A single-camera multi-projection dual-axis projection structured light 3D reconstruction measurement system

Abstract

The present invention discloses a single-camera multi-projection dual-axis projection structured light three-dimensional reconstruction measurement system, which relates to the technical field of structured light measurement. The method comprises: using a structured light projector to project structured light to a detection object to generate a dual-axis phase-coded fringe image and collect it, reconstructing a height map of the detection object according to the image, eliminating the noise of the occluded area according to the installation position of the structured light projector group and the collected image and obtaining a height map of the unobstructed real height area, fusing the height map of the real height area to obtain a complete and accurate height map, obtaining point cloud data of the detection object according to the height map and performing three-dimensional reconstruction; the present invention can project a 45-degree rotated structured light through a structured light projector and present dual-axis phase-coded fringe on the detection object, eliminate the influence of shadows and high reflections, increase the number of effective point clouds, improve the measurement precision and accuracy of the system, and obtain a complete and accurate measurement result of the detection object.

Description

Technical Field

[0001] The present invention relates to the technical field of structured light measurement, and in particular to a single-camera multi-projection dual-axis projection structured light three-dimensional reconstruction measurement system. Background Art

[0002] Ball Grid Array (BGA) packaging technology is a surface bonding technology used on integrated circuits. This technology is often used to permanently fix devices such as microprocessors. During the BGA packaging process, problems such as bridging, false soldering, cold soldering, bubbles, dirt, cracking, offset, and gradual soldering often occur, causing the chip to be short-circuited and unusable. The detection equipment needs to be regularly inspected and maintained to ensure its performance and reliability. Optical detection plays a pivotal role in the field of BGA packaging and testing. It is an indispensable link in the production process of semiconductor chips. It is usually distributed in wafer detection, particle appearance defect detection, patch / wire bonding detection, and plastic packaging appearance detection. Users can sort the chips according to the detection results. However, due to the high reflectivity and close arrangement of the solder ball surface, how to solve the high reflectivity and shadow problems of BGA detection has always been a difficulty in the industry. In order to solve the above difficulties, people in this field urgently need to develop a dual-axis projection structured light three-dimensional reconstruction measurement method that can eliminate the influence of shadows and high reflectivity and increase the number of effective point clouds. Summary of the invention

[0003] The present invention provides a single-camera multi-projection dual-axis projection structured light three-dimensional reconstruction measurement system, comprising: a main support (1), a structured light projector group (2) and a high-speed industrial acquisition camera (3);

[0004] A structured light projector bracket (11) is fixedly mounted on the periphery of the main bracket (1), and the structured light projector group (2) is fixed to the periphery of the main bracket (1) via the structured light projector bracket (11); a camera bracket (12) is fixedly mounted on the upper portion of one side of the main bracket, and a high-speed industrial acquisition camera (3) is fixed to the inside of the main bracket (1) via the camera bracket;

[0005] The structured light projector group (2) comprises four structured light projectors, and the four structured light projectors are all rotated 45 degrees along the optical axis of their respective projection lenses and then fixed on corresponding structured light projector brackets (11);

[0006] The high-speed industrial acquisition camera (3) is fixed inside the main bracket (1), a high-precision double telecentric lens (31) is installed at the lower end, and the detection object is placed directly below the high-precision double telecentric lens (31);

[0007] Four structured light projectors sequentially project encoded structured light grating stripes at ±45 degrees onto the object to be detected. After the structured light grating stripes reach the object to be detected, dual-axis phase-encoded stripes are formed. A high-speed industrial acquisition camera (3) photographs the dual-axis phase-encoded stripes on the object to be detected through a high-precision dual-telecentric lens (31) to form a dual-axis phase-encoded stripe image. Structured light three-dimensional reconstruction is performed based on the dual-axis phase-encoded stripe image to obtain the three-dimensional morphology of the object to be detected.

[0008] A single-camera multi-projection dual-axis projection structured light three-dimensional reconstruction and measurement system as described above, wherein each structured light projector in the structured light projector group (2) is rotated 45 degrees and then projects structured light grating stripes onto the detection object, so that it presents a ±45 degree dual-axis phase-encoded stripe image on the surface of the detection object; wherein the rotation angle of the structured light projector can be set to other rotation angles, and structured light grating stripes with corresponding rotation angles are projected onto the detection object.

[0009] A single-camera multi-projection dual-axis projection structured light three-dimensional reconstruction measurement system as described above, wherein the structured light projector group (2) includes a first structured light projector (21), a second structured light projector (22), a third structured light projector (23) and a fourth structured light projector (24); during measurement, the first structured light projector (21) is first turned on according to the rotation angle to project structured light grating stripes onto the detection object, and the structured light grating stripes present a dual-axis phase-encoded stripe image in the projection direction of the first structured light projector (21) on the surface of the detection object, and then the high-speed industrial acquisition camera (3) acquires the image through a high-precision dual telecentric lens (31). A first dual-axis phase-coded fringe image is collected, and then a second structured light projector (22), a third structured light projector (23), and a fourth structured light projector (24) are sequentially turned on according to the rotation angle to sequentially project structured light grating fringes in corresponding directions onto the detection object. The structured light grating fringes sequentially present dual-axis phase-coded fringe images in the projection directions of the second structured light projector (22), the third structured light projector (23), and the fourth structured light projector (24) on the surface of the detection object. A high-speed industrial acquisition camera (3) sequentially acquires the second, third, and fourth dual-axis phase-coded fringe images through a high-precision dual telecentric lens (31).

[0010] The present invention also provides a single-camera multi-projection dual-axis projection structured light 3D reconstruction measurement method, which is applied to any of the dual-axis projection structured light 3D reconstruction measurement systems described above, and the dual-axis projection structured light 3D reconstruction measurement method comprises:

[0011] Step S1, using four structured light projectors to sequentially project structured light onto the surface of the detection object, and sequentially generating four full-frame clear biaxial phase-coded fringe images with equal width and biaxial projection of ±45 degrees;

[0012] Step S2, using a high-speed industrial acquisition camera and a high-precision dual telecentric lens to sequentially acquire four dual-axis phase-coded fringe images;

[0013] Step S3, reconstructing four height maps of the detection object according to the four dual-axis phase-encoded fringe images;

[0014] Step S4, according to the installation position of the structured light projector group and the four collected dual-axis phase-encoded fringe images, obtain the blocked noise area and the unblocked real height area in the four height maps and remove the noise in the blocked area to obtain four unblocked real height area height maps;

[0015] Step S5: fuse the four unobstructed real height area height maps to obtain a complete and accurate height map, obtain point cloud data of the detected object based on the height map, and perform three-dimensional reconstruction of the detected object.

[0016] As described above, a single-camera multi-projection dual-axis projection structured light 3D reconstruction measurement method is described, wherein four structured light projectors are used to sequentially project structured light onto the surface of the object being tested, and the sub-steps of sequentially generating four full-frame clear dual-axis phase-coded fringe images with equal width and dual-axis projection of ±45 degrees are as follows:

[0017] According to the system setting, four structured light projectors are fixed and rotated 45 degrees according to the structured light projector fixing bracket to project structured light grating stripes and present a ±45 degree dual-axis phase coded stripe image on the surface of the detection object;

[0018] The dual-axis phase-encoded stripes are corrected by fringe correction technology to make them appear as full-frame clear dual-axis phase-encoded stripes with equal width and dual-axis projection of ±45 degrees.

[0019] The above-mentioned single-camera multi-projection dual-axis projection structured light 3D reconstruction measurement method, wherein the sub-steps of sequentially collecting four dual-axis phase-coded fringe images using a high-speed industrial acquisition camera in conjunction with a high-precision dual telecentric lens are as follows:

[0020] Calculate the placement parallel deviation value according to the placement position of the high-speed industrial acquisition camera and the detection object, obtain the wrong placement device according to the placement parallel deviation value and adjust its position so that the adjusted high-speed industrial acquisition camera and the detection object are in a horizontal position;

[0021] Calculate the placement vertical deviation value according to the placement position of the high-precision dual telecentric lens and the detection object, obtain the wrong placement device according to the placement vertical deviation value and adjust its position so that the adjusted high-precision dual telecentric lens and the detection object are in a vertical position;

[0022] Use an adjusted height industrial acquisition camera and a high-precision dual telecentric lens to sequentially acquire four dual-axis phase-coded fringe images of the inspection object at four angular directions.

[0023] As described above, in the single-camera multi-projection dual-axis projection structured light 3D reconstruction measurement method, the sub-steps of reconstructing four height maps of the detection object according to four dual-axis phase-coded fringe images are as follows:

[0024] Calculate image phase values ​​based on four dual-axis phase-encoded fringe images;

[0025] The height of the detected object is calculated according to the phase value of each point of the phase-coded fringe image and four height maps of the detected object are reconstructed.

