatial frequencies, coordinate closure information, and geometric depth information, thereby determining the target propagation distance and target projection coordinates. The residual complex transfer parameters corresponding to the target propagation distance are used to correct the phase shift fundamental frequency complex data. This achieves the coordinated determination of local refocusing distance and structured light phase, solving the problem that local defocusing in existing surface structured light 3D measurements easily causes structured light phase mixing, which in turn affects the accurate determination of target propagation distance and structured light phase.

[0007] Preferably, the step of acquiring surface structured light holographic interferometric data of the object under test corresponding to different spatial frequencies and different phase shift states includes: Set up multiple spatial frequency-corresponding surface structured light projection sequences; Multiple phase shift states are set for each spatial frequency; Structured light is projected onto the object under test in spatial frequency order and phase shift state order. The object light returned by the object under test and the reference light form a holographic interferogram at the image sensor, and the holographic interferogram data corresponding to each spatial frequency and each phase shift state are collected. The acquired holographic interference data are classified and stored according to spatial frequency identifiers and phase shift state identifiers.

[0008] Preferably, extracting the optical complex amplitude from the holographic interferometric data includes: The holographic interference data is frequency domain transformed, and the frequency domain transformed data is divided into spectral regions to extract the target spectral region containing object light information; The object light complex amplitude of the sensor plane is obtained by performing an inverse transformation based on the target spectral region. The object optical complex amplitude is corrected by a reference light to obtain the corrected object optical complex amplitude.

[0009] Preferably, generating multiple candidate propagation distances based on the propagation distance range includes: The object optical complex amplitude corresponding to the first spatial frequency with the smallest frequency value among multiple spatial frequencies is selected as the initial measurement data; The initial measurement data is propagated along a preset propagation distance; Obtain the phase shift coding state corresponding to different propagation distances; The initial propagation distance is determined based on the phase shift coding state; Generate the propagation distance range based on the discrete range corresponding to the initial propagation distance; Within the propagation distance range, a set of candidate propagation distances is generated at preset distance intervals.

[0010] Preferably, obtaining phase shift reconstruction data corresponding to different candidate propagation distances includes: Obtain a candidate propagation distance set, and propagate the object light complex amplitude corresponding to each spatial frequency according to the candidate propagation distance set; The coordinates of the reconstructed data corresponding to different candidate propagation distances are uniformly processed. Phase shift reconstruction data corresponding to candidate propagation distances are generated based on the reconstructed data after coordinate unification.

[0011] Preferably, obtaining the candidate correction phase includes: Extract the phase shift reconstruction data corresponding to each candidate propagation distance; Phase information is extracted from the phase shift reconstruction data according to a preset phase shift state; Obtain phase shift coding matching information corresponding to different candidate propagation distances; The coding consistency result corresponding to the candidate propagation distance is determined based on the phase shift coding matching information; Obtain the phase-shifted fundamental frequency complex data corresponding to each spatial frequency; Based on the candidate propagation distance, spatial frequency, and image location, the target residual complex transfer parameters are extracted from the pre-calibrated set of residual complex transfer parameters; The phase-shifted fundamental frequency complex data is subjected to amplitude and phase correction based on the target residual complex transfer parameters to form a candidate correction phase.

[0012] Preferably, the generation of coordinate closure information includes: Establish the correspondence between spatial frequency and projected coordinates; Candidate projection coordinates are generated based on the candidate correction phase corresponding to each spatial frequency. Consistency judgment is performed on candidate projection coordinates corresponding to different spatial frequencies; Determine the common projection coordinates that satisfy the multi-spatial frequency constraints; The matching states of each spatial frequency relative to the common projection coordinates are collected to form coordinate closure information.

[0013] Preferably, determining the target propagation distance and target projection coordinates includes: Obtain camera imaging parameters and surface structured light projection parameters; Establish the correspondence between the camera imaging position and the projection position based on the common projection coordinates; Based on the correspondence, the spatial location of the point to be measured is determined, and candidate three-dimensional location data is generated. Extract geometric depth information from the candidate 3D location data; Based on the pre-defined correspondence between geometric depth and propagation distance, a predicted propagation distance corresponding to the geometric depth information is generated; The predicted propagation distance, the candidate propagation distance, the phase shift coding consistency information, and the coordinate closure information are correlated to determine the target propagation distance, and the common projection coordinates corresponding to the target propagation distance are determined as the target projection coordinates.

[0014] Preferably, obtaining the corrected structured light phase includes: Obtain the residual complex transfer parameters corresponding to the target propagation distance; Extract the target phase shift fundamental frequency complex data from the phase shift reconstruction data corresponding to the target propagation distance; The target phase-shift fundamental frequency complex data is corrected based on the residual complex transfer parameters; Obtain the corrected structured light complex phase information; A corrected structured light phase is generated based on the structured light complex phase information.

