Monocular vision pose measurement method and system based on circular feature target point

Abstract

The application discloses a monocular vision pose measurement method and system based on circular feature target points, and comprises the following steps: camera calibration, image acquisition, image preprocessing, feature point detection and matching; the Hough circle detection is adopted to extract the center two-dimensional pixel coordinates and radius of the circular feature target points, and the unique relationship between the feature target points is obtained according to the center coordinates and the radius information; pose measurement, the rotation matrix R and the translation vector T of the target coordinate system conversion to the camera coordinate system in each collected pose image are obtained according to the PnP algorithm; and the relative pose is solved. The application uses a specially designed cooperative target containing five circular feature target points as a target object, combines a fast detection algorithm realized according to the target characteristics, realizes the center matching and sequencing, and improves the precision. The method has the advantages of simple structure, simple camera calibration, large measurement field of view, strong real-time performance and the like, and also avoids the defects of small field of view and difficult stereo matching in stereo vision.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of visual detection, and particularly relates to a monocular vision pose measurement method and system based on a circular feature target point. BACKGROUND

[0002] The aviation turbine shaft engine is an aviation engine which outputs shaft power to the outside through a power output shaft. The coupling is a mechanical part which firmly connects the driving shaft and the driven shaft in different mechanisms to rotate together and transmits motion and torque, and is an important power transmission part of the small and medium-sized aviation turbine shaft engine, and directly affects the power performance of the engine. At present, there are various couplings, which can be classified in different ways according to different principles. Generally, the coupling is divided into rigid coupling and flexible coupling. Unlike the rigid coupling, the flexible coupling has great elasticity, and can not only transmit torque to the driving shaft and the driven shaft, but also has certain tolerance for the inconsistent motion of the driving shaft and the driven shaft on other degrees of freedom except the axial rotation, and will not accumulate excessive stress to cause the coupling to break or be damaged, so it is an important power transmission part of the engine.

[0003] Based on the principle of elasticity, there is a one-to-one correspondence between the relative displacement and rotation of the two ends of the flexible coupling and the load. Therefore, the motion data of the other five degrees of freedom except the axial rotation can represent the load performance of the flexible coupling which can be accommodated by the driving shaft and the driven shaft during the torque transmission process, and is an important performance index of the flexible coupling, and the corresponding measuring equipment is needed for data acquisition and analysis. At the same time, although there are various existing pose measurement methods, the application objects of most of them are large-scale targets, such as the real-time measurement of the end pose of industrial equipment such as machine tools and robots, and the laser measuring equipment is used in the traditional method. However, the internal space of the aviation engine is small, and at present there is a lack of unified measuring means for small objects in a compact environment. SUMMARY

[0004] The purpose of the application is to overcome the defects of the prior art and provide a monocular vision pose measurement method and system based on a circular feature target point.

[0005] The purpose of the application can be achieved by the following technical solutions:

[0006] The application provides a monocular vision pose measurement method based on a circular feature target point, which comprises the following steps:

[0007] Step S1: fix the target on the measured target surface, fix the monocular camera on the static platform corresponding to the target, obtain and record the monocular camera parameters, the target includes five high-reflective circular target points, one of which is located at the center of the target, and the other four are located at the upper left, lower left, upper right and lower right of the center target point, respectively, and the five high-reflective circular target points form a completely central symmetric pattern, the four high-reflective circular target points have the same diameter, and the center high-reflective circular target point is 0, the upper left corner is 1, the upper right corner is 2, the lower right corner is 3, and the lower left corner is 4;

[0008] Step S2: control the illumination unit to illuminate the target, and use the camera to obtain an image sequence of the target;

[0009] Step S3: pre-process the obtained image sequence;

[0010] Step S4: detecting, matching and sorting the feature target points of the target based on the target information after pre-processing the image sequence, obtaining the two-dimensional pixel coordinate information of all feature target points, including: obtaining the center two-dimensional pixel coordinates and radius information of all circles in the image through the Hough circle detection algorithm, comparing the radius of all circles, taking the circle with the largest radius as the center circular target point, i.e. 0, calculating the distance between all circle centers and the origin of the pixel coordinate system, taking the circle with the smallest distance as 1, and the circle with the largest distance as 3, calculating the slope of the line connecting all circle centers and the origin of the pixel coordinate system, taking the circle with the largest slope as 4, and the circle with the smallest slope as 2, obtaining the two-dimensional pixel coordinate information and corresponding number of all feature target point circles;

[0011] Step S5: obtaining the three-dimensional coordinate information of the feature target points in the target coordinate system according to the target design parameters, and calculating the pose relationship of each image through PnP algorithm according to the two-dimensional pixel coordinate information, three-dimensional coordinate information and monocular camera parameters;

[0012] Step S6: calculating the relative pose results of adjacent time images through matrix multiplication and matrix inverse operation according to the pose relationship of adjacent time images.

[0013] Further, the diameter of the center target point is n times the diameter of the four high-reflective circular target points, where n is greater than 1.

[0014] Further, in step S1, the monocular camera is fixed on the static platform corresponding to the target, and the target is perpendicular to the optical axis of the monocular camera.

[0015] Further, in step S1, the monocular camera parameters are obtained and recorded, including camera intrinsic matrix K and camera distortion coefficient D, wherein the camera intrinsic matrix K is:

[0016]

[0017] wherein f is the focal length of the camera, dx, dy are intrinsic parameters of the camera sensor, representing the actual size of each pixel, u0, v0 are the principal point horizontal and vertical coordinates, representing the offset of the image coordinate system origin relative to the pixel coordinate system origin;

[0018] The camera distortion coefficient D is:

[0019]

[0020] wherein k1, k2 and k3 are the first, second and third order coefficients of radial distortion respectively, and p1 and p2 are the first and second order coefficients of tangential distortion respectively.

[0021] Further, the lighting unit in step S2 is two LED lighting lamps, and the LED lighting lamps are installed on both sides of the monocular camera.

[0022] Further, the step S3 preprocessing includes: using the camera intrinsic parameters and the distortion coefficient to correct the image distortion, carrying out gray scale on the corrected image, and using Gaussian filter to filter the gray scale image.

[0023] Further, the pixel coordinate system origin is the first pixel point at the top left corner of the image.

[0024] Further, the step S5 specifically includes: obtaining three-dimensional coordinates of all circle centers in a target coordinate system according to target design parameters, wherein the target coordinate system origin is the circle center of the central circular target point, the X-axis of the coordinate system passes through the midpoint of the connecting line between the second and third circle centers, the XOY plane is located on the upper surface of the target, and the Z-axis is perpendicular to the upper surface of the target, the target design parameters are the circle center positions and radii of all circles, a matrix P o is composed according to the three-dimensional coordinates of all circle centers, a matrix P u is composed according to the two-dimensional pixel coordinates of all circles obtained in step S4, a coordinate matrix P o of the five feature target points in the target coordinate system, a coordinate matrix P u of the five feature target points in the pixel coordinate system, the camera intrinsic parameter matrix K and the camera distortion coefficient D are substituted into the formula:

[0025] [R, T] = PnP(P o , P u , K, D)

[0026] The pose relationship of the image is obtained: the rotation matrix R and the translation vector T.

[0027] Further, the step S6 calculation process includes: at time 1, the rotation matrix R1 and the translation vector T1 between the camera coordinate system and the target coordinate system can be obtained by PnP algorithm, at time 2, the rotation matrix R2 and the translation vector T2 between the camera coordinate system and the target coordinate system can be obtained by the same method, the transformation matrix of the target coordinate system at time 1 to the camera coordinate system is And the transformation matrix of the target coordinate system at time 2 to the camera coordinate system is

[0028]

[0029] There is a relationship between the transformation matrices:

[0030]

[0031] Then according to the formula:

[0032]

[0033] The relative pose result can be obtained through matrix multiplication, matrix inversion operation, including the rotation matrix R and the translation vector T:

[0034] R=R1 · R2

[0035] T=R1 · (T2-T1).

[0036] The application also provides a monocular vision pose measurement system based on a circular feature target point, which comprises a monocular camera, two LED lighting units, a target and a processor, the processor is connected with the monocular camera and is used for executing the above method.

