Method and system for flow field measurement based on structured light coding and dual-view light field imaging

Abstract

This invention discloses a flow field measurement method and system based on structured light coding and dual-view optical field imaging. The method includes calibrating the resolution and point spread function of the imaging system to generate orthogonal structured light fringes; constructing an excitation optical path to generate three orthogonal structured lights of different colors with a phase difference of 120°; acquiring the three phase structured light fringes within a single camera exposure time; constructing an experimental optical path for dual-view optical field imaging and reconstructing the optical field from the imaging results under the illumination of the three different phase structured lights; performing structured light super-resolution reconstruction on the optical field images under dual-view to obtain super-resolution images of particles in the Y and Z directions; superimposing the dual-view images of the super-resolution images to obtain a super-resolution three-dimensional particle distribution image; performing cross-correlation calculation on the super-resolution three-dimensional particle distribution image and post-processing the three-dimensional velocity vector field of the flow field.

Description

Technical Field

[0001] This invention relates to the field of optical measurement methods, and specifically to a flow field measurement method and system based on structured light coding and dual-view optical field imaging. Background Technology

[0002] Two-dimensional particle image velocimetry (PIV) is a fluid velocimetry method developed in the late 1970s. Compared with single-point velocimetry methods such as laser Doppler velocimetry and hot-wire anemometers, it has the advantages of being transient, multi-point, and non-contact. However, since many flow phenomena in basic scientific research and practical engineering applications have complex three-dimensional characteristics, the velocity information of two-dimensional flow fields is insufficient to reflect the mathematical and physical essence of flow phenomena. Therefore, how to accurately measure three-dimensional velocity fields has become a research hotspot in the field of particle imaging velocimetry (PIV) in recent decades.

[0003] Current three-dimensional flow field testing technologies mainly include the following:

[0004] (1) Scanning PIV. Scanning PIV uses a two-dimensional laser plane and a specially designed mechanical device to scan a three-dimensional flow field to achieve three-dimensional measurement of the flow field. However, its structure is complex and it is limited by the scanning speed, so it cannot achieve high-speed three-dimensional imaging.

[0005] (2) Defocused PIV. It eliminates blurring when particles are not in the focal plane by using a specific aperture near the camera lens. However, the tracer particle density of defocused PIV technology is severely limited, making it unable to image and measure high-speed flowing fields.

[0006] (3) Holographic PIV. It reconstructs the volume position of the tracer particle by recording the interference pattern (hologram) generated when a coherent beam passes through the particle volume. Typically, a special holographic projection film is needed to record the interference pattern, a cumbersome and time-consuming process. In recent years, digital holographic PIV has made significant progress, but it still suffers from low resolution and insufficient velocity vector reconstruction. Tomographic PIV uses multiple cameras to record tracer particle information from different angles and reconstructs the three-dimensional velocity vector field through tomographic imaging. It has been widely used in the field of flow, but requires a complex and expensive camera array system and tedious system debugging.

[0007] (4) Synthetic aperture PIV. This is another multi-camera 3D PIV technique, but it requires more cameras compared to chromatographic PIV.

[0008] Unlike the aforementioned 3D PIV testing techniques, Light Field PIV (LF-PIV) is a relatively recent development that can accurately measure 3D velocity fields using only a single camera. It eliminates the need for complex optical systems and multiple cameras, significantly simplifying experimental operations and reducing hardware costs. It is particularly suitable for measuring fluid 3D velocity fields in situations with limited optical windows. Compared to traditional methods, light field imaging can simultaneously acquire position and angle information, requiring only a single image for 3D position data acquisition, facilitating rapid 3D imaging and greatly simplifying the PIV system. However, because a light field camera is essentially a compact camera array with low parallax, the overall observation angle of the tracer particles during image capture is relatively small. This causes the reconstructed particles to be elongated along the optical axis, resulting in a conical field distribution and thus lower spatial resolution along the optical axis. Furthermore, light field imaging increases imaging depth of field by sacrificing lateral resolution, so its lateral resolution is worse than that of traditional methods. Although PIV imaging can improve the axial resolution of light field imaging by using two light field cameras arranged vertically, it can only be improved to a level comparable to its lateral resolution, and cannot achieve super-resolution imaging. Therefore, it limits the measurement of fine flow by the PIV system. Summary of the Invention

[0009] Purpose of the invention: To address the shortcomings of existing technologies, such as the need for numerous devices and low spatial resolution, this invention provides a flow field measurement method and system based on structured light coding and dual-view light field imaging.