[0015] Preferably, generating the three-dimensional surface data of the object to be tested includes the following steps: Obtain the projection coordinate information corresponding to the phase of the corrected structured light; The spatial point position is determined based on the projection coordinate information and the camera projection calibration relationship; Combine the positions of multiple spatial points; Generate the three-dimensional surface data corresponding to the object under test.

[0016] This invention provides a 3D measurement method for surface structured light based on holographic interferometry. It has the following advantages: 1. This invention extracts the complex amplitude of the object beam from holographic interferometric data of surface structured light corresponding to different spatial frequencies and phase shift states, performs numerical propagation at multiple candidate propagation distances, and combines phase shift coding consistency information, common projection coordinates corresponding to multiple spatial frequencies, coordinate closure information, and geometric depth information to jointly constrain the candidate propagation distances, thereby determining the target propagation distance and target projection coordinates. Furthermore, it uses the residual complex transfer parameters corresponding to the target propagation distance to correct the phase shift fundamental frequency complex data, achieving the coordinated determination of local refocusing distance and structured light phase. This solves the problem that in existing surface structured light 3D measurements, local defocusing easily causes structured light phase mixing, which in turn affects the accurate determination of the target propagation distance and structured light phase.

[0017] 2. This invention utilizes phase-shift coding rules, the consistency relationship of projection coordinates between phases of multiple spatial frequencies, and the geometric measurement relationship of structured light to mutually verify the local reconstructed position. This makes the determination of propagation distance no longer dependent on the single image sharpness or the phase result of a single frequency, thereby reducing the impact of surface reflection differences, insufficient local texture, and phase mixing on refocusing judgment, and reducing the generation of erroneous propagation distances and erroneous projection coordinates.

[0018] 3. This invention addresses the spatial frequency response attenuation and phase shift that still exist after numerical refocusing by performing residual correction on the structured light coding components and using the correction results for three-dimensional spatial point recovery. This can reduce the phase error caused by local defocusing and residual optical path aberrations, and improve the continuity and reliability of three-dimensional measurement results for complex surfaces. Attached Figure Description

[0019] Figure 1 This is a flowchart of a three-dimensional measurement method based on holographic interference using surface structured light proposed in this invention; Figure 2 This is an architecture diagram of a surface structured light three-dimensional measurement system based on holographic interference, as proposed in an embodiment of the present invention. Figure 3 This is a flowchart illustrating the initial propagation distance determination and candidate propagation distance generation proposed in an embodiment of the present invention. Figure 4 This is a flowchart of the numerical propagation and phase shift reconstruction data generation process proposed in an embodiment of the present invention; Figure 5 This is a flowchart of the consistency evaluation process for candidate correction phase and phase shift coding proposed in an embodiment of the present invention. Detailed Implementation

[0020] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] Example 1: In a first embodiment of the present invention, the present invention provides a three-dimensional measurement method based on holographic interference and structured light, such as... Figure 1 As shown, it includes the following steps: Acquire surface structured light holographic interference data of the object under test corresponding to different spatial frequencies and different phase shift states, and extract optical complex amplitude from the holographic interference data; Furthermore, acquire surface structured light holographic interferometry data of the object under test corresponding to different spatial frequencies and different phase shift states, including: Set up multiple spatial frequency-corresponding surface structured light projection sequences; Multiple phase shift states are set for each spatial frequency; Structured light is projected onto the object under test in spatial frequency order and phase shift state order. The object light returned by the object under test and the reference light form a holographic interferogram at the image sensor, and the holographic interferogram data corresponding to each spatial frequency and each phase shift state are collected. The acquired holographic interference data are classified and stored according to spatial frequency identifiers and phase shift state identifiers.

[0022] Furthermore, the optical recovery amplitude was extracted from the holographic interferometry data, including: The holographic interferometric data is transformed into the frequency domain, and the transformed data is divided into spectral regions to extract the target spectral region containing object light information. The object light complex amplitude on the sensor plane is obtained by performing an inverse transformation based on the target spectral region. The object optical complex amplitude is corrected by a reference light to obtain the corrected object optical complex amplitude.

[0023] Specifically, this step is used to establish the correspondence between the projection coding state, the holographic interferogram, and the complex amplitude of the object light. By sequentially changing the spatial frequency and phase shift state of the surface structured light, each holographic interferogram carries distinguishable fringe coding information. Then, off-axis spectral separation and reference light correction are used to recover the complex light field of the sensor plane, providing data input with complete amplitude and phase for subsequent propagation distance search.