[0037] Compared with the prior art, the application has the following advantages:

[0038] (1) The application uses a specially designed target as a target object, combines a fast detection algorithm realized according to the target characteristics, realizes the center matching sorting, and improves the precision.

[0039] (2) The application has the advantages of simple structure, simple camera calibration, large measurement field of view, strong real-time performance and the like.

[0040] (3) The application avoids the shortcomings of small field of view and difficult stereo matching in stereo vision.

[0041] (4) The present application has high measurement accuracy. The position and posture of the target in the camera coordinate system can be obtained by using a specially designed target, which has high accuracy in small-range measurement, and the measurement accuracy can reach 0.2mm. The specially designed target can be fixed on the measurement target, and can be widely used in modern industrial equipment (machine tools, industrial mechanical arms) measurement, meeting the demand of modern industrial equipment for small-range high-precision pose measurement. Compared with traditional measurement equipment, the device cost of the method is low, and the pose measurement accuracy is high. BRIEF DESCRIPTION OF DRAWINGS

[0042] Figure 1 The method flowchart of the present application is shown in the figure;

[0043] Figure 2 The system schematic diagram of the present application is shown in the figure;

[0044] Figure 3 The specially designed target of the present application is shown in the figure;

[0045] Figure 4 The detection matching sorting algorithm flowchart of the present application is shown in the figure;

[0046] Figure 5 The coordinate system pose relationship diagram of the present application is shown in the figure. DETAILED DESCRIPTION

[0047] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should belong to the protection scope of the present application.

[0048] Embodiment 1:

[0049] The present application provides a monocular vision pose measurement method based on circular feature target points, as shown in the figure, which comprises: Figure 1

[0050] Step S1: camera calibration before pose measurement.

[0051] Step S2: obtaining the image sequence of the cooperative target.

[0052] Step S3: preprocessing the collected images.

[0053] Step S4: matching and sorting the five circular feature target points based on the center coordinates and radius information.

[0054] Step S5: obtaining the pose relationship of the target coordinate system converted to the camera coordinate system by PnP algorithm.

[0055] ​Step S6: Obtain the relative pose of the target coordinate system in the image sequence by the Euclidean transformation relationship.

[0056] Further, the monocular vision pose measurement method based on the circular feature target point comprises the following steps: S1: Obtain and record the camera parameters. Before the vision measurement, the camera parameters are calibrated by using a high-precision circular point calibration board, and the camera intrinsic matrix K is recorded:

[0057]

[0058] Wherein f is the focal length of the camera, dx and dy are the inherent parameters of the camera sensor, representing the actual size of each pixel, u0 and v0 are the horizontal and vertical coordinates of the principal point, representing the offset of the image coordinate system origin relative to the pixel coordinate system origin; record the distortion coefficient D of the camera:

[0059]

[0060] Wherein k1, k2 and k3 are the first, second and third order coefficients of radial distortion, respectively, and p1 and p2 are the first and second order coefficients of tangential distortion, respectively.

[0061] S2: Real-time image acquisition of the cooperative target using the camera. The cooperative target is specially designed, containing 5 circular feature target points, with a high-reflective circular target point at the center and 4 high-reflective circular target points symmetrically and uniformly distributed around it. The diameter of the central circular target point is twice that of the surrounding circular target points; the origin O of the target coordinate system is located at the center of the central circular target point, the XOY plane is located on the upper surface of the target, and the Z axis is perpendicular to the upper surface of the target.

[0062] S3: Preprocess the collected image. Use the camera calibration parameters to correct the image distortion; then grayscale the image; finally, filter the image using a filter.

[0063] S4: Match and sort the 5 circular feature target points based on the center coordinates and radius information. Detect all circular features in the cooperative target image to obtain the two-dimensional pixel coordinates and radius of all circular features. Identify the central circular target point as the 0th circle by the maximum radius; identify the surrounding circular target points as the 1st, 3rd, 2nd, 4th and 5th circles by the minimum, maximum, minimum, maximum and minimum distances from the pixel coordinate system origin, respectively; and match and sort the 5 circles in the order of 0th to 4th.

[0064] S5: The pose relationship of the target coordinate system converted to the camera coordinate system is solved by the PnP algorithm. Using the PnP algorithm in visual measurement, the coordinate matrix P of the 5 feature targets on the cooperative target corresponding to the target coordinate system o , the coordinate matrix P of the 5 feature targets corresponding to the pixel coordinate system u , the camera intrinsic parameter matrix K and the camera distortion coefficient D are used as the PnP algorithm parameters Formula (1), to obtain the pose relationship of the target coordinate system relative to the camera coordinate system in a single image: the rotation matrix (R), the translation vector (T).

[0065] [R, T] = PnP(P o , P u , K, D) (1).

[0066] S6: The relative pose of the target coordinate system between the image sequences is obtained by the Euclidean transformation relationship. According to the pose relationship of the target coordinate system relative to the camera coordinate system in a single image, the relative pose relationship of the target coordinate system between two images is obtained using the Euclidean transformation principle.

[0067] Example 2:

[0068] The part not mentioned in this embodiment is the same as in Example 1.

[0069] The system schematic diagram of this embodiment is shown in Figure 2 The target is fixed on the surface of the measured object, and the monocular camera is fixed on the static platform. The LED lighting unit installed around the camera illuminates the target, and the five high-reflectivity marker points installed on the surface of the target reflect the incident light back into the camera field of view for imaging. When the measured object moves, the target moves, and the camera takes real-time images of the target to obtain target images at the same time interval. After image feature processing, effective feature point information is extracted, and then a pose solving algorithm is used to obtain the pose information of the target.

[0070] The monocular camera is a Mercury 2 digital camera of Daheng Image Company, model MER2-502-79U3M, resolution 2448x2048, data interface USB3.0, and can be powered by USB; the lens is a HN-P-20M series fixed focus lens of Daheng Image Company, model HN-P-2524-20M-C1.1 / 1, focal length 25mm, field of view angle 37.31°x30.47°x22.53°, and the lens and the camera are connected through a C-type interface.

[0071] The proposed method is implemented on the Windows operating system using the programming language C++, which also uses some data structures and common image processing algorithms of OpenCV, and the software environment is Visual Studio 2017, OpenCVv4.5.1.

[0072] The hardware environment is Intel i7-7700HQ processor, 16GB running memory.

[0073] This example uses a specially designed target, as shown in the target schematic Figure 3 The diameter of the central circular target is 16mm, the diameter of the surrounding four circular targets is 8mm, and the distance between adjacent targets is 34mm. The five circular targets form a completely central symmetric pattern. Figure 3 The XYZ coordinate system in the figure is the target coordinate system, with the origin O at the center of the central circular target, the X-axis passing through the midpoint of the line connecting the centers of the 2nd and 3rd targets, the XOY plane on the upper surface of the target, and the Z-axis perpendicular to the upper surface of the target.

[0074] S1: Before the pose measurement, use the high-precision circular point calibration board shown in Table 1 to calibrate the camera used according to the Zhang Zhengyou camera calibration method.

[0075] Table 1 Circular point calibration board related parameters

[0076]

[0077] The camera intrinsic matrix is obtained as:

[0078]

[0079] The camera distortion coefficient is:

[0080]

[0081] S2: Use the camera to obtain an image containing the target, with the target perpendicular to the camera optical axis as much as possible and in the clear imaging position of the camera, and the image occupying more than 1 / 5 of the entire frame.

[0082] S3: First, use the camera intrinsic and distortion coefficients to correct the image distortion; then grayscale the corrected image; finally, use Gaussian filtering to filter the grayscale image. According to the test, the Gaussian filter size is selected as (9, 9), the X-direction Gaussian filter standard deviation is selected as 2, and the Y-direction Gaussian filter standard deviation is selected as 2. At this time, the noise removal effect and the clarity of the filtered image are balanced, and the filtering effect is good.

[0083] S4: Detect, match and sort the five circular feature targets in the preprocessed image, and the process is shown in Figure 4The center coordinates (u, v) and radius information of all circles in the picture are detected and extracted. According to the test, the resolution is inversely proportional to the resolution of the image, and the minimum distance between the two circle centers in the detection result is selected as 300, the larger value of the two threshold values passed to the Canny edge detector is selected as 100, the accumulator threshold value for detecting circles is selected as 20, the minimum radius for detecting circles is selected as 50, and the maximum radius for detecting circles is selected as 160, so that 5 circular feature target points can be accurately detected.

[0084] By comparing the radii of all circles, the circle with the largest radius is selected, which is No. 0.

[0085] In the collected image, the origin of the pixel coordinate system is the first pixel point in the upper left corner of the image, and the distance between the extracted circle centers and the origin of the pixel coordinate system is calculated, and the distance calculation formula is as follows:

[0086]

[0087] Among them, the farthest is No. 3, and the closest is No. 1.

[0088] The slope of the line connecting all the extracted circle centers and the origin of the pixel coordinate system is calculated, and the slope calculation formula is as follows:

[0089] k = u / v (2)

[0090] Among them, the largest slope is No. 4, and the smallest slope is No. 2.

[0091] Since the angle of rotation of the target around the Z axis is very small in actual measurement, the corresponding relationship between the sorted two-dimensional circle centers in the image and the three-dimensional circle centers in the target coordinate system is uniquely determined.

[0092] S5: The XOY plane of the target coordinate system is located on the upper surface of the target, so for the circle center three-dimensional coordinates (X, Y, Z) of the five circular feature targets, Z = 0. The coordinate array of the five circle centers from No. 0 to No. 4 in the target coordinate system can be obtained from the target design parameters, which are (0, 0, 0), (-17, 17, 0), (17, 17, 0), (17, -17, 0), and (-17, -17, 0) in order, arranged as a 5x3 matrix P o .

[0093] According to the detection and matching sorting algorithm, the coordinate array of the five circle centers from No. 0 to No. 4 in the pixel coordinate system can be obtained, which is arranged as a 5x2 matrix P u .

[0094] The PnP (Perspective-n-Point) algorithm in visual measurement is used to obtain the coordinate matrix P oThe coordinate matrix P of the five feature target points in the pixel coordinate system u Substitute the camera intrinsic parameter matrix K and the camera distortion coefficient D into formula (3) to obtain the pose relationship between the target coordinate system and the camera coordinate system in a single image: rotation matrix (R) and translation vector (T).

[0095] [R,T]=PnP(P o ,P u ,K,D) (3).

[0096] S6: As Figure 5 As shown, at time 1, the target coordinate system is O1-X1Y1Z1, and at time 2 after the target has moved, its coordinate system becomes O2-X2Y2Z2, while the camera coordinate system is O2-X1Y1Z1. c -X c Y c Z c .

[0097] At time 1, the rotation matrix R1 and translation vector T1 between the camera coordinate system and the target coordinate system can be obtained by the PnP algorithm.

[0098] At time 2, the rotation matrix R2 and translation vector T2 between the camera coordinate system and the target coordinate system can be obtained by the same method.

[0099] The transformation matrix of the target coordinate system from time 2 to time 1 It can be represented as:

[0100]

[0101] Similarly, the transformation matrix from the target coordinate system at time 1 to the camera coordinate system can be obtained. Transformation matrix from target coordinate system to camera coordinate system at time 2

[0102]

[0103] There are relationships between the transformation matrices:

[0104]

[0105] According to formula (7), we have:

[0106]

[0107] The relative pose result, including the rotation matrix R and translation vector T, can be obtained through matrix multiplication and matrix inversion.

[0108] R = R1 · R2 (9)

[0109] T = R1· (T2-T1) (10)

[0110] An experimental platform is built to verify the monocular vision pose measurement system and method based on a circular feature target.

[0111] The translational range is ±10mm, and the movement is performed along the X, Y and Z axes with a step of 1mm, respectively. Each position is stopped for several seconds, and the image at the current pose is captured by the monocular camera. The rotational range is ±5°, and the rotation is performed along the X, Y and Z axes with a step of 1°, respectively. Each position is stopped for several seconds, and the image at the current pose is captured by the monocular camera.

[0112] The measurement value is obtained by image solving, and the measurement error can be obtained by comparing the theoretical value with the measurement value.

[0113] The error indicators in each direction are shown in Tables 2 and 3.

[0114] Table 2 translational error indicators

[0115]

[0116] Table 3 rotational error indicators

[0117]

[0118] The experimental results show that the translational measurement accