A monocular microscopic imaging surface micro-nano defect 3D reconstruction detection device and method

Abstract

This invention discloses a monocular microscopic imaging device and method for 3D reconstruction detection of micro- and nano-scale surface defects. The detection device includes an illumination imaging module, a displacement module, and a laser ruler ranging system. The displacement module comprises, from bottom to top, a series of sequentially stacked axial guide rails, a spin axis, another axial guide rail, another axial guide rail, a stage, and an axial guide rail positioned above the stage. The axial guide rails are mounted on a swing axis. The illumination imaging module is mounted on the axial guide rails and includes an electrically operated zoom imaging module and several stroboscopic light sources uniformly arranged in a ring around it. The illumination angle and aperture angle of each stroboscopic light source are electrically adjustable. The laser ruler ranging system measures the actual movement distance of the axial guide rails. Using this invention, micro- and nano-scale defect detection and 3D reconstruction can be achieved using only monocular vision, while ensuring detection efficiency and accuracy.

Description

Technical Field

[0001] This invention belongs to the field of optical detection of surface defects in materials, and in particular relates to a device and method for 3D reconstruction detection of micro- and nano-defects on a surface using monocular microscopy. Background Technology

[0002] On certain ultra-smooth material surfaces and polished metal surfaces, there are stringent requirements for the size detection of defects, necessitating the detection of defects at the sub-micron or even nanometer scale. The presence of surface defects can cause damage such as energy scattering and stress concentration inside the equipment, shortening its lifespan. Therefore, timely detection and quantitative evaluation of these micro- and nano-sized defects are crucial.

[0003] Optical inspection methods are widely used in surface defect detection due to their non-contact, digital evaluation algorithms, and automated inspection features. Typical optical performance defect detection methods are represented by vision inspection systems with cameras as the core tool, which are divided into monocular (single camera) and multi-view (multiple cameras) vision inspection based on the number of cameras.

[0004] Currently, traditional monocular vision inspection systems for surface defects can only identify lateral dimensions such as width and length when quantitatively evaluating defect size, but cannot obtain longitudinal dimension information such as defect depth.

[0005] Multi-view vision detection methods are based on the triangulation principle of acquiring the same target image from multiple perspectives (similar to the principle of binocular parallax in human eyes). They can calculate the spatial three-dimensional model of the target object containing depth information and have been applied to the extraction of spatial three-dimensional information in fields such as robot vision, autonomous driving, and motion capture.

[0006] For example, Chinese patent document CN111551559A discloses a defect detection method for LCD screens based on a multi-view vision system. The system uses multiple cameras to form a multi-view vision system, which acquires 2D images from different angles to achieve multi-angle image comparison and analysis, and restores 3D information such as depth information of the detected object.

[0007] Chinese patent document CN120339413A discloses a method and system for capturing moving targets based on AI multi-view vision, including: calibrating the camera parameters of each vision system; extracting 2D feature points of the moving target through each calibrated vision system; reconstructing each 2D feature point according to the calibrated camera parameters, combining a multi-view geometric reconstruction algorithm and the least squares method to obtain 3D feature points; adding constraints to the 3D feature points; reconstructing the constrained 3D feature points through a multi-view geometric analytical algorithm to obtain the three-dimensional motion information of the moving target; and outputting the three-dimensional motion information of the moving target to complete the capture of the moving target.

[0008] However, multi-view vision systems have stringent requirements for camera configuration, necessitating the use of multiple cameras located at specific spatial positions. This requires extremely rigorous and complex methods for calibrating system intrinsic and extrinsic parameters, along with spatial coordinate demodulation algorithms, to obtain the spatial 3D information of the object under test. The highest lateral resolution is at the micrometer level, and the vertical resolution is at the 10-micrometer level, making it difficult to apply to 3D reconstruction of surface micro- and nano-defects. Therefore, it is essential to develop 3D reconstruction visual inspection systems specifically for surface micro- and nano-defects. Summary of the Invention

[0009] This invention provides a device and method for detecting and reconstructing micro- and nano-defects on a surface using monocular microscopy. Under the premise of ensuring detection efficiency and accuracy, it can detect micro- and nano-defects and achieve 3D reconstruction using only monocular vision.