[0024] In this embodiment, the first The spatial frequency and the first The projection pattern corresponding to each phase shift state is represented as follows: ,in, Represents projected coordinates The projected grayscale at that location, Indicates the average gray level. Indicates the modulation amplitude. Indicates the first Spatial frequency, Indicates the first A preset phase shift amount; in this embodiment, the spatial frequency is preferably 1 / 64, 1 / 32, and 1 / 16 of a period per projected pixel, the phase shift step number is preferably four steps, and the phase shift amount is 0, ... , and .

[0025] In this embodiment, the object light returned by the object under test With reference light A holographic interferogram is formed at the image sensor. ,in, Represents the coordinates of the image sensor. This represents the holographic intensity data corresponding to the spatial frequency and phase shift state. Generally, the reference light is incident at an angle to separate the object-light cross term from the zero-order and conjugate terms in the frequency domain; the wavelength of the coherent light source is preferably 532 nm, and the exposure time of the image sensor is preferably 2 ms.

[0026] In this embodiment, a two-dimensional frequency domain transformation is performed on the holographic interferogram, and the target spectral region carrying the optical information of the object is extracted. ,in, and These represent the two-dimensional Fourier transform and the two-dimensional inverse Fourier transform, respectively. Represents the sensor's frequency domain coordinates. The filter window represents the positive first-order cross-spectrum of the covered object light. This represents the initial object light complexation data including the influence of the reference light. In some embodiments, the filter window is a two-dimensional circular window whose center is determined by the carrier frequency position, and its radius is preferably one-eighth of the spectral width.

[0027] Correction is performed based on the pre-acquired reference optical complex amplitude and the system's static aberration phase. ,in, This represents the corrected complex amplitude of the sensor plane object light. Indicates the static aberration phase of the system. This represents the reference light amplitude stability parameter; in this embodiment, Preferably 10 times the maximum intensity of the reference light -6 Subsequently, the complex amplitudes of each object's light are organized according to the spatial frequency identifier and phase shift state identifier, thereby ensuring that the complete phase-shifted complex light field sequence at the same spatial frequency can be called during subsequent numerical propagation.

[0028] Please see the appendix Figure 3 The initial propagation distance and propagation distance range are determined based on the complex amplitude of the object light, and multiple candidate propagation distances are generated based on the propagation distance range; Furthermore, multiple candidate propagation distances are generated based on the propagation distance range, including: The object optical complex amplitude corresponding to the first spatial frequency with the smallest frequency value among multiple spatial frequencies is selected as the initial measurement data; The initial measurement data is propagated according to the preset propagation distance; Obtain the phase shift coding state corresponding to different propagation distances; The initial propagation distance is determined based on the phase shift coding state; Generate the propagation distance range based on the discrete range corresponding to the initial propagation distance; Within the propagation distance range, a set of candidate propagation distances is generated at preset distance intervals.

[0029] Specifically, this step utilizes the characteristic that the fundamental frequency attenuation of the first spatial frequency is relatively small in the defocus state to perform a wide-range coarse propagation search on the aforementioned corrected object light complex amplitude, and determines the search center by the degree of conformity of the phase shift sequence to the preset coding model, thereby limiting the propagation range for subsequent multi-spatial frequency joint reconstruction.

[0030] In this embodiment, the complete two-dimensional object optical complex amplitude corresponding to the first spatial frequency with the smallest frequency value is selected, and the object optical complex amplitudes corresponding to each phase shift state are propagated to a coarse propagation distance. : ; ; in, For the first spatial frequency The complex amplitude of the sensor plane object light in each phase-shifted state. For the distance of transmission The corresponding angular spectral transfer function, For coherent light wavelengths, and is the spatial frequency of the light field. In this embodiment, The preferred value is 532 nanometers, and the preferred coarse propagation distance interval is 0.2 millimeters.

[0031] Calculate the background intensity and phase-shifted fundamental frequency complex data based on the reconstructed intensity after propagation: ; ; And construct phase shift coding cost ,in, For the first Each phase shift state over the coarse propagation distance Under the reconstruction intensity, For the number of phase shift steps, To preset the phase shift, For the cost of phase shift coding, For stable parameters, To adjust the system weights. In this embodiment, The preferred value is 4. Preferably 10 -6 , The preferred value is 0.1.

[0032] The initial propagation distance is determined by the coarse propagation distance that minimizes the cost. ,by The propagation distance range is formed by extending two coarse propagation distance intervals to both sides of the center, and a candidate propagation distance set is generated according to a preset distance interval of 0.05 mm. The propagation distance range can be expanded according to the cost change of adjacent coarse propagation distances to avoid the real object surface being excluded due to the initial estimation deviation.

