A large batch of reflective plane surface type automatic measurement method

Abstract

The application provides a large-batch reflective plane surface automatic measurement method, and relates to the technical field of surface measurement, which comprises the following steps: constructing a binocular phase deflection measurement system, collecting multiple phase shift fringe images of a plurality of to-be-measured parts based on the binocular phase deflection measurement system; respectively performing phase calculation on the collected fringe images to determine the position corresponding relationship between the camera pixel points and the screen pixel points of the two cameras; extracting the corresponding effective areas of the plurality of to-be-measured parts according to the fringe change conditions of the fringe images collected by the two cameras; establishing the corresponding relationship between the effective areas and the to-be-measured parts in the two cameras according to the constraints of the polar lines and normal vectors, fitting the predetermined positioning plane of the to-be-measured parts based on binocular vision; performing calculation based on the predetermined positioning plane to obtain the normal vector direction of the surface of the to-be-measured part, and integrating and calculating the surface shape according to the normal vector direction of the surface of the to-be-measured part. The application effectively solves the demand of large-batch detection of the surface shape of the reflective plane device.

Description

Technical Field

[0001] This invention relates to the field of surface shape measurement technology, and specifically to an automatic method for mass measurement of reflective flat surface shapes. Background Technology

[0002] Flatness is a crucial indicator in the industrial field, directly related to the performance of many industrial components. For example, the flatness of a sealing end face directly affects the sealing effect of a part. The flatness of the contact surface between the substrate and the heat sink directly affects its heat dissipation efficiency. The flatness of the flange face directly affects the fit and preload between the two flanges. Therefore, flatness has a significant impact on the industrial application of components, making flatness measurement essential during the manufacturing and quality inspection processes. Furthermore, with the increasing speed of industrial production, higher demands are placed on measurement efficiency. The ability to quickly and simultaneously inspect multiple planes will significantly improve inspection efficiency, making it a research area of ​​great importance.

[0003] Currently, common methods for flatness inspection include profilometry and interferometry. Profilometry uses a point-to-point measurement method, which, while unaffected by the material of the workpiece and capable of measuring steep surfaces, is slow and costly due to its point-by-point measurement nature. Interferometry offers high measurement accuracy and is commonly used for measuring standard parts. However, interferometric measurement systems are expensive, require precise leveling of the workpiece, and are sensitive to environmental factors such as vibration, hindering improved measurement efficiency and wider application. With the development of related measurement technologies, phase deflection measurement methods have been proposed and are gradually being applied to the measurement of reflective planes. This method uses a surface measurement approach, offering a large measurement field of view, high flatness measurement range and accuracy, and eliminates the need for precise adjustment of the surface being measured.

[0004] Many researchers have conducted in-depth studies on the challenges of measurement efficiency, accuracy, and discrete surface measurement in deflection measurement systems. Single-camera measurement methods based on a reference plane can be used to measure near-planar surfaces, offering fast reconstruction speeds and high system integrability. However, this method requires pre-determining the reference plane position, limiting the application scenarios of the measurement system. Some literature proposes binocular measurement systems that determine the normal vector of the measured surface by matching the normal vectors calculated by the two cameras, achieving high measurement accuracy. However, the matching and search process between the two cameras is very time-consuming, hindering the improvement of detection efficiency. The above methods, because they require first calculating the normal vector direction of the surface to be measured and then performing integration to solve the surface shape, cannot directly achieve the measurement of batches of discrete surfaces. Some literature proposes a method that directly calculates depth from the phase, eliminating the need for normal vector solving; however, this measurement system requires two screens and a beam splitter, and the measurement accuracy is not high. Some literature describes binocular deflection systems that propose performing slope integration in each continuous curved surface region to obtain higher measurement accuracy, and then calculating the absolute position of the height reference point of each continuous curved surface to evaluate the relative positions between different curved surfaces. However, the study did not address the discrete case of multiple test pieces, nor did it solve the problem of the algorithm complexity of binocular matching.

[0005] In summary, although phase deflection is suitable for measuring the flatness of reflective surfaces, there is still a lack of mature solutions for rapid detection of large numbers of reflective surfaces. Summary of the Invention

[0006] This invention aims to at least partially solve one of the technical problems in the prior art. Therefore, the object of this invention is to provide an automatic method for mass production of reflective planar surface profiles, effectively addressing the need for mass production inspection of reflective planar devices.

[0007] Technical solution: This invention provides an automatic measurement method for large-scale reflective planar surface profiles, comprising:

[0008] A binocular phase deflection measurement system is constructed, and multiple phase shift fringe images are simultaneously acquired for several test objects based on the binocular phase deflection measurement system.

[0009] Phase calculations were performed on the acquired stripe images to determine the positional correspondence between the camera pixels of the two cameras and the screen pixels.

[0010] Based on the stripe changes in the stripe images acquired by the two cameras, the corresponding effective areas of several test pieces are extracted.

[0011] Based on the constraints of epipolar lines and normal vectors, the correspondence between the effective areas of the two cameras and the test object is established, and the pre-positioning plane of the test object is fitted based on binocular vision.

[0012] The normal vector direction of the surface of the test piece is obtained by solving based on the pre-positioned plane, and the surface shape is solved by integrating based on the normal vector direction of the surface of the test piece.

[0013] Furthermore, establishing the correspondence between the effective regions in the two cameras and the test piece based on the constraints of the epipolar line and the normal vector includes: selecting a pixel on the contour of any effective region in the fringe image captured by camera one of the binocular cameras as the selected point; drawing a corresponding epipolar line in the fringe image captured by camera two; the epipolar line and the contour of the effective region in the fringe image captured by camera two form several intersection points; calculating the surface normal vector at each intersection point; selecting the intersection point with the smallest deviation in the normal vector direction from the selected point as the corresponding point; and the effective region corresponding to the contours where the selected point and the corresponding point are located is the surface region of the corresponding test piece.

[0014] Furthermore, after establishing the correspondence between the effective areas of the two cameras and the test object based on the constraints of the epipolar line and the normal vector, the method further includes: selecting the neighborhood range of the corresponding epipolar line for the contour of each effective area, searching for the point with the smallest difference in the normal vector within the neighborhood range and defining it as the contour fine positioning point, and forming a fine positioning contour from several contour fine positioning points.

[0015] Furthermore, the pre-positioning plane of the test piece based on binocular vision fitting includes: calculating the coordinates of the actual contour of the test piece in the camera coordinate system based on the fine positioning contour calculated in camera one and camera two of the binocular camera, and generating the pre-positioning plane of the test piece by performing plane fitting based on the coordinates of the actual contour of the test piece in the camera coordinate system.

[0016] Furthermore, the binocular phase deflection measurement system includes:

[0017] A display screen is used to display phase-shifted fringe images;

[0018] A fixture is used to hold several test pieces simultaneously.

[0019] A binocular camera is used to collect reflected light from several test objects, wherein the fixture is located within the common field of view of the binocular camera.

[0020] Furthermore, the step of performing phase calculation on the acquired stripe images to determine the positional correspondence between the camera pixels and the screen pixels of the two cameras includes: encoding the screen pixels using sinusoidal stripes, employing a four-step phase-shift dephase calculation and a multi-frequency method to unwrap the stripes, determining the stripe phase value corresponding to each pixel in the camera, and thus determining the position of the screen pixel corresponding to the camera pixel.

[0021] Furthermore, the step of extracting the corresponding effective regions of several test objects based on the stripe changes in the stripe images acquired by the two cameras includes: performing grayscale processing on the stripe images acquired by the two cameras respectively, and then extracting the effective regions based on the changes in regional connectivity and contrast.

[0022] Beneficial effects: The measurement method of the present invention improves upon the traditional phase deflection measurement method, solving the problem that it cannot achieve discrete measurement of the test surface. It realizes the functions of rapid synchronous measurement of multiple planes and automatic surface shape calculation, effectively solving the need for large-scale detection of the surface shape of reflective planar devices. Attached Figure Description

[0023] Figure 1 A flowchart illustrating an automatic measurement method for large-scale reflective planar surface profiles provided by this invention;

[0024] Figure 2 This is a schematic diagram of the binocular phase deflection measurement system in a method for automatically measuring the shape of reflective planes in large batches, provided by the present invention.

[0025] Figure 3 This is a schematic diagram of the fixture in the binocular phase deflection measurement system of the automatic measurement method for large-scale reflective planar surface shapes provided by the present invention;

[0026] Figure 4 A flowchart illustrating an automatic measurement method for large-scale reflective planar surface profiles provided by this invention;

[0027] Figure 5 This is a schematic diagram of the contour positioning structure in a method for automatic measurement of reflective planar surface shapes in large batches, provided by the present invention.

[0028] Figure 6 This is a schematic diagram of the normal vector calculation based on a pre-positioned plane in an automatic measurement method for large-scale reflective planar surface shapes provided by the present invention.

[0029] In the diagram: 1. Display screen; 2. Camera 1; 3. Camera 2; 4. Housing; 5. Adapter plate; 6. Fixture; 7. Precision positioning search area; 8. Outer rough extraction contour of the part under test; 9. Epipolar line; 10. Pre-positioning plane; 11. Schematic diagram of pixel extraction from camera image. Detailed Implementation

[0030] The specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0031] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, 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.

[0032] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0033] like Figure 1 As shown, the first embodiment of the present invention provides an automatic measurement method for the surface shape of reflective planes in large quantities, comprising:

[0034] S1. Construct a binocular phase deflection measurement system, and simultaneously acquire multiple phase shift fringe images of several test pieces based on the binocular phase deflection measurement system;

[0035] The binocular phase deflection measurement system includes:

[0036] A display screen is used to display phase-shifted fringe images;

[0037] A fixture is used to hold several test pieces simultaneously.

[0038] A binocular camera is used to collect reflected light from several test objects, wherein the fixture is located within the common field of view of the binocular camera.

[0039] S2. Perform phase calculation on the acquired stripe images to determine the positional correspondence between the camera pixels of the two cameras and the screen pixels.

[0040] Specifically, the step of performing phase calculation on the acquired stripe images to determine the positional correspondence between camera pixels and screen pixels of the two cameras includes: encoding screen pixels using sinusoidal stripes, employing a four-step phase-shift dephase method and a multi-frequency method to dewrap the image, determining the stripe phase value corresponding to each pixel in the camera, and thus determining the position of the screen pixel corresponding to the camera pixel.

[0041] S3. Based on the stripe changes in the stripe images acquired by the two cameras, extract the corresponding effective areas of several test pieces to generate the outer coarse extraction contour of the test piece 8.

[0042] Specifically, the step of extracting the corresponding effective regions of several test objects based on the stripe changes in the stripe images acquired by the two cameras includes: performing grayscale processing on the stripe images acquired by the two cameras respectively, and then extracting the effective regions based on the changes in regional connectivity and contrast.

[0043] S4. Based on the constraints of epipolar lines and normal vectors, establish the correspondence between the effective areas of the two cameras and the test piece, and fit the pre-positioning plane of the test piece based on binocular vision.

[0044] The process of establishing the correspondence between the effective regions in the two cameras and the test piece based on the constraints of the epipolar line and the normal vector includes: selecting a pixel on the contour of any effective region in the fringe image captured by camera one of the binocular cameras as the selected point; drawing a corresponding epipolar line in the fringe image captured by camera two; the epipolar line and the contour of the effective region in the fringe image captured by camera two form several intersection points; calculating the surface normal vector at each intersection point; selecting the intersection point with the smallest deviation in the normal vector direction from the selected point as the corresponding point; and the effective region corresponding to the contours of the selected point and the corresponding point is the surface region of the corresponding test piece.

[0045] The pre-positioning plane of the test piece based on binocular vision fitting includes: calculating the coordinates of the actual contour of the test piece in the camera coordinate system based on the fine positioning contour calculated by camera one and camera two of the binocular camera, and generating the pre-positioning plane of the test piece by performing plane fitting based on the coordinates of the actual contour of the test piece in the camera coordinate system.

[0046] S5. Solve based on the pre-positioned plane to obtain the normal vector direction of the surface of the part under test, and solve the surface shape by integrating based on the normal vector direction of the surface of the part under test.

[0047] Furthermore, after establishing the correspondence between the effective areas of the two cameras and the test object based on the constraints of the epipolar line and the normal vector, the method further includes: selecting the neighborhood range of the corresponding epipolar line for the contour of each effective area, searching for the point with the smallest difference in the normal vector within the neighborhood range and defining it as the contour fine positioning point, and forming a fine positioning contour from several contour fine positioning points.

[0048] Beneficial effects: The measurement method of the present invention improves upon the traditional phase deflection measurement method, solving the problem that it cannot achieve discrete measurement of the test surface. It realizes the functions of rapid synchronous measurement of multiple planes and automatic surface shape calculation, effectively solving the need for large-scale detection of the surface shape of reflective planar devices.

[0049] Specific implementation: The specific implementation scheme adopted in this invention is as follows, including the following steps:

[0050] 1. Construction of a Binocular Phase Deflection Measurement System: Select appropriate cameras, lenses, and screens based on the required field of view and accuracy. Design the relative poses of each component rationally according to the actual situation, ensuring system compactness while meeting measurement requirements. Simultaneously, an integrated system design is necessary to achieve system integration. The field of view stems from specific measurement needs. It must be ensured that the common field of view of the cameras covers the device under test (DUT), and that the surface of the DUT reflects light from the screen to the camera. This requires the screen area to be at least twice the area of ​​the DUT surface. While ensuring that the common field of view of the two cameras covers the entire measurement area, and that the displayed stripes on the screen are reflected from all parts of the DUT surface and captured by the camera, the camera should be positioned as close to the edge of the screen as possible.

[0051] 2. Stripe Image Acquisition: Multiple test pieces (simultaneously clamped by fixtures) are placed within the measurable area so that both cameras can image them and capture their reflected images on the screen. Then, using custom projection software, phase-shifted stripes are displayed on the screen, and the cameras are controlled to acquire images one by one, resulting in a series of images to be processed.

[0052] 3. Automated Surface Separation: Since patterns from multiple test surfaces are acquired simultaneously, it is necessary to separate the acquired surfaces to obtain each surface pattern individually. This invention proposes a matching method based on contrast characteristics and epipolar and normal vector constraints, solving the problem of difficult matching caused by surface separation and high similarity.

[0053] 4. Fast surface shape calculation: In order to solve the problem of low efficiency of existing algorithms and meet the requirements of high-efficiency measurement and calculation, this invention proposes a surface shape reconstruction method based on a pre-positioned plane, which eliminates the need for time-consuming processes such as matching and searching, and realizes fast surface shape calculation of the surface to be measured.

[0054] 5. Processing and Evaluation of Measurement Results for the Surface Under Test: Through software programming, the measurement results for each surface are automatically leveled and analyzed, enabling batch detection of flat surfaces and subsequent evaluation as needed. Ultimately, this achieves simultaneous detection and automatic analysis of batch flat surfaces.

[0055] The present invention has the following advantages due to the adoption of the above technical solutions:

[0056] 1. Based on the concept of this invention, the need for mass testing of reflective planar devices can be effectively addressed. At the same time, multiple test pieces can be measured without a complicated leveling process, which significantly improves the efficiency of measurement.

[0057] 2. Based on the algorithm proposed in this invention, the surface shape separation and matching of multiple reflective planar devices can be realized, as well as the rapid surface shape calculation. Ultimately, it can automatically perform surface shape separation and rapid calculation for each test piece, meeting the speed requirements of modern manufacturing development in the field of measurement.

[0058] 3. By calculating the precise positioning profile, this measurement system can achieve sub-micron measurement accuracy for surface shapes, ensuring both high efficiency and high measurement precision and analytical accuracy.

[0059] Ultimately, this system can achieve batch testing and automatic evaluation functions according to actual needs. The entire system has significant advantages in terms of high efficiency and high automation.

[0060] Figure 2 This is a schematic diagram of the binocular phase deflection measurement system of the present invention. Figure 3 This is a schematic diagram of the fixture of the binocular phase deflection measurement system of the present invention. This measurement system can be used for the measurement of a large number of reflective planes.

[0061] The binocular phase deflection measurement system includes:

[0062] Display screen 1 is used to display a phase-shifted fringe image;

[0063] Fixture 6 is used to simultaneously clamp several test pieces;

[0064] A binocular camera, including camera 2 and camera 3, is used to collect reflected light from several test objects, wherein the fixture 6 is located within the common field of view of the binocular camera.

[0065] The display screen 1 and the binocular camera are both mounted on the housing 4, which is fixed to the mounting surface by an adapter plate 5.

[0066] The entire system is controlled by a host computer. Display screen 1 shows stripes of different frequencies, and images are acquired by a binocular camera. The acquired data is then processed and analyzed by the host computer. By fixing the workpiece under test within the field of view of the measurement system using a custom fixture 6, and ensuring that the camera can capture the screen stripes reflected from its surface, the host computer can control the entire process of displaying stripes, acquiring images, calculating surface features, and analyzing surface features.

[0067] The image-based analysis algorithm is another core component of this invention, and the specific process is as follows: Figure 4 As shown, the specific process and method are as follows:

[0068] (1) Perform phase calculation on the acquired stripe image to determine the position of the screen pixel corresponding to the camera pixel for subsequent calculation.

[0069] (2) Specifically, the screen pixels are encoded by sinusoidal stripes, and the four-step phase shift dephase and multi-frequency de-wrapping method are used to determine the stripe phase value corresponding to each pixel of the camera, and then the screen pixel position corresponding to the camera pixel is determined.

[0070] (3) Extract the effective region (the region where the stripe changes) based on the stripe changes in the image captured by the camera, and segment the region based on its connectivity (the extracted region consists of multiple discretely distributed test objects, and the image of each test surface is an independent connected whole; the process of extracting and segmenting these wholes separately is called connected component segmentation). Since the screen displays stripes using the phase-shifting method, there are significant contrast changes in different images within the effective stripe region; the effective region can be extracted and segmented based on the contrast.

[0071] (4) Based on the edge of the segmented shape and the constraint information of the epipolar line 9 and the normal vector, corresponding surface matching is achieved to obtain the test surfaces corresponding to the two cameras. Since the reflective test surfaces are similar and have few surface features, in order to solve this problem, the method used in this invention achieves the matching of the test surfaces based on the constraint of the epipolar line 9 and the normal vector. Specifically, any point on the image contour of camera 12 will have several intersection points with the image contour of camera 2 on the corresponding epipolar line 9. Each intersection point can calculate a surface normal vector. The point with the smallest deviation from the normal vector direction calculated by camera 12 is the corresponding point, and the corresponding contour and region are the matching regions.

[0072] (5) After matching is completed, for each face shape to be solved, such as Figure 5 As shown, the contour extraction accuracy of the surface is further improved by searching within the neighborhood of epipolar line 9. To ensure the measurement accuracy of this invention, the contour position of the surface to be measured needs to be accurately located. Therefore, the contour positioning accuracy is improved by using neighborhood search to ensure that the measurement accuracy of the measurement system meets the requirements. Specifically, after determining the coarsely extracted contour 8 of the part to be measured selected by camera 2 (3) and camera 1 (2), in order to improve the contour search accuracy, it is necessary to search for corresponding points on both sides of the point determined by camera 2 (3) along the direction of epipolar line 9. In this embodiment, the search range extending on both sides is three pixels. The point with the smallest difference in normal vector within the search area is the fine contour positioning point.

[0073] (6) Figure 6 As shown, the spatial position of the surface to be measured is located based on a pre-defined contour. A solution method based on planar pre-positioning is used to calculate the intersection points of the rays corresponding to the effective pixels of the camera and the pre-positioning plane 10. This further calculates the normal direction of the surface to be measured and integrates to solve the surface shape. The stripe image acquired by camera 2 is processed by an image analysis algorithm to generate a pixel extraction diagram 11. Existing methods have low reconstruction efficiency. To address this issue, this invention replaces the matching search process with pre-positioning, which can significantly improve the efficiency of surface shape calculation.

[0074] Specifically, after achieving precise contour localization, the coordinates of the actual contour in the camera coordinate system can be calculated using the corresponding points of the contour in the two cameras. Further plane fitting based on the contour allows for pre-positioning of the surface under test in the camera coordinate system. Based on the pre-positioned plane position, the intersection of the camera ray and the fitted plane, and the corresponding point of the ray emitted from the screen calculated in step S2, the incident and reflected rays passing through the surface under test can be calculated, further enabling the calculation of the surface normal direction. Finally, integration completes the surface shape calculation. This invention replaces the matching search process with pre-positioning, significantly improving the efficiency of surface shape calculation.

[0075] (7) Finally, this measurement system can also quickly and automatically analyze and process the solved surface profile. This includes leveling the surface to be measured, calculating peak and valley values ​​and root mean square (RMS) values, and determining whether a product is qualified based on threshold parameters. For example, after solving for the surface profile, the position of the surface profile in the camera coordinate system is obtained. The coordinates of the plane equation can be calculated through plane fitting. Furthermore, the surface to be measured can be made almost parallel to the XOY surface through plane rotation. Then, based on the difference between the maximum and minimum values ​​of the surface to be measured, its peak and valley values ​​can be calculated. Other operations such as RMS and qualified product judgment can be further optimized by adding corresponding algorithms as needed.

[0076] In summary, this invention designs and builds a measurement system capable of simultaneously detecting a batch of reflective surfaces, and also designs supporting algorithms for surface shape separation, surface shape calculation, and automatic evaluation. Ultimately, it can obtain the surface shape measurement results for each surface under test, as well as information such as peak-valley values ​​and root mean square values, achieving the function of simultaneous batch detection and automatic evaluation analysis.

[0077] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media containing computer-usable program code. These computer program instructions can also be stored in a computer-readable storage medium capable of directing a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means implemented in a process. Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0078] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

Claims

1. A method for automatically measuring the surface shape of reflective planes in large quantities, characterized in that, include: A binocular phase deflection measurement system is constructed, and based on the binocular phase deflection measurement system, multiple phase shift fringe images are simultaneously acquired for several test objects. The binocular phase deflection measurement system includes: A display screen is used to display phase-shifted fringe images; A fixture is used to hold several test pieces simultaneously. A binocular camera is used to collect reflected light from several test objects, wherein the fixture is located within the common field of view of the binocular camera; Phase calculations were performed on the acquired stripe images to determine the positional correspondence between the camera pixels of the two cameras and the screen pixels. Based on the stripe changes in the stripe images acquired by the two cameras, the corresponding effective areas of several test pieces are extracted. Based on the constraints of epipolar lines and normal vectors, the correspondence between the effective areas in the two cameras and the test piece is established, including: selecting a pixel on the contour of any effective area in the fringe image captured by camera one of the binocular cameras as the selected point; drawing a corresponding epipolar line in the fringe image captured by camera two; the epipolar line and the contour of the effective area in the fringe image captured by camera two form several intersection points; calculating the surface normal vector at each intersection point; selecting the intersection point with the smallest deviation in the normal vector direction from the selected point as the corresponding point; and the effective area corresponding to the contours where the selected point and the corresponding point are located is the surface area of ​​the corresponding test piece. For each effective region, a neighborhood range of the corresponding epipolar line is selected. Within the neighborhood range, the point with the smallest difference in normal vectors is selected as the fine positioning point of the contour. Several fine positioning points of the contour form a fine positioning contour. The pre-positioning plane of the test piece is fitted based on binocular vision, including: calculating the coordinates of the actual contour of the test piece in the camera coordinate system based on the fine positioning contour calculated in camera one and camera two of the binocular camera, and performing plane fitting based on the coordinates of the actual contour of the test piece in the camera coordinate system to generate the pre-positioning plane of the test piece. The normal vector direction of the surface of the test piece is obtained by solving based on the pre-positioned plane, and the surface shape is solved by integrating based on the normal vector direction of the surface of the test piece.

2. The method according to claim 1, wherein, The step of performing phase calculation on the acquired stripe images to determine the positional correspondence between camera pixels and screen pixels includes: encoding screen pixels using sinusoidal stripes, employing a four-step phase-shift dephase calculation and a multi-frequency de-wrapping method to determine the stripe phase value corresponding to each pixel in the camera, and then determining the position of the screen pixel corresponding to the camera pixel.

3. The method of claim 2, wherein the method is used for measuring a large number of flat reflective surfaces. The step of extracting the corresponding effective regions of several test objects based on the stripe changes in the stripe images acquired by the two cameras includes: performing grayscale processing on the stripe images acquired by the two cameras respectively, and then extracting the effective regions based on the changes in regional connectivity and contrast.

Images