[0010] A monocular microscopic imaging surface micro-nano defect 3D reconstruction detection device includes an illumination imaging module, a displacement module and a laser ruler ranging system; The displacement modules include those arranged sequentially from bottom to top. Shaft guide rail, Shaft guide rail, Spin axis, Shaft guide rail, Shaft guide rail, stage, and the components mounted on top of the stage Shaft guide rail; the Shaft guide rail mounted on On the swing axis; The illumination imaging module is installed in On the axial guide rail, there is a motorized zoom imaging module and components evenly arranged around it in a ring. Individual stroboscopic sources; illumination angle of each stroboscopic source Illumination aperture angle Electronically adjustable, under given parameters , Each strobe light source and the motorized zoom imaging module together form one light field, which can form a maximum of [number] light fields at any given time. A light field; The laser ruler distance measuring system is used for measurement The actual travel distance of the axis guide rail.

[0011] Furthermore, the laser ruler ranging system includes a laser, a beam splitter, a reflector, a reference mirror, and an interferometer. Among them, the laser is installed at The swing axis, beam splitter, reflector, and reference mirror are mounted on... The axis guide rail and interferometer are mounted on the illumination imaging module.

[0012] Furthermore, the number of strobe sources It is an odd number greater than 1.

[0013] A method for 3D reconstruction and detection of surface micro-nano defects using monocular microscopy, based on the aforementioned monocular microscopy surface micro-nano defect 3D reconstruction and detection device, includes the following steps: Step 1, the illumination imaging module at low imaging magnification Next, using Manually focus the axis guide rail to find the highest point of the surface profile of the defect to be measured. focal plane and lowest position The focal plane; Step 2: Accurately locate the highest and lowest positions of the defect surface contour using a sharpness evaluation algorithm, and record the corresponding positions as follows: , ; Step 3, along to Range, continuous acquisition Images at different heights, based on The distance traveled by the axis guide rail records the distance of each image. Axis position information, where, A safe distance that can be customized by the user; Step 4: Use a depth-of-field stitching algorithm to... Images with different depths of field are stitched together and fused to obtain an initial 3D image of the defect to be tested. Step 5: Based on the initial 3D image of the defect to be tested, plan the imaging of the defect at high magnification. The scan path below generates Group by , , , Composition of scan parameters; Step 6: Based on the scanning parameters, use the illumination imaging module to perform a high-magnification scan of the defects in the sample under test. By controlling the displacement module, the sample under test is moved sequentially relative to its initial position to... The coordinate positions corresponding to the group scan parameters; At each coordinate location, each strobe light source is illuminated via time-division strobe control, along... to Range, continuous acquisition Images at different contour height positions; wherein, the number of images acquired from each stroboscopic source is [number missing]. Laser ruler distance measurement system The precise movement distance of the axis guide rail is recorded for each image. Axis position information; Step 7: Using a depth-of-field stitching algorithm, at each coordinate location, collect data from each stroboscopic light source. Images with different depths of field are stitched together and merged, and the result is recorded as follows: At the i-th coordinate position, the th The stitched and fused image obtained from each light field is , ; Step 8: Using a field-of-view stitching algorithm, the [image / data] will be stitched together in the [image / data] section. Under a light field The images are stitched together to obtain the final result. Under a light field 3D full-aperture stitched image of the defect .

[0014] Furthermore, in step 6, the number of images acquired from each stroboscopic light source... More than twice .

[0015] Furthermore, steps 4 and 7 use the same depth-of-field stitching algorithm, the specific process of which is as follows: S1, Assuming the motorized zoom imaging module obtains... Initial grayscale image ;in, express The first image Image, corresponding The axis position information is ; These are the column coordinates of the pixels in the image. These are the row coordinates of the pixels in the image; S2. Apply Gaussian filtering to the initial grayscale image. Processing is performed to obtain the image. Mean filtering is used on the initial grayscale image. Processing is performed to obtain the image. ; S3, Initial grayscale image With images Perform the difference operation to obtain the difference image. ; S4, Differential Image The image is obtained by processing it using the following formula. ; ; In the formula, To find the maximum value function, For gradient operators; S5, Image-based For the initial grayscale image The image is processed to enhance the focus of the pixels, resulting in a processed image. ; S6, By comparing pixels Location Image Find the maximum value, and the position of the maximum value. Images, for Image The sharp pixels within the depth of field are stitched together to form a single depth-of-field image. , ; S7. Stitch images based on depth of field Each pixel The place Initial grayscale image corresponding Axis position information Calculate to obtain the 3D contour of the defect. , .

[0016] Further, in step S5, the initial grayscale image is... The processing is carried out, and the specific process is as follows: S5-1. First, calculate the coefficients according to the following formula. : ; ; In the formula, Represents the two-dimensional coordinates of image pixels. ,gather pixel size And includes pixel position The area For threshold; It is an integer; To characterize the image The middle pixel is located in the set Image regions in; To characterize the image The middle pixel is located in the set Image regions in; For image The average grayscale value of all pixels; For image The average grayscale value of all pixels; For image The middle pixel is located in the set The average grayscale value of all pixels; For image The standard deviation of grayscale values ​​for all pixels; S5-2. Calculate the coefficients according to the following formula. : ; S5-3. Calculate the coefficients according to the following formula. and : ; ; In the formula, the set pixel size And includes pixel position The region, pixel two-dimensional coordinates lie in Image center coordinates The pixel location; S5-4. Calculate the image according to the following formula. and : ; ; S5-5, will and Substitute into the following formula to calculate : .

[0017] Furthermore, in step 8, a field-of-view stitching algorithm is used, with the following formula: ; In the formula, This is the image field-of-view stitching function.

[0018] Compared with the prior art, the present invention has the following beneficial effects: This invention innovatively overcomes the limitation of the inability to identify the longitudinal depth of defects, building upon the traditional monocular vision inspection system which can only be used for two-dimensional lateral dimension identification of defects. The designed laser ruler ranging system can improve the identification accuracy in the longitudinal depth direction of defects, with longitudinal accuracy and resolution far superior to current ultra-depth-of-field microscopes based on grating rulers and high-precision motors. The established depth-of-field stitching algorithm can reconstruct the spatial 3D model of defects. Combined with multi-field time-division stroboscopic imaging, it can obtain spatial 3D models of defects with different contrasts under different illumination angles and illumination aperture angles, providing technical support for the quantitative evaluation of the spatial three-dimensional dimensions of surface micro- and nano-defects. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a schematic diagram of a monocular microscopic imaging surface micro-nano defect 3D reconstruction detection device according to an embodiment of the present invention.

[0021] Figure 2 This is a flowchart of a method for 3D reconstruction and detection of surface micro-nano defects using monocular microscopy, according to an embodiment of the present invention.

[0022] Figure 3 The defect depth-of-field stitched image obtained in an embodiment of the present invention .

[0023] Figure 4 The defect depth-of-field stitched 3D image obtained in an embodiment of the present invention . Detailed Implementation

[0024] 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, 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.

[0025] It should be noted that, unless otherwise specified, the features in the following embodiments and implementation methods can be combined with each other.

[0026] like Figure 1 As shown, a monocular microscopic imaging surface micro-nano defect 3D reconstruction detection device includes an illumination imaging module 1, a displacement module and a laser ruler ranging system.

[0027] The illumination imaging module 1 includes an electrically operated zoom imaging module 2 and seven strobe light sources 3 arranged around it; the strobe light sources 3 use strobe LEDs, and the illumination angle of each strobe light source 3 is... Illumination aperture angle The electronically adjustable light source 3 can form an illumination field with different optical parameters, and each stroboscopic light source 3 is arranged in a uniform ring at equal angular intervals.

[0028] The displacement modules consist of sequentially stacked units arranged from bottom to top. 4-axis guide rail 5-axis guide rail Spin axis 6 7. Shaft guide rail The guide rail 8, the stage 9, and the components positioned above the stage 9 are also present. 10-axis guide rail; Shaft guide rail 10 is installed on On the swing shaft 11.

[0029] The laser ruler ranging system includes a laser 12, a beam splitter 13, a reflector 14, a reference mirror 15, and an interferometer 16. The laser 12 is mounted on... The swing shaft 11, beam splitter 13, reflector 14, and reference mirror 15 are mounted on... The axial guide rail 10 and the interferometer 16 are mounted on the illumination imaging module 1.

[0030] In this embodiment, the sample to be tested is placed on the stage 9, and the specific detection method is as follows: Figure 2 As shown, it includes the following steps: 1) Adjustment Shaft guide rail 4 and Shaft guide rail 5, so that The axis of spin axis 6 is collinear with the optical axis 2-1 of motorized zoom imaging module 2.

[0031] 2) Turn on the laser 12, and adjust the beam splitter 13 and the interferometer 16 so that the reflected light EC from the laser 12 is reflected back to the receiver position at point E on the laser 12 by the interferometer 16. The incident light AB and the reflected light EC, and the optical axis 2-1 of the motorized zoom imaging module 2 are aligned. The directions of motion of the axis guide rails 10 are parallel to each other. Adjust the reflector 14 and the reference mirror 15 so that the reflected light DE reflected back to point E by the reference mirror 15 is collinear with the light ray EC.

[0032] 3) At 5x magnification, manually focus to find the focal plane at the highest point of the defect surface profile (10.018mm) and the lowest point of the defect surface profile (10.015mm).

[0033] 4) The highest and lowest positions of the defect surface contour are accurately located using a sharpness evaluation algorithm. The corresponding positions are calculated to be 10.016 mm and 10.013 mm, respectively.

[0034] 5) Select the feature point at the highest position of the defect surface contour as the initial reference point. By adjusting 7. Shaft guide rail Shaft guide rail 8, so that the initial reference point O Move to the line where the optical axis 2-1 of the motorized zoom imaging module 2 is located.

[0035] 6) Control the seven stroboscopic light sources 3, which are arranged in a uniform ring, to all be at an incident angle. , Provide lighting.

[0036] 7) Along Fifty images were continuously acquired at different contour height positions within a range of 10.012mm to 10.017mm, and the corresponding data for each image were recorded. Axis position , This location information is provided by The grating ruler on the axis guide rail 10 provides data.

[0037] 8) Using a depth-of-field stitching algorithm, 50 images with different depths of field are stitched and fused to obtain an initial 3D image of the defect to be tested.

[0038] 9) Based on the initial 3D image of the defect to be tested, plan the scanning path of the defect at 50x imaging magnification and generate... , , , The scan parameters generated a total of 75 based on the path planning algorithm. coordinate.

[0039] 10) Set the illumination parameters for the 7-beam stroboscopic light source 3 as follows: ; Each stroboscopic light source 3 and the motorized zoom imaging module 2 form a light field.

[0040] 11) For each generated in step 9) Position, move sequentially 4-axis guide rail 5-axis guide rail Spin axis 6 Swing axis 11 to the specified Position, adjustment Shaft guide rail 10 along Within a coordinate range of 10.012mm to 10.017mm, 200 images at different contour heights are continuously acquired at equal intervals. Simultaneously, during acquisition, time-division stroboscopic control is used to operate the motorized zoom imaging modules 2 in seven light fields, ensuring that each... The coordinates allow for the acquisition of seven images under different lighting fields. During scanning, the coordinates corresponding to each image are recorded. Axis position , This location information is measured by a laser ruler ranging system. The precise movement distance of the axis guide rail 10 is obtained. Seven sets of images were obtained under seven light fields within the range of coordinates from 10.012 mm to 10.017 mm, with each set containing 200 images.

[0041] 12) At the scanning position Next, using a depth-of-field stitching algorithm, 200 images with different depths of field from each of the 7 groups of images are stitched and fused to obtain the 3D sub-aperture image of the defect at that scanning location. Let the first image be denoted as the first sub-aperture image. indivual The coordinate position of the first The 3D sub-aperture image of the depth-of-field stitching and fusion defects obtained from the light field is as follows: , .

[0042] 13) Using a field-of-view stitching algorithm, obtain 7 stitched 3D full-aperture images of the defect under test in 7 light fields. : ; In the formula, This is the image field-of-view stitching function.

[0043] In this embodiment of the invention, steps 8) and 12) employ the same depth-of-field stitching algorithm, and the specific process is as follows: Step 1: First, the motorized zoom imaging module 2 is used to acquire pixel sizes of... Initial grayscale image of the sample Among them, for the low-magnification case in step 8), Indicates the first of 50 images Zhang image; for the high magnification case in step 12), Represents the first of 200 images Zhang image.

[0044] Step 2: Next, apply Gaussian filtering to the initial grayscale image. Processing is performed to obtain the image. Mean filtering is used on the initial grayscale image. Processing is performed to obtain the image. .

[0045] Step 3: Next, take the initial grayscale image... With images Perform the difference operation to obtain the difference image. : .

[0046] Step 4: Further obtain the image using the following formula. : ; In the formula, To find the maximum value function, This is the gradient operator.