[0033] Please see the appendix Figure 4 The object light complex amplitude is numerically propagated according to the candidate propagation distance to obtain phase shift reconstruction data corresponding to different candidate propagation distances; Furthermore, phase shift reconstruction data corresponding to different candidate propagation distances are obtained, including: Obtain a set of candidate propagation distances, and propagate the object light complex amplitude corresponding to each spatial frequency according to the set of candidate propagation distances; The coordinates of the reconstructed data corresponding to different candidate propagation distances are uniformly processed. Phase shift reconstruction data corresponding to candidate propagation distances are generated based on the reconstructed data after coordinate unification.

[0034] Specifically, this step reconstructs the complete two-dimensional complex optical field for each candidate propagation distance and eliminates the differences in magnification and coordinate offset caused by changes in propagation distance, so that the reconstruction results of different spatial frequencies and phase shift states maintain spatial correspondence.

[0035] For the first in the candidate propagation distance set Candidate propagation distance , will the The spatial frequency and the first angular spectrum propagation of the object light complex amplitude corresponding to each phase shift state. ,in, The complex amplitude of the optical wave on the sensor plane. To reconstruct the complete two-dimensional complex optical field for candidate propagation distances, Let be the angular spectral transfer function corresponding to the candidate propagation distance. and These are two-dimensional Fourier transform and two-dimensional inverse Fourier transform, respectively; in this embodiment, the candidate propagation distance interval is preferably 0.05 mm.

[0036] In this embodiment, instead of performing independent propagation on individual pixels, the complete object light complex amplitude is propagated for each candidate propagation distance to preserve the spatial coupling relationship between adjacent light fields during diffraction; a pre-calibrated coordinate remapping operator is used for coordinate unification. ,in, Indicates candidate propagation distance The corresponding object-space coordinate remapping operator, The reconstructed complex optical field is represented under unified object-space coordinates. Coordinate remapping uses bicubic interpolation, and the optimal sampling interval for unified coordinates is 5 micrometers.

[0037] Phase-shift reconstruction data is generated based on the reconstructed complex light field after unifying coordinates. ,in, To determine the reconstruction intensity corresponding to the spatial frequency, phase shift state, and candidate propagation distance, the reconstruction intensity is associated and stored according to the spatial frequency identifier, phase shift state identifier, and candidate propagation distance identifier, for subsequent phase shift coding consistency analysis.

[0038] Please see the appendix Figure 5 Phase shift coding consistency information is obtained from the phase shift reconstruction data, and the phase shift fundamental frequency complex data in the phase shift reconstruction data is corrected according to the residual complex transfer parameters corresponding to each candidate propagation distance to obtain the candidate correction phase; Furthermore, candidate correction phases are obtained, including: Extract the phase shift reconstruction data corresponding to each candidate propagation distance; Phase information is extracted from the phase shift reconstruction data based on the preset phase shift state; Obtain phase shift coding matching information corresponding to different candidate propagation distances; Determine the coding consistency result corresponding to the candidate propagation distance based on the phase shift coding matching information; Obtain the phase-shifted fundamental frequency complex data corresponding to each spatial frequency; Based on the candidate propagation distance, spatial frequency, and image location, the target residual complex transfer parameters are extracted from the pre-calibrated set of residual complex transfer parameters. The amplitude and phase of the phase-shifted fundamental frequency complex data are corrected based on the target residual complex transfer parameters to form a candidate correction phase.

[0039] Specifically, this step evaluates the local refocusing state by assessing the degree of conformity between the phase shift sequence at each candidate propagation distance and the preset coding model, and compensates for the spatial frequency response attenuation and phase shift that still exist after numerical propagation, thereby obtaining candidate correction phases for subsequent multi-spatial frequency projection coordinate closure.

[0040] In this embodiment, the same candidate propagation distance and the same spatial frequency The corresponding phase-shift reconstructed data are sorted, and the background intensity and phase-shift fundamental frequency complex data are calculated: ; ; in, For the first The reconstruction intensity corresponding to each phase shift state For the number of phase shift steps, This is a preset phase shift amount; in this embodiment, Preferably 4, and .

[0041] Encoded reconstruction data is generated based on background intensity and phase-shifted fundamental frequency complex data, and the phase-shifted coding residual is calculated: ; ; in, To reconstruct the data, For phase shift coding matching information, For stable parameters. In this embodiment, Preferably 10 -6 ,when If the distance is not greater than 0.15, the corresponding candidate propagation distance will be marked as consistent with the encoding.

[0042] The residual complex transfer parameters are determined by the calibrated phase-shifted fundamental frequency complex data and the ideal phase-shifted fundamental frequency complex data. ,in, These are the target residual complex transfer parameters corresponding to the candidate propagation distance, spatial frequency, and image position. Preferably 10 -6 .

[0043] The phase-shifted fundamental frequency complex data is corrected using a regularization method: ; ; in, To encode the mean square residual between the reconstructed data and the phase-shifted reconstructed data, For regularization parameters, The preferred value is 0.01; the final candidate correction phase is expressed as This preserves the effective phase shift coding information while suppressing the amplification of noise components during the residual complex transfer correction process.

[0044] The common projection coordinates are determined based on the candidate correction phases corresponding to different spatial frequencies, and coordinate closure information is generated. Furthermore, coordinate closure information is generated, including: Establish the correspondence between spatial frequency and projected coordinates; Candidate projection coordinates are generated based on the candidate correction phase corresponding to each spatial frequency. Consistency judgment is performed on candidate projection coordinates corresponding to different spatial frequencies; Determine the common projection coordinates that satisfy the multi-spatial frequency constraints; The matching status of each spatial frequency relative to the common projected coordinates is collected to form coordinate closure information.

[0045] Specifically, this step utilizes the physical relationship between the same test point and the same projection position at different spatial frequencies to jointly match each candidate correction phase, thereby providing coordinate basis for the subsequent correlation between propagation distance and structured light geometric information.

[0046] In this embodiment, the first Spatial frequencies in candidate propagation distance The following satisfies ,in, As a candidate correction phase, The spatial frequency of structured light. To calibrate the initial phase, These are the continuous projected coordinates of the point to be measured.

[0047] According to different phase orders Generate candidate projection coordinates ,in, For the first The set of candidate projected coordinates corresponding to each spatial frequency. To ensure that the candidate projection coordinates are within the phase order of the calibration projection range.

[0048] In this embodiment, common projection coordinates are determined within the common range covered by each candidate projection coordinate set: ; ; ; in, For the number of spatial frequencies, For coordinate closure residuals, The circumferential phase difference is used to characterize the deviation between the actual corrected phase and the theoretical phase of the projected coordinates.

[0049] When the circumferential phase difference corresponding to each spatial frequency is not greater than the closure threshold, the corresponding matching state is recorded as valid, and the common projection coordinates, coordinate closure residuals, and matching states of each spatial frequency are collected into coordinate closure information; in this embodiment, Preferably 3, The preferred value is 0, the preferred closure threshold is 0.12 radians, and the preferred projection coordinate search interval is 0.1 projection pixels.

[0050] Geometric depth information is determined based on common projection coordinates, and joint constraint relationships are constructed by combining candidate propagation distance, phase shift coding consistency information, and coordinate closure information to determine the target propagation distance and target projection coordinates; Furthermore, determining the target propagation distance and target projected coordinates includes: Obtain camera imaging parameters and surface structured light projection parameters; Establish the correspondence between the camera imaging position and the projection position based on the common projection coordinates; Determine the spatial location of the point to be measured based on the correspondence, and generate candidate 3D location data; Extract geometric depth information from candidate 3D location data; Based on the pre-defined correspondence between geometric depth and propagation distance, a predicted propagation distance corresponding to the geometric depth information is generated; The predicted propagation distance, candidate propagation distance, phase shift coding consistency information, and coordinate closure information are correlated to determine the target propagation distance, and the common projection coordinates corresponding to the target propagation distance are determined as the target projection coordinates.

[0051] Specifically, this step converts the common projection coordinates into candidate three-dimensional positions of the point to be measured, and uses the calibration relationship between geometric depth and holographic propagation distance to jointly judge the phase shift coding state, cross-space frequency coordinate closure state, and propagation distance matching state, thereby determining the propagation distance and projection coordinates corresponding to the real object surface.

[0052] In this embodiment, the image position is determined based on the camera imaging parameters. Corresponding camera imaging rays ,in, For the camera optical center, Image location The corresponding unit imaging ray, For ray distance parameters, Candidate propagation distance The corresponding candidate 3D position.

[0053] Common projection coordinates The corresponding projection plane satisfies Thus we obtain ,in, Let be the unit normal vector of the projection plane corresponding to the target projection coordinates. This is the calibration constant for the projected light plane.

[0054] Extracting geometric depth from candidate 3D locations ,in, To calibrate the depth direction, This serves as a measurement reference point. The predicted propagation distance is generated based on a pre-calibrated mapping between geometric depth and propagation distance. ,in, The holographic propagation distance calibration mapping represents the image position and geometric depth, and the mapping adopts a quadratic calibration model; in this embodiment, the calibration depth interval is preferably 0.5 mm.

[0055] Construct joint constraint values ​​for candidate propagation distances ,in, The phase shift coding consistency residual corresponding to the candidate propagation distance. For coordinate closure residuals, To allow for deviation in propagation distance, For stable parameters, , and There are three types of constraint weights; in this embodiment, the preferred weights for the three types of constraints are 0.3, 0.4, and 0.3, respectively. Preferably 0.1 mm, Preferably 10 -6 .

[0056] The candidate propagation distance with the minimum joint constraint value and valid phase shift coding matching state and coordinate closure state is determined as the target propagation distance: ; ; And The target projection coordinates are determined. This process ensures that the target propagation distance is simultaneously constrained by the holographic reconstruction state, the surface structured light encoding rules, and the three-dimensional measurement geometry.

[0057] The phase-shifted reconstruction data is subjected to residual complex transfer correction based on the target propagation distance to obtain the corrected structured light phase. Furthermore, obtaining the corrected structured light phase includes: Obtain the residual complex transfer parameters corresponding to the target propagation distance; Extract the fundamental frequency complex data of the target phase shift from the phase shift reconstruction data corresponding to the target propagation distance; The target phase-shift fundamental frequency complex data is corrected based on the residual complex transfer parameters; Obtain the corrected structured light complex phase information; generate the corrected structured light phase based on the structured light complex phase information.

[0058] Specifically, this step involves re-extracting the structured light coding components from the reconstructed object surface corresponding to the target propagation distance, and using the residual complex transfer parameters corresponding to the spatial frequency, image position, and target propagation distance for final correction to reduce the amplitude attenuation and phase shift that still exist after numerical refocusing.

[0059] In this embodiment, the target propagation distance Extracting the first from the corresponding phase-shift reconstruction data Target phase shift fundamental frequency complex data at each spatial frequency ,in, For the target propagation distance, the first The reconstruction intensity of each phase shift state For the number of phase shift steps, This is a preset phase shift amount. In this embodiment, The preferred value is 4.

[0060] Based on the target propagation distance, spatial frequency, and image location, the residual complex transfer parameters of the target are extracted from the residual complex transfer parameter set. And perform regularization correction: ; ; in, This is the corrected structured light complex phase information. This represents the noise variance corresponding to the phase-shift coding residual. For regularization parameters, and For stable parameters. In this embodiment, The preferred value is 0.01. and All are preferably 10 -6 .

[0061] Corrected structured light phase is represented as In some embodiments, the phase order is determined based on the target projection coordinates, and the corrected structured light phase is converted into a continuous phase for use in the subsequent generation of three-dimensional surface data.

[0062] Based on the corrected structured light phase, target projection coordinates, and the joint calibration relationship between the camera and the projection unit, the three-dimensional surface data of the object under test is generated.

[0063] Furthermore, generating the three-dimensional surface data of the object under test includes the following steps: Obtain the projection coordinate information corresponding to the phase of the corrected structured light; Determine the spatial point location based on the projection coordinate information and the camera projection calibration relationship; Combine the positions of multiple spatial points; Generate the three-dimensional surface data corresponding to the object under test.

[0064] Specifically, this step uses the target projection coordinates to determine the phase order of the corrected structured light phase, and recovers the coordinates of the point to be measured by the spatial intersection relationship between the camera imaging ray and the projection light plane, thereby forming the three-dimensional surface data of the object to be measured.

[0065] In this embodiment, based on the target projection coordinates Determine the first Phase order corresponding to each spatial frequency And obtain the verification projection coordinates. ,in, To correct the phase of structured light, To calibrate the initial phase, For spatial frequency, In order to make and The phase order with the smallest distance. Generally, when the difference between the two is no greater than 0.2 projected pixels, the projected coordinate information is marked as valid.

[0066] Determine the spatial point position based on the joint calibration relationship between the camera and the projection unit. ,in, The average value of the verification projected coordinates corresponding to each effective spatial frequency. For the camera optical center, The unit imaging ray corresponding to the image position. Let be the unit normal vector of the projected light plane. For the calibration constant of the projection light plane, The spatial location of the point to be measured.

[0067] Traverse all valid image locations and aggregate the resulting spatial point locations into a 3D point set. ,in, The effective number of spatial points; in some embodiments, three-dimensional points are connected according to the adjacency relationship of image positions to form a depth map or triangular mesh, thereby generating three-dimensional surface data of the object under test.

[0068] Example 2: In a second embodiment of the present invention, the present invention provides a surface structured light three-dimensional measurement system based on holographic interference, such as... Figure 2 As shown, it includes the following modules: Acquisition module: used to acquire surface structured light holographic interference data of the object under test corresponding to different spatial frequencies and different phase shift states, and extract the complex amplitude from the holographic interference data; Initial distance module: used to determine the initial propagation distance and propagation distance range based on the object light complex amplitude, and to generate multiple candidate propagation distances based on the propagation distance range; The propagation module is used to propagate the complex amplitude of the object light numerically according to the candidate propagation distances to obtain phase shift reconstruction data corresponding to different candidate propagation distances. Candidate correction module: used to obtain phase shift coding consistency information from phase shift reconstruction data, and perform candidate correction on the phase shift fundamental frequency complex data in phase shift reconstruction data according to the residual complex transfer parameters corresponding to each candidate propagation distance, to obtain candidate correction phase; Closure module: used to determine common projection coordinates based on candidate correction phases corresponding to different spatial frequencies, and generate coordinate closure information; Joint constraint module: used to determine geometric depth information based on common projection coordinates, and to construct joint constraint relationships by combining candidate propagation distance, phase shift coding consistency information and coordinate closure information, thereby determining the target propagation distance and target projection coordinates; Phase correction module: used to perform residual complex transfer correction on phase shift reconstruction data based on target propagation distance to obtain corrected structured light phase; 3D Reconstruction Module: Used to generate 3D surface data of the object under test based on the corrected structured light phase, target projection coordinates, and the joint calibration relationship between the camera and the projection unit.

[0069] In the surface inspection of precision devices, the object under test may have continuous curved surfaces, deep grooves, steps, and tiny protrusions. The distances between different surface regions and the imaging system vary significantly, making it difficult to maintain clarity within the same imaging plane. When using conventional surface structured light measurement, local defocusing can cause a decrease in fringe modulation and mix the phases of structured light carried at adjacent positions, leading to a deviation between the phase demodulation result and the actual projected position. If the reconstruction position is selected solely based on the clarity of a single image, incorrect propagation distances may be chosen due to differences in surface reflection or insufficient texture, thus affecting the continuity and reliability of the 3D surface reconstruction. To address these issues, this invention employs a holographic interferometry-based surface structured light 3D measurement system, the architecture of which is as follows: Figure 2 As shown. The specific implementation process of this system is as follows: First, the acquisition module sequentially acquires holographic interference data corresponding to different spatial frequencies and different phase shift states, and extracts the object beam complex amplitude under each projection state; Subsequently, the initial distance module uses the spatial frequency with the smallest frequency value to perform a coarse propagation search, determine the initial propagation distance range, and generate candidate propagation distances; Next, the propagation module propagates the optical complex amplitude of each object to different candidate object surfaces and unifies the coordinates to form phase-shift reconstruction data; The candidate correction module evaluates each candidate surface based on the phase shift coding matching state and corrects the phase shift fundamental frequency complex data using the residual complex transfer parameters; The closure module determines the common projection coordinates based on the candidate correction phases of multiple spatial frequencies, forming coordinate closure information; The joint constraint module correlates geometric depth, candidate propagation distance, phase shift coding consistency information, and coordinate closure information to determine the target propagation distance and target projection coordinates; The phase correction module completes the final residual complex transfer correction at the target propagation distance to obtain the corrected structured light phase; Finally, the 3D reconstruction module generates multiple 3D measurement points based on the target projection coordinates and joint calibration relationships, and aggregates them to form the 3D surface data of the object under test.

[0070] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A three-dimensional measurement method based on holographic interference and structured light, characterized in that, Includes the following steps: Acquire surface structured light holographic interference data of the object under test corresponding to different spatial frequencies and different phase shift states, and extract the optical complex amplitude from the holographic interference data; The initial propagation distance and propagation distance range are determined based on the object light complex amplitude, and multiple candidate propagation distances are generated based on the propagation distance range; The object light complex amplitude is propagated numerically according to the candidate propagation distance to obtain phase shift reconstruction data corresponding to different candidate propagation distances; Phase shift coding consistency information is obtained from the phase shift reconstruction data, and the phase shift fundamental frequency complex data in the phase shift reconstruction data is corrected according to the residual complex transfer parameters corresponding to each candidate propagation distance to obtain the candidate correction phase; The common projection coordinates are determined based on the candidate correction phases corresponding to different spatial frequencies, and coordinate closure information is generated. Geometric depth information is determined based on the common projection coordinates, and a joint constraint relationship is constructed by combining the candidate propagation distance, the phase shift coding consistency information, and the coordinate closure information to determine the target propagation distance and target projection coordinates; Based on the target propagation distance, residual complex transfer correction is performed on the phase-shift reconstruction data to obtain the corrected structured light phase; Based on the corrected structured light phase, the target projection coordinates, and the joint calibration relationship between the camera and the projection unit, three-dimensional surface data of the object under test are generated.

2. The method for three-dimensional measurement of surface structured light based on holographic interferometry according to claim 1, characterized in that: The acquisition of surface structured light holographic interferometry data of the object under test corresponding to different spatial frequencies and different phase shift states includes: Set up multiple spatial frequency-corresponding surface structured light projection sequences; Multiple phase shift states are set for each spatial frequency; Structured light is projected onto the object under test in spatial frequency order and phase shift state order. The object light returned by the object under test and the reference light form a holographic interferogram at the image sensor, and the holographic interferogram data corresponding to each spatial frequency and each phase shift state are collected. The acquired holographic interference data are classified and stored according to spatial frequency identifiers and phase shift state identifiers.

3. The method for three-dimensional measurement of surface structured light based on holographic interferometry according to claim 1, characterized in that: Extracting the optical recovery amplitude from the holographic interferometric data includes: The holographic interference data is frequency domain transformed, and the frequency domain transformed data is divided into spectral regions to extract the target spectral region containing object light information; The object light complex amplitude of the sensor plane is obtained by performing an inverse transformation based on the target spectral region. The object optical complex amplitude is corrected by a reference light to obtain the corrected object optical complex amplitude.

4. The method for three-dimensional measurement of surface structured light based on holographic interferometry according to claim 1, characterized in that: The generation of multiple candidate propagation distances based on the propagation distance range includes: The object optical complex amplitude corresponding to the first spatial frequency with the smallest frequency value among multiple spatial frequencies is selected as the initial measurement data; The initial measurement data is propagated along a preset propagation distance; Obtain the phase shift coding state corresponding to different propagation distances; The initial propagation distance is determined based on the phase shift coding state; Generate the propagation distance range based on the discrete range corresponding to the initial propagation distance; Within the propagation distance range, a set of candidate propagation distances is generated at preset distance intervals.

5. The method for three-dimensional measurement of surface structured light based on holographic interferometry according to claim 1, characterized in that: The process of obtaining phase shift reconstruction data corresponding to different candidate propagation distances includes: Obtain a candidate propagation distance set, and propagate the object light complex amplitude corresponding to each spatial frequency according to the candidate propagation distance set; The coordinates of the reconstructed data corresponding to different candidate propagation distances are uniformly processed. Phase shift reconstruction data corresponding to candidate propagation distances are generated based on the reconstructed data after coordinate unification.

6. The method for three-dimensional measurement of surface structured light based on holographic interferometry according to claim 1, characterized in that: The process of obtaining the candidate correction phase includes: Extract the phase shift reconstruction data corresponding to each candidate propagation distance; Phase information is extracted from the phase shift reconstruction data according to a preset phase shift state; Obtain phase shift coding matching information corresponding to different candidate propagation distances; The coding consistency result corresponding to the candidate propagation distance is determined based on the phase shift coding matching information; Obtain the phase-shifted fundamental frequency complex data corresponding to each spatial frequency; Based on the candidate propagation distance, spatial frequency, and image location, the target residual complex transfer parameters are extracted from the pre-calibrated set of residual complex transfer parameters; The phase-shifted fundamental frequency complex data is subjected to amplitude and phase correction based on the target residual complex transfer parameters to form a candidate correction phase.

7. The method for three-dimensional measurement of surface structured light based on holographic interference according to claim 1, characterized in that: The generation of coordinate closure information includes: Establish the correspondence between spatial frequency and projected coordinates; Candidate projection coordinates are generated based on the candidate correction phase corresponding to each spatial frequency. Consistency judgment is performed on candidate projection coordinates corresponding to different spatial frequencies; Determine the common projection coordinates that satisfy the multi-spatial frequency constraints; The matching states of each spatial frequency relative to the common projection coordinates are collected to form coordinate closure information.

8. The method for three-dimensional measurement of surface structured light based on holographic interferometry according to claim 1, characterized in that: Determining the target propagation distance and target projection coordinates includes: Obtain camera imaging parameters and surface structured light projection parameters; Establish the correspondence between the camera imaging position and the projection position based on the common projection coordinates; Based on the correspondence, the spatial location of the point to be measured is determined, and candidate three-dimensional location data is generated. Extract geometric depth information from the candidate 3D location data; Based on the pre-defined correspondence between geometric depth and propagation distance, a predicted propagation distance corresponding to the geometric depth information is generated; The predicted propagation distance, the candidate propagation distance, the phase shift coding consistency information, and the coordinate closure information are correlated to determine the target propagation distance, and the common projection coordinates corresponding to the target propagation distance are determined as the target projection coordinates.

9. The method for three-dimensional measurement of surface structured light based on holographic interferometry according to claim 1, characterized in that: The process of obtaining the corrected structured light phase includes: Obtain the residual complex transfer parameters corresponding to the target propagation distance; Extract the target phase shift fundamental frequency complex data from the phase shift reconstruction data corresponding to the target propagation distance; The target phase-shift fundamental frequency complex data is corrected based on the residual complex transfer parameters; Obtain the corrected structured light complex phase information; A corrected structured light phase is generated based on the structured light complex phase information.

10. A method for three-dimensional measurement of surface structured light based on holographic interferometry according to claim 1, characterized in that: The process of generating the three-dimensional surface data of the object to be tested includes the following steps: Obtain the projection coordinate information corresponding to the phase of the corrected structured light; The spatial point position is determined based on the projection coordinate information and the camera projection calibration relationship; Combine the locations of multiple spatial points; Generate the three-dimensional surface data corresponding to the object under test.

Images