[0026] As described above, a single-camera multi-projection dual-axis projection structured light 3D reconstruction measurement method is described, wherein, according to the installation position of the structured light projector group and the four collected dual-axis phase-coded fringe images, the occluded noise area and the unobstructed true height area in the four height maps are obtained and the noise in the occluded area is removed. The sub-steps of obtaining the four unobstructed true height area height maps are as follows:

[0027] Obtain shadow points and high-reflection points based on the installation positions of each structured light projector and mark them in four height maps;

[0028] The switch threshold method is used to identify the occluded noise area and the unoccluded true height area in the four height maps;

[0029] The marked points in the dual-axis phase-encoded fringe image and the height map are removed from the corresponding four height maps to obtain four unobstructed real height area height maps.

[0030] In the above-mentioned single-camera multi-projection dual-axis projection structured light 3D reconstruction measurement method, four unobstructed real height area height maps are fused to obtain a complete and accurate height map, and the sub-steps of obtaining point cloud data of the detection object according to the height map and performing 3D reconstruction of the detection object are as follows:

[0031] The four unobstructed real height area height maps are merged to obtain a complete and accurate height map;

[0032] Generate point cloud data of the detected object based on the height map;

[0033] The object to be inspected is reconstructed in three dimensions based on its point cloud data.

[0034] The beneficial effects achieved by the present invention are as follows: the present invention can utilize a single acquisition camera and multiple structured light projectors, and each structured light projector is fixedly rotated 45 degrees to project structured light and present dual-axis phase-coded stripes on the surface of the detection object. The acquisition camera sequentially acquires four dual-axis phase-coded stripe images rotated 45 degrees. Since the projection direction is rotated 45 degrees, this dual-axis phase-coded stripe image eliminates the influence of shadows and high reflections, increases the number of effective point clouds, thereby obtaining a 3D point cloud with higher integrity, and improves the measurement precision and accuracy of the system, thereby obtaining complete and accurate measurement and detection results at all angles of the detection object. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0036] Figure 1 It is a schematic diagram of a single-camera multi-projection dual-axis projection structured light three-dimensional reconstruction measurement system provided in Example 1 of the present application;

[0037] Figure 2 This is a schematic diagram of a camera collecting images after the structured light projector rotates in a single-camera multi-projection dual-axis projection structured light three-dimensional reconstruction measurement system provided in Example 1 of the present application;

[0038] Figure 3 This is a flow chart of a single-camera multi-projection dual-axis projection structured light three-dimensional reconstruction measurement method provided in Example 2 of the present application.

[0039] Figure numerals: 1. Main bracket; 2. Structured light projector group; 3. High-speed industrial acquisition camera; 11. Structured light projector bracket; 12. Camera bracket; 21. First structured light projector; 22. Second structured light projector; 23. Third structured light projector 24. Fourth structured light projector; 31. High-precision dual telecentric lens. DETAILED DESCRIPTION

[0040] The following is a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0041] Embodiment 1

[0042] like Figure 1-2As shown, the first embodiment of the present application provides a single-camera multi-projection dual-axis projection structured light three-dimensional reconstruction measurement system, including: a main support 1, a structured light projector group 2 and a high-speed industrial acquisition camera 3;

[0043] A structured light projector bracket 11 is fixedly installed around the main bracket 1, and the structured light projector group 2 is fixed to the outside of the main bracket 1 through the structured light projector bracket 11; a camera bracket 12 is fixedly installed on the upper part of one side of the main bracket, and a high-speed industrial acquisition camera 3 is fixed inside the main bracket 1 through the camera bracket;

[0044] The structured light projector group 2 includes four structured light projectors, and the four structured light projectors are all rotated 45 degrees along the optical axis of their respective projection lenses and then fixed on the corresponding structured light projector bracket 11;

[0045] The high-speed industrial acquisition camera 3 is fixed inside the main bracket 1, and a high-precision double telecentric lens 31 is installed at the lower end, and the detection object is placed directly below the high-precision double telecentric lens 31;

[0046] Four structured light projectors project encoded structured light grating stripes of ±45 degrees onto the object to be tested in turn. The structured light grating stripes form dual-axis phase-coded stripes when reaching the object to be tested. The high-speed industrial acquisition camera 3 shoots the dual-axis phase-coded stripes on the object to be tested through a high-precision dual-telecentric lens 31 to form a dual-axis phase-coded stripe image. Structured light three-dimensional reconstruction is performed based on the dual-axis phase-coded stripe image to obtain the three-dimensional morphology of the object to be tested.