[0010] Technical Solution: To solve the above problems, this invention employs a flow field measurement method based on structured light coding and dual-view optical field imaging, comprising the following steps:

[0011] Step 1: Calibrate the resolution and point spread function of the imaging system to generate three orthogonal structured light fringes with different phases;

[0012] Step 2: Construct the excitation optical path, modulate the trigger signal between the digital micromirror array and the three lasers, and sequentially generate three orthogonal structured lights of different colors with a phase difference of 120° in the imaging area;

[0013] Step 3: Adjust the relationship between the camera exposure time and the digital micromirror array to acquire structured light fringes of three phases within a single camera exposure time;

[0014] Step 4: Construct the experimental optical path for dual-view imaging of the light field. Use a light field camera to image the flow field under test in the frontal view and the top view reflected by a 45-degree mirror. Reconstruct the light field from the imaging results under structured light illumination of three different phases.

[0015] Step 5: Obtain the spatial information of structured light fringes at different depths from two perspectives: the frontal view and the top view reflected by a 45-degree mirror from the light field reconstruction pattern. Extract the frequency and phase information of the structured light. Based on the frequency and phase of the structured light, perform structured light super-resolution reconstruction on the light field images under the two perspectives to obtain the three-dimensional frontal view image of the particle super-resolution image and the three-dimensional top view image of the particle super-resolution image.

[0016] Step 6: Superimpose the three-dimensional front view and the three-dimensional top view of the particle super-resolution image to obtain a three-dimensional particle distribution image with super-resolution.

[0017] Step 7: Perform cross-correlation calculation on the three-dimensional particle distribution image to obtain the three-dimensional velocity vector field of the flow field to be measured, and perform post-processing on the three-dimensional velocity vector field of the flow field.

[0018] Furthermore, the specific steps for constructing the excitation optical path in step 2 are as follows: the three-color lasers emitted by the three lasers are converged through the optical path, and the surface of the modulated digital micromirror array is illuminated by a reflector. The structured light is generated by the stripes loaded by the modulated digital micromirror array, the zero-order stripe is selected by a spatial filter, and then the light is amplified by a lens combination to excite and illuminate the flow field space.

[0019] Furthermore, the wavelengths of the three-color lasers are 405nm, 532nm, and 650nm, respectively.

[0020] Furthermore, the specific operation of structured light super-resolution reconstruction described in step 5 is as follows: Wiener filtering is applied to the image to remove background noise and improve the signal-to-noise ratio of the reconstructed image; then, a deconvolution algorithm is used to reconstruct the frequency components. and The centers are offset to their correct positions in the frequency domain space (+p) θ and -p θ This method collects twice the amount of information, enabling super-resolution reconstruction of refocused images.

[0021] Furthermore, the super-resolution particle three-dimensional distribution image mentioned in step 6 refers to the superposition of two-view particle three-dimensional images to eliminate the axial stretching effect of the single-view particle three-dimensional image, thereby achieving super-resolution reconstruction of the position and shape of three-dimensional particles and obtaining the three-dimensional flow field velocity.

[0022] Furthermore, the specific steps of the cross-correlation calculation in step 7 are as follows: the three-dimensional distribution image of particles is divided into multiple sampling volumes of the same size. By performing Fourier transform multiplication on the corresponding sampling volumes of two consecutive voxel matrices, and performing inverse Fourier transform on the product in the frequency domain, the cross-correlation function of the sampling region can be obtained, and the three-dimensional velocity vector field of the entire flow field can be further calculated.

[0023] Furthermore, the specific steps for post-processing the three-dimensional velocity vector field of the flow field in step 7 are as follows: remove erroneous velocity vectors from the velocity vector field and insert the correct velocity vectors based on adjacent velocity vectors.

[0024] The present invention also provides a system for measuring flow fields using the aforementioned method based on structured light coding and dual-view optical field imaging, comprising:

[0025] (1) Light field camera - an industrial camera including macro lens, microlens array and main lens;

[0026] (2) Structured light illumination and dual-view light field imaging system, the illumination of which includes three lasers of different wavelengths, a digital microlens array and an optical system of lens combination;

[0027] (3) Data processor: The three-dimensional flow field velocity vector field is reconstructed by the data processor based on the obtained temporal particle light field image.

[0028] Furthermore, it also includes a 45-degree reflector, which is placed above and facing the light field, while one side is tilted upwards at a 45-degree angle.

[0029] Furthermore, the three lasers emit wavelengths of 405nm, 532nm, and 650nm, respectively.

[0030] Beneficial effects: Compared with the prior art, the present invention has the following significant advantages:

[0031] (1) The measurement requires few devices, and only a single light field camera is needed to achieve three-dimensional super-resolution fast imaging;

[0032] (2) High imaging spatial resolution: First, the spatial resolution in the Z direction is improved by using dual-view light field imaging method, and then super-resolution imaging is achieved by using structured light super-resolution imaging.

[0033] (3) Fewer images need to be acquired. The innovative use of structured light color coding method reduces the number of images to be acquired from the original nine images in three directions and three phases to one image in a single direction containing three phases, significantly improving the imaging speed and enabling high-precision and rapid measurement of three-dimensional flow fields. Attached Figure Description

[0034] Figure 1 This is a schematic diagram of the process of this invention;

[0035] Figure 2 This is a diagram of the experimental apparatus for this invention;

[0036] Figure 3 This is a schematic diagram illustrating the principle of structured light color encoding in this invention;

[0037] Figure 4 This is a schematic diagram illustrating the principle of three-dimensional flow field reconstruction in this invention. Detailed Implementation

[0038] like Figure 1 As shown in the figure, the specific steps of the flow field measurement method based on structured light coding and dual-view optical field imaging in this embodiment are as follows:

[0039] Step 1: Calibrate the resolution and point spread function of the imaging system to generate three orthogonal structured light stripes with different phases.

[0040] Step 2: Construct the excitation optical path. Three-color lasers with wavelengths of 405nm, 532nm, and 650nm emitted by three lasers are converged through the optical path and illuminated on the surface of the modulated digital micromirror array using a mirror. Collimated structured light is generated by the stripes loaded by the modulated digital micromirror array. The zero-order stripe is selected using a spatial filter and then amplified by a lens combination to excite and illuminate the flow field space. The trigger signal between the modulated digital micromirror array and the three lasers is modulated to generate three orthogonal structured lights of different colors with a phase difference of 120° in the imaging area.

[0041] Step 3: Adjust the relationship between the camera exposure time and the digital micromirror array to acquire structured light fringes of three phases within a single camera exposure time.

[0042] Step 4: Construct the experimental optical path for dual-view imaging of the optical field. Scatter tracer particles in the flow field to be tested. Use the optical path excitation system to illuminate the test flow field. Use the optical field camera to image the flow field to be tested in the frontal view and the top view reflected by the 45-degree mirror to obtain the optical field image of the tracer particles illuminated by structured light of three phases. Reconstruct the optical field from the imaging results.

[0043] Step 5: Obtain spatial information of structured light fringes at different depths from two viewpoints—the frontal view and the top view (reflected by a 45-degree specular surface)—from the reconstructed light field image. Extract the frequency and phase information of the structured light. Based on the frequency and phase of the structured light, perform structured light super-resolution reconstruction on the light field images from both viewpoints to obtain a 3D frontal view of the particle super-resolution image and a 3D top view of the particle super-resolution image. The specific steps for structured light super-resolution reconstruction are as follows:

[0044] Wiener filtering is applied to the image to remove background noise and improve the signal-to-noise ratio of the reconstructed image. Then, a deconvolution algorithm is used to separate the frequency components. and The centers are offset to their correct positions in the frequency domain space (+p) θ and -p θ This method collects twice the amount of information, enabling super-resolution reconstruction of refocused images.

[0045] Step 6: Rotate the super-resolution 3D image obtained from the top-view direction with 45-degree mirror reflection by 90 degrees, overlap the center of the rotated image with the center of the front view image, and fuse and superimpose the two images to obtain a super-resolution 3D particle distribution image.

[0046] Step 7: Perform cross-correlation calculations on the super-resolution 3D particle distribution image to obtain the 3D velocity vector field of the flow field. Divide the 3D particle distribution image into multiple sampling volumes of uniform size. By performing Fourier transform multiplication on the corresponding sampling volumes of two consecutive voxel matrices, and then performing an inverse Fourier transform on the product in the frequency domain, the cross-correlation function of the sampling region can be obtained. Further, the 3D velocity vector field of the entire flow field can be calculated. Then, post-process the 3D velocity vector field of the flow field to remove erroneous velocity vectors and insert the correct velocity vectors based on adjacent velocity vectors to obtain a high-precision 3D flow field.

[0047] This embodiment also provides a measurement system for a flow field measurement method based on structured light coding and dual-view optical field imaging, including:

[0048] (1) Light field camera - an industrial camera including macro lens, microlens array and main lens;

[0049] (2) Structured light illumination and dual-view light field imaging system, the illumination of which includes three lasers of different wavelengths, a digital microlens array and an optical system of lens combination;

[0050] (3) Data processor: The three-dimensional flow field velocity vector field is reconstructed by the data processor based on the obtained temporal particle light field image.

[0051] The measurement system also includes a 45-degree reflector, which is placed above and facing the light field, with one side tilted upward at a 45-degree angle.

Claims

1. A flow field measurement method based on structured light coding and dual-view optical field imaging, characterized in that, Includes the following steps: Step 1: Calibrate the resolution and point spread function of the imaging system, and generate three orthogonal structured light fringes with different phases; Step 2: Construct the excitation optical path, modulate the trigger signal between the digital micromirror array and the three lasers, and sequentially generate three orthogonal structured lights of different colors with a phase difference of 120° in the imaging area; Step 3: Adjust the relationship between the camera exposure time and the digital micromirror array to acquire structured light fringes of three phases within a single camera exposure time; Step 4: Construct the experimental optical path for dual-view imaging of the light field. Use a light field camera to image the flow field under test in the frontal view and the top view reflected by a 45-degree mirror. Reconstruct the light field from the imaging results under structured light illumination of three different phases. Step 5: Obtain the spatial information of structured light fringes at different depths from two perspectives: the frontal view and the top view reflected by a 45-degree mirror from the light field reconstruction pattern. Extract the frequency and phase information of the structured light. Based on the frequency and phase of the structured light, perform structured light super-resolution reconstruction on the light field images under the two perspectives to obtain the three-dimensional frontal view image of the particle super-resolution image and the three-dimensional top view image of the particle super-resolution image. Step 6: Superimpose the three-dimensional front view and the three-dimensional top view of the particle super-resolution image to obtain a three-dimensional particle distribution image with super-resolution. Step 7: Perform cross-correlation calculation on the three-dimensional particle distribution image to obtain the three-dimensional velocity vector field of the flow field to be measured, and perform post-processing on the three-dimensional velocity vector field of the flow field.

2. The measurement method according to claim 1, characterized in that, The specific steps for constructing the excitation optical path in step 2 are as follows: the three-color lasers emitted by the three lasers are converged through the optical path, and the surface of the modulated digital micromirror array is illuminated by a reflector. The structured light is generated by the stripes loaded by the modulated digital micromirror array, the zero-order stripe is selected by a spatial filter, and then the light is amplified by a lens combination to excite and illuminate the flow field space.

3. The measurement method according to claim 2, characterized in that, The wavelengths of the three-color lasers are 405nm, 532nm, and 650nm, respectively.

4. The measurement method according to claim 1, characterized in that, The specific operation of structured light super-resolution reconstruction described in step 5 is as follows: Wiener filtering is applied to the image to remove background noise in the obtained image, improve the signal-to-noise ratio of the reconstructed image, and then the center of the frequency components is shifted to their correct positions in the frequency domain space by a deconvolution algorithm, thereby collecting twice the amount of information and realizing super-resolution reconstruction of the refocused image.

5. The measurement method according to claim 1, characterized in that, The super-resolution particle 3D distribution image mentioned in step 6 refers to the superposition of dual-view particle 3D images to eliminate the axial stretching effect of single-view particle 3D images, thereby achieving super-resolution reconstruction of the position and shape of 3D particles and obtaining the 3D flow field velocity.

6. The measurement method according to claim 1, characterized in that, The specific steps for cross-correlation calculation in step 7 are as follows: the three-dimensional distribution image of particles is divided into multiple sampling volumes of the same size. By performing Fourier transform multiplication on the corresponding sampling volumes of two consecutive voxel matrices, and performing inverse Fourier transform on the product in the frequency domain, the cross-correlation function of the sampling region can be obtained. The three-dimensional velocity vector field of the entire flow field can then be calculated.

7. The measurement method according to claim 1, characterized in that, The specific steps for post-processing the three-dimensional velocity vector field of the flow field in step 7 are as follows: remove erroneous velocity vectors from the velocity vector field and insert the correct velocity vectors based on adjacent velocity vectors.

8. A system for measuring a flow field using the flow field measurement method based on structured light coding and dual-view optical field imaging as described in any one of claims 1 to 7, characterized in that, include: (1) Light field camera – an industrial camera including macro lenses, microlens arrays, and a main lens; (2) Structured light illumination and dual-view light field imaging system, the illumination of which includes three lasers of different wavelengths, a digital microlens array and an optical system of lens combination; (3) Data processor: The three-dimensional flow field velocity vector field is reconstructed by the data processor based on the obtained temporal particle light field image.

9. The system according to claim 8, characterized in that, It also includes a 45-degree reflector, which is placed above and facing the light field, with one side tilted upward at a 45-degree angle.

10. The system according to claim 9, characterized in that, The three lasers emitted wavelengths of 405nm, 532nm, and 650nm, respectively.

Images