[0047] Step 5: Process the initial grayscale image Process and enhance the image The processed image is obtained by focusing on the clearly focused pixel area. The specific calculation steps are as follows: Step 5-1: Calculate the coefficients according to the following formula. : ; ; In the formula, Represents the two-dimensional coordinates of image pixels. ,gather pixel size And includes pixel position The area.

[0048] Step 5-2: Calculate the coefficients according to the following formula. : ; Step 5-3: Calculate the coefficients according to the following formula. and : ; In the formula, the set pixel size And includes pixel position region, pixel position lie in Represented pixel size Image center coordinates The pixel location.

[0049] Step 5-4: Calculate the image according to the following formula. and : ; Step 5-5: Calculate according to the following formula : .

[0050] Step 6: By comparing pixels Location Image Find the maximum value, and the position of the maximum value. Images, for Image The sharp pixels within the depth of field are stitched together to form a single depth-of-field image. , For the low-magnification case in step 8), This represents 50 images; for the high magnification case in step 12), This represents 200 images.

[0051] Step 7: Stitch the images according to depth of field Each pixel The place Initial grayscale image corresponding Axis position information Calculate to obtain the 3D contour of the defect. , .

[0052] about Axis position information For the low-magnification case in step 8), by The grating ruler on the axis guide rail 10 provides data; for the high magnification case in step 12), the measurement is performed by the laser ruler ranging system. The precise movement distance of the axis guide rail 10 is obtained.

[0053] like Figure 3 The image shown illustrates a defect depth-of-field stitched image obtained according to an embodiment of the present invention. ,like Figure 4 The image shown illustrates a defect depth-of-field stitched 3D image obtained according to an embodiment of the present invention. In traditional optics, high resolution and large depth of field are contradictory; increasing magnification to achieve high resolution inevitably sacrifices depth of field. The method described in this invention "bypasses" this physical limitation. While maintaining the highest lateral resolution of the optical system, it artificially creates a depth of field far exceeding physical limits. This allows for effective observation of samples with millimeter- or even centimeter-level undulations at sub-micron resolution. Figure 3 and Figure 4 As can be seen from the image, the method can better restore the three-dimensional morphological information of surface defects outside the depth of field of the original imaging system. The final image obtained by image fusion is clear and sharp from the highest point to the lowest point of the defect contour.

[0054] The embodiments described above provide a detailed explanation of the technical solutions and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the present invention. Any modifications, additions, and equivalent substitutions made within the scope of the principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A monocular microscopic imaging surface micro-nano defect 3D reconstruction detection method, characterized in that, Includes the following steps: Step 1, the illumination imaging module (1) in low imaging magnification Next, use The shaft guide rail (10) manual focus, find the highest position of the surface profile of the measured defect The focal plane and the lowest position The focal plane; Step 2: Accurately locate the highest and lowest positions of the defect surface contour using a sharpness evaluation algorithm, and record the corresponding positions as follows: , ; Step 3, along to Range, continuous acquisition Images at different heights, based on The travel distance of the guide rail (10) records the distance of each image. Axis position information, where, A safe distance that can be customized by the user; Step 4: Use a depth-of-field stitching algorithm to... Images with different depths of field are stitched together and fused to obtain an initial 3D image of the defect to be tested. Step 5: Based on the initial 3D image of the defect to be tested, plan the imaging of the defect at high magnification. The scan path below generates Group by , , , Composition of scan parameters; Step 6: According to the scanning parameters, use the illumination imaging module (1) to perform high-magnification scanning of the defects in the sample under test. By controlling the displacement module, the sample under test is moved sequentially relative to the initial position to... The coordinate positions corresponding to the group scan parameters; At each coordinate location, each strobe light source is illuminated via time-division strobe control, along... to Range, continuous acquisition Images at different contour height positions; wherein, the number of images acquired from each stroboscopic source is [number missing]. Laser ruler distance measurement system The precise movement distance of the axis guide rail (10) is recorded for each image. Axis position information; Step 7: Using a depth-of-field stitching algorithm, at each coordinate location, collect data from each stroboscopic light source. Images with different depths of field are stitched together and merged, and the result is recorded as follows: At the i-th coordinate position, the th The stitched and fused image obtained from each light field is , ; Step 8, using a field-of-view stitching algorithm, will stitch the data in the first... Under a light field The images are stitched together to obtain the final result. Under a light field 3D full-aperture stitched image of the defect ; Steps 4 and 7 use the same depth-of-field stitching algorithm. The specific process is as follows: S1, Assuming the motorized zoom imaging module obtains... Initial grayscale image ;in, express The first image Image, corresponding The axis position information is ; These are the column coordinates of the pixels in the image. These are the row coordinates of the pixels in the image; S2. Apply Gaussian filtering to the initial grayscale image. Processing is performed to obtain the image. Mean filtering is used on the initial grayscale image. Processing is performed to obtain the image. ; S3, Initial grayscale image With images Perform the difference operation to obtain the difference image. ; S4, Differential Image The image is obtained by performing the following formula processing. ; ； In the formula, To find the maximum value function, For gradient operators; S5, Image-based For the initial grayscale image The image is processed to enhance the focus of the pixels, resulting in a processed image. ; S6, By comparing pixels Location Image Find the maximum value, and the position of the maximum value. Images, for Image The sharp pixels within the depth of field are stitched together to form a single depth-of-field image. , ; S7. Stitch images based on depth of field Each pixel The place Initial grayscale image corresponding Axis position information Calculate to obtain the 3D contour of the defect. , .

2. The method for 3D reconstruction and detection of surface micro / nano defects using monocular microscopy according to claim 1, characterized in that, In step 6, the number of images acquired from each stroboscopic light source More than twice .

3. The method for 3D reconstruction and detection of surface micro / nano defects using monocular microscopy according to claim 1, characterized in that, In step S5, the initial grayscale image is processed. The processing is carried out, and the specific process is as follows: S5-1. First, calculate the coefficients according to the following formula. ； ； In the formula, Represents the two-dimensional coordinates of image pixels. ,gather pixel size And includes pixel position The area For the threshold; It is an integer; To characterize the image The middle pixel is located in the set Image regions in; To characterize the image The middle pixel is located in the set Image regions in; For image The average grayscale value of all pixels; For image The average grayscale value of all pixels; For image The middle pixel is located in the set The average grayscale value of all pixels; For image The standard deviation of grayscale values ​​for all pixels; S5-2. Calculate the coefficients according to the following formula. ； S5-3. Calculate the coefficients according to the following formula. and ； ； In the formula, the set pixel size And includes pixel position The region, pixel two-dimensional coordinates lie in Image center coordinates The pixel location; S5-4. Calculate the image according to the following formula. and ； ； S5-5, will and Substitute into the following formula to calculate 。 4. The method for 3D reconstruction and detection of surface micro / nano defects using monocular microscopy according to claim 1, characterized in that, In step 8, a field-of-view stitching algorithm is used, with the following formula: ； In the formula, This is the image field-of-view stitching function.

5. A monocular microscopic imaging surface micro / nano defect 3D reconstruction detection device, used to implement the monocular microscopic imaging surface micro / nano defect 3D reconstruction detection method according to any one of claims 1 to 4, characterized in that, Includes an illumination imaging module (1), a displacement module, and a laser ruler ranging system; The displacement modules include those arranged sequentially from bottom to top. Shaft guide rail (4) Shaft guide rail (5) Spin axis (6) Shaft guide rail (7) Shaft guide rail (8), stage (9), and a mounting plate positioned above the stage (9). Shaft guide rail (10); the The shaft guide rail (10) is installed on On the swing axis (11); The illumination imaging module (1) is installed on On the axial guide rail (10), there is an electrically operated zoom imaging module (2) and other components uniformly arranged around it in a ring. Individual stroboscopic sources; illumination angle of each stroboscopic source Illumination aperture angle Electronically adjustable, under given parameters , Each strobe light source and the motorized zoom imaging module (2) together form one light field, which can form a maximum of [number] light fields at the same time. A light field; The laser ruler distance measuring system is used for measurement The actual travel distance of the guide rail (10).

6. The monocular microscopic imaging surface micro / nano defect 3D reconstruction detection device according to claim 5, characterized in that, The laser ruler ranging system includes a laser (12), a beam splitter (13), a reflector (14), a reference mirror (15), and an interferometer (16). Among them, the laser (12) is installed in The swing axis (11), beam splitter (13), reflector (14), and reference mirror (15) are mounted on the... The axial guide rail (10) and the interferometer (16) are mounted on the illumination imaging module (1).

7. The monocular microscopic imaging surface micro / nano defect 3D reconstruction detection device according to claim 5, characterized in that, Number of strobe light sources It is an odd number greater than 1.

Images