[0047] In the embodiment of the present application, each structured light projector in the structured light projector group 2 is rotated 45 degrees according to the rotation angle setting and then projects structured light grating stripes on the detection object, so that it presents a rotated ±45-degree dual-axis phase-encoded stripe image on the surface of the detection object, wherein the rotation angle of the structured light projector can be set to other rotation angles, and the structured light grating stripes of the corresponding rotation angle are projected to the detection object, and other rotation angles also belong to the content of the present invention. At the same time, if other structured light projectors are replaced to directly project ±45-degree or other angle stripes, dual-axis projection also belongs to the content of the present invention, which is not specifically limited here.

[0048] The structured light projector group 2 includes a first structured light projector 21, a second structured light projector 22, a third structured light projector 23 and a fourth structured light projector 24; during measurement, the first structured light projector 21 is first turned on according to the rotation angle to project structured light grating stripes to the detection object, and the structured light grating stripes present a dual-axis phase-coded stripe image in the projection direction of the first structured light projector 21 on the surface of the detection object, and then the high-speed industrial acquisition camera 3 collects the first dual-axis phase-coded stripe image through the high-precision dual telecentric lens 31, and then the second structured light projector 22, the third structured light projector 23 and the fourth structured light projector 24 are turned on according to the rotation angle to project structured light grating stripes in corresponding directions to the detection object in turn, and the structured light grating stripes present a dual-axis phase-coded stripe image in the projection direction of the second structured light projector 22, the third structured light projector 23 and the fourth structured light projector 24 on the surface of the detection object in turn, and the high-speed industrial acquisition camera 3 collects the second, third and fourth dual-axis phase-coded stripe images in turn through the high-precision dual telecentric lens 31, and the dual-axis phase-coded stripe images are as shown in FIG. Figure 2 shown.

[0049] The four collected dual-axis phase-coded fringe images are dual-axis phase-coded fringe images at four different angles of the detection object, which enables them to project and collect all points of the detection object. In addition, since the structured light projector group is rotated 45 degrees, it can effectively eliminate the influence of shadows and high reflections, and obtain point cloud data with higher integrity, thereby performing complete three-dimensional reconstruction and measurement of the detection object.

[0050] Embodiment 2

[0051] like Figure 3 As shown, the second embodiment of the present application provides a single-camera multi-projection dual-axis projection structured light 3D reconstruction measurement method, the method comprising the following steps:

[0052] Step S1, using four structured light projectors to sequentially project structured light grating stripes onto the surface of the object to be detected, and sequentially generating four full-frame clear biaxial phase-encoded stripe images with equal width and biaxial projection of ±45 degrees;

[0053] Furthermore, the sub-steps of using four structured light projectors to sequentially project structured light grating stripes onto the surface of the object to be detected, and sequentially generating four full-frame clear biaxial phase-encoded stripe images with equal width and biaxial projection of ±45 degrees are as follows:

[0054] Step S11, according to the system setting, four structured light projectors are fixed and rotated 45 degrees according to the structured light projector fixing bracket to project structured light grating stripes and present a ±45 degree dual-axis phase coded stripe image on the surface of the detection object;

[0055] Specifically, traditional structured light projection can only project horizontal and vertical phase-coded stripe images at 0 and 90 degrees. This system rotates the structured light projector 45 degrees and projects structured light on the surface of the object to be detected, presenting ±45-degree dual-axis phase-coded stripes on the surface of the object to eliminate shadows and high reflections. The fixed rotation angle of 45 degrees can also be set to other fixed angles.

[0056] Step S12, correcting the dual-axis phase-encoded fringes by using a fringe correction technology so that the dual-axis phase-encoded fringes appear as full-frame clear dual-axis phase-encoded fringes with equal width and dual-axis projection at ±45 degrees;

[0057] Specifically, in order to make the phase-encoded stripes presented on the surface of the detected object full-frame clear and equal-width dual-axis phase-encoded stripes and biaxially projected into ±45 degrees, the dual-axis phase-encoded stripes are corrected using fringe correction technology.

[0058] Among them, the fringe correction technology optimizes the rotation degree and fringe width of the dual-axis phase-encoded stripes to the optimal value, thereby adjusting the dual-axis phase-encoded stripes to make them appear as full-frame clear dual-axis phase-encoded stripes with equal width and dual-axis projection of ±45 degrees.

[0059] The stripe correction technology is implemented as follows:

[0060] Based on the rotation angle of the structured light projector through the function The dual-axis phase-encoded stripes are adjusted so that each stripe projected by the structured light projector is a rotational projection angle, where is the rotation adjustment function, For the The number of degrees before the stripes rotate, For the The degree of rotation of the stripes, The value range is , is the number of phase-encoded fringes, is the rotation degree of the structured light projector, For the The rotation of the stripes affects the adjustment characteristics. is the rotation adjustment value, Adjust the index for fringe rotation, Adjust the exponent for the rotation angle.

[0061] In order to make the dual-axis phase-encoded fringes clear and easy to analyze, the formula The fringe width that can make the dual-axis phase encoding fringe clearest is calculated, where: is the optimal stripe width, is the wavelength of the structured light, To select the optimal distance from the object to the high-precision bi-telecentric lens, To select the optimal distance from the object to the high-speed industrial acquisition camera, is selected as the optimal distance between two adjacent stripes.

[0062] Based on the optimal stripe width Through the function Adjust the width of each stripe so that the width of each stripe is the optimal stripe width, where: is the width adjustment function, For the Stripe width, The value range is , is the total number of phase-encoded fringes, is the optimal stripe width, Adjust the eigenvalue for the stripe width, The scaling exponent that adjusts the feature value for the fringe width.

[0063] Step S2, using a high-speed industrial acquisition camera and a high-precision dual telecentric lens to sequentially acquire four dual-axis phase-coded fringe images;

[0064] Furthermore, the sub-steps of sequentially collecting four dual-axis phase-coded fringe images using a high-speed industrial acquisition camera and a high-precision dual telecentric lens are as follows:

[0065] Step S21, calculating the placement parallel deviation value according to the placement position of the high-speed industrial acquisition camera and the detection object, obtaining the wrong placement device according to the placement parallel deviation value and adjusting its position so that the adjusted high-speed industrial acquisition camera and the detection object are in a horizontal position;

[0066] Specifically, in order to ensure that the object to be detected and the high-speed industrial acquisition camera are in parallel, the formula Calculate the placement parallel deviation value to obtain the non-parallel position of the inspection object and the high-speed industrial acquisition camera and adjust the position of the wrongly placed device according to the placement parallel deviation value. To place the parallel deviation value, a parallel gap is established between the inspection object and the high-speed industrial acquisition camera according to the position of the high-speed industrial acquisition camera fixed bracket. vertical lines, and For the The high-speed industrial acquisition camera and the horizontal coordinate point of the detected object on the vertical line, and For the The high-speed industrial acquisition camera and the vertical coordinate point of the detection object on the vertical line, and For the The high-speed industrial acquisition camera and the horizontal coordinate point of the detected object on the vertical line, and For the The high-speed industrial acquisition camera and the vertical coordinate point of the detection object on the vertical line, The value range is , The number of vertical lines between the camera and the inspection object is collected for high-speed industry.

[0067] Step S22, calculating the placement vertical deviation value according to the placement position of the high-precision dual telecentric lens and the detection object, obtaining the wrong placement device according to the placement vertical deviation value and adjusting its position so that the adjusted high-precision dual telecentric lens and the detection object are in a vertical position;

[0068] Specifically, in order to ensure that the detection object and the optical axis of the high-precision bi-telecentric lens are in a vertical position, the formula Calculate the placement vertical deviation value to obtain the non-vertical position of the detection object and the optical axis of the high-precision bi-telecentric lens and adjust the position of the wrong device according to the placement vertical deviation value. To place the vertical deviation value, obtain the placement straight line between the center point of the upper plane and the center point of the lower plane according to the optical axis position of the high-precision double telecentric lens. To select two coordinate points from a line and The horizontal and vertical coordinate points are obtained according to the placement position of the detection object, and the placement straight line between the center point of the left plane and the center point of the right plane is obtained. To select two coordinate points from a line and The horizontal and vertical coordinates of .

[0069] Step S23, using the adjusted height industrial acquisition camera and the high-precision dual telecentric lens to sequentially acquire four dual-axis phase-encoded fringe images of the detection object at four angles;

[0070] Specifically, the collected dual-axis phase-coded fringe images are labeled according to the serial number of the structured light projector that projects the structured light, so as to distinguish the collection direction of the detection object.

[0071] Step S3, reconstructing four height maps of the detection object according to the four dual-axis phase-encoded fringe images;

[0072] Furthermore, the sub-steps of reconstructing four height maps of the detection object according to the four dual-axis phase-encoded fringe images are as follows:

[0073] Step S31, calculating image phase values ​​according to four dual-axis phase-encoded fringe images;

[0074] Specifically, the grayscale value of the image pixel is calculated based on the four dual-axis phase-encoded fringe images, and its expression is as follows:

[0075]

[0076] in, The object corresponding to the deformed image Gray value at the point; is the intensity of ambient light; is the light wave amplitude value related to the projected light intensity; For point The corresponding phase value; is the phase shift of the phase shift fringes.

[0077] Since the four dual-axis phase-encoded fringe images collected are moved each time cycles, which means the phase shift each time is Therefore, the phase value is constructed according to the gray value to solve the equation group, and its expression is as follows:

[0078] in, Four dual-axis phase-encoded fringe images are used to detect points on the object. The corresponding grayscale value.

[0079] Combining the four equations, we can get Solve for each point The phase value at .

[0080] Since the inverse tangent function is used in the phase calculation, the phase value range is , such a phase is called a wrapped phase or a truncated phase. In order to reconstruct a continuous phase distribution, we need to unfold the wrapped phase. Usually, this is done by comparing the truncated phase values ​​of two adjacent pixels and adding or subtracting to restore the continuous phase.

[0081] Step S32, calculating the height of the detected object according to the phase value of each point of the phase-coded fringe image and reconstructing four height maps of the detected object;

[0082] According to the formula Calculate the height of each point of the test object ,in, To collect the intersection point of the light from the camera to the object to be detected, The intersection point where the projector projects light to the object to be tested. To collect the distance between the light collected by the camera and the light projected by the projector to the object to be detected, is the angle between the projector’s projection light and the horizontal plane of the object being tested, For point and Point The phase difference, The stripe interval.

[0083] Step S4, according to the installation position of the structured light projector group and the four collected dual-axis phase-encoded fringe images, obtain the blocked noise area and the unblocked real height area in the four height maps and remove the noise in the blocked area to obtain four unblocked real height area height maps;

[0084] Furthermore, according to the installation position of the structured light projector group and the four collected dual-axis phase-encoded fringe images, the blocked noise area and the unblocked real height area in the four height maps are obtained and the noise in the blocked area is removed. The sub-steps of obtaining the four unblocked real height area height maps are as follows:

[0085] Step S41, obtaining shadow points and high reflection points according to the installation positions of each structured light projector and marking them in four height maps;

[0086] Step S42, identifying the blocked noise area and the unblocked real height area in the four height maps by using the switch threshold method;

[0087] Specifically, through the formula Determine each pixel Is it a noise point? is a pixel, For unobstructed points, is the noise point, is the pixel gray value, is the average value of all grayscale values ​​in a height map, is the switching threshold, the switching threshold The expression is .

[0088] Each pixel point is divided into two parts according to the judgment result: an occluded noise area and an unobstructed true height area, and the occluded noise area is marked in each height map.

[0089] Step S43, removing the marked points in the dual-axis phase-encoded fringe image and the height map from the corresponding four height maps to obtain four unobstructed real height area height maps;

[0090] Step S5, fusing the four unobstructed real height area height maps to obtain a complete and accurate height map, obtaining point cloud data of the detected object based on the height map and performing three-dimensional reconstruction of the detected object;

[0091] Furthermore, the four unobstructed real height area height maps are fused to obtain a complete and accurate height map, and the sub-steps of obtaining the point cloud data of the detected object according to the height map and performing three-dimensional reconstruction of the detected object are as follows:

[0092] Step S51, fusing four unobstructed real height area height maps to obtain a complete and accurate height map;

[0093] Obtain the pixel point sets of four unobstructed real height area height maps, and fuse the pixels representing the same point in the four height maps through weighted processing to generate clearer and more accurate pixel points, and then generate a complete and accurate height map through each pixel point set.

[0094] Step S52, generating point cloud data of the detected object according to the height map;

[0095] Specifically, detailed data of each point is obtained based on the height map and the pixel point set, including but not limited to the coordinate information, state information, and height information of each point, and the detailed data of each point is fused to generate point cloud data of the detected object.

[0096] Step S53, performing three-dimensional reconstruction of the detection object according to the point cloud data of the detection object;

[0097] Specifically, a three-dimensional reconstruction model of the inspection object is generated according to the point cloud data of the inspection object by using point cloud modeling technology, and the inspection object is inspected according to the three-dimensional reconstruction model.

[0098] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made on the basis of the technical solution of the present invention should be included in the scope of protection of the present invention.

Claims

1. A single-camera multi-projection dual-axis projection structured light 3D reconstruction measurement system, characterized in that: include: A main support (1), a structured light projector group (2) and a high-speed industrial acquisition camera (3); A structured light projector bracket (11) is fixedly mounted on the periphery of the main bracket (1), and the structured light projector group (2) is fixed to the periphery of the main bracket (1) via the structured light projector bracket (11); a camera bracket (12) is fixedly mounted on the upper portion of one side of the main bracket, and a high-speed industrial acquisition camera (3) is fixed to the inside of the main bracket (1) via the camera bracket; The structured light projector group (2) comprises four structured light projectors, and the four structured light projectors are all rotated 45 degrees along the optical axis of their respective projection lenses and then fixed on corresponding structured light projector brackets (11); The high-speed industrial acquisition camera (3) is fixed inside the main bracket (1), a high-precision double telecentric lens (31) is installed at the lower end, and the detection object is placed directly below the high-precision double telecentric lens (31); Four structured light projectors sequentially project encoded structured light grating stripes at ±45 degrees onto the object to be tested. After the structured light grating stripes reach the object to be tested, they form dual-axis phase-encoded stripes. A high-speed industrial acquisition camera (3) photographs the dual-axis phase-encoded stripes on the object to be tested through a high-precision dual-telecentric lens (31) to form a dual-axis phase-encoded stripe image. Structured light three-dimensional reconstruction is performed based on the dual-axis phase-encoded stripe image to obtain the three-dimensional morphology of the object to be tested. In order to make the phase-encoded fringes on the surface of the detected object clear in full frame with equal width and projected in two axes at ±45 degrees, the fringe correction technology is used to correct the two-axis phase-encoded fringes; Among them, the fringe correction technology optimizes the rotation degree and fringe width of the dual-axis phase-encoded fringe to the optimal value, and adjusts the dual-axis phase-encoded fringe to a full-frame clear width with equal width and dual-axis projection of ±45 degrees. The stripe correction technology is implemented as follows: Based on the rotation angle of the structured light projector through the function Adjust the rotational projection angle of the dual-axis phase-encoded fringes, where: is the rotation adjustment function, For the The number of degrees before the stripes rotate, For the The degree of rotation of the stripes, The value range is , is the number of phase-encoded fringes, is the rotation degree of the structured light projector, For the The rotation of the stripes affects the adjustment characteristics. is the rotation adjustment value, Adjust the index for fringe rotation, Adjust the exponent for the rotation angle; By formula The clear fringe width of the dual-axis phase-encoded fringe is calculated, where: is the optimal stripe width, is the wavelength of the structured light, To select the optimal distance from the object to the high-precision bi-telecentric lens, To select the optimal distance from the object to the high-speed industrial acquisition camera, Select the optimal distance between two adjacent stripes; Based on the optimal stripe width Through the function Adjust the stripe width, where is the width adjustment function, For the Stripe width, The value range is , is the total number of phase-encoded fringes, is the optimal stripe width, Adjust the eigenvalue for the stripe width, The scaling exponent that adjusts the feature value for the fringe width.

2. The single-camera multi-projection dual-axis projection structured light 3D reconstruction measurement system according to claim 1, characterized in that: Each structured light projector in the structured light projector group (2) is rotated 45 degrees to project structured light grating stripes onto the detection object, so that a ±45 degree dual-axis phase-encoded stripe image is presented on the surface of the detection object.

3. The single-camera multi-projection dual-axis projection structured light 3D reconstruction measurement system according to claim 1, characterized in that: The structured light projector group (2) comprises a first structured light projector (21), a second structured light projector (22), a third structured light projector (23) and a fourth structured light projector (24); during measurement, the first structured light projector (21) is first turned on according to a rotation angle to project structured light grating stripes onto the detection object, and the structured light grating stripes present a dual-axis phase-coded stripe image in the projection direction of the first structured light projector (21) on the surface of the detection object, and then a high-speed industrial acquisition camera (3) acquires the first dual-axis phase-coded stripe image through a high-precision dual telecentric lens (31), and then The second structured light projector (22), the third structured light projector (23) and the fourth structured light projector (24) are sequentially turned on according to the rotation angle to sequentially project structured light grating stripes in corresponding directions onto the detection object, and the structured light grating stripes sequentially present dual-axis phase-coded stripe images in the projection directions of the second structured light projector (22), the third structured light projector (23) and the fourth structured light projector (24) on the surface of the detection object, and the high-speed industrial acquisition camera (3) sequentially acquires the second, third and fourth dual-axis phase-coded stripe images through a high-precision dual telecentric lens (31).

4. A single-camera multi-projection dual-axis projection structured light 3D reconstruction measurement method, characterized in that: The method is applied to the dual-axis projection structured light three-dimensional reconstruction measurement system according to any one of claims 1 to 3, and the method comprises: Step S1, using four structured light projectors to sequentially project structured light onto the surface of the detection object, and sequentially generating four full-frame clear biaxial phase-coded fringe images with equal width and biaxial projection of ±45 degrees; Step S2, using a high-speed industrial acquisition camera and a high-precision dual telecentric lens to sequentially acquire four dual-axis phase-coded fringe images; Step S3, reconstructing four height maps of the detection object according to the four dual-axis phase-encoded fringe images; Step S4, according to the installation position of the structured light projector group and the four collected dual-axis phase-encoded fringe images, obtain the blocked noise area and the unblocked real height area in the four height maps and remove the noise in the blocked area to obtain four unblocked real height area height maps; Step S5: fuse the four unobstructed real height area height maps to obtain a complete and accurate height map, obtain point cloud data of the detected object based on the height map, and perform three-dimensional reconstruction of the detected object.

5. The single-camera multi-projection dual-axis projection structured light 3D reconstruction measurement method according to claim 4, characterized in that: The sub-steps of using four structured light projectors to sequentially project structured light onto the surface of the object to be detected, and sequentially generating four full-frame clear biaxial phase-coded fringe images with equal width and biaxial projection at ±45 degrees, are as follows: According to the system setting, four structured light projectors are fixed and rotated 45 degrees according to the structured light projector fixing bracket to project structured light grating stripes and present a ±45 degree dual-axis phase coded stripe image on the surface of the detection object; The dual-axis phase-encoded stripes are corrected by fringe correction technology to make them appear as full-frame clear dual-axis phase-encoded stripes with equal width and dual-axis projection of ±45 degrees.

6. The single-camera multi-projection dual-axis projection structured light 3D reconstruction measurement method according to claim 5, characterized in that: The sub-steps of using a high-speed industrial acquisition camera and a high-precision bi-telecentric lens to sequentially acquire four dual-axis phase-coded fringe images are as follows: Calculate the placement parallel deviation value according to the placement position of the high-speed industrial acquisition camera and the detection object, obtain the wrong placement device according to the placement parallel deviation value and adjust its position so that the adjusted high-speed industrial acquisition camera and the detection object are in a horizontal position; Calculate the placement vertical deviation value according to the placement position of the high-precision dual telecentric lens and the detection object, obtain the wrong placement device according to the placement vertical deviation value and adjust its position so that the adjusted high-precision dual telecentric lens and the detection object are in a vertical position; Use an adjusted height industrial acquisition camera and a high-precision dual telecentric lens to sequentially acquire four dual-axis phase-coded fringe images of the inspection object at four angular directions.

7. The single-camera multi-projection dual-axis projection structured light 3D reconstruction measurement method according to claim 6, characterized in that: The sub-steps of reconstructing four height maps of the detected object based on four dual-axis phase-encoded fringe images are as follows: Calculate image phase values ​​based on four dual-axis phase-encoded fringe images; The height of the detected object is calculated according to the phase value of each point of the phase-coded fringe image and four height maps of the detected object are reconstructed.

8. The single-camera multi-projection dual-axis projection structured light 3D reconstruction measurement method according to claim 7, characterized in that: According to the installation position of the structured light projector group and the four collected dual-axis phase-encoded fringe images, the blocked noise area and the unblocked true height area in the four height maps are obtained and the noise in the blocked area is removed. The sub-steps of obtaining the four unblocked true height area height maps are as follows: Obtain shadow points and high-reflection points based on the installation positions of each structured light projector and mark them in four height maps; The switch threshold method is used to identify the occluded noise area and the unoccluded true height area in the four height maps; The marked points in the dual-axis phase-encoded fringe image and the height map are removed from the corresponding four height maps to obtain four unobstructed real height area height maps.

9. The single-camera multi-projection dual-axis projection structured light 3D reconstruction measurement method according to claim 8, characterized in that: The four unobstructed real height area height maps are fused to obtain a complete and accurate height map. The sub-steps of obtaining the point cloud data of the detected object based on the height map and performing 3D reconstruction of the detected object are as follows: The four unobstructed real height area height maps are merged to obtain a complete and accurate height map; Generate point cloud data of the detected object based on the height map; The object to be inspected is reconstructed in three dimensions based on its point cloud data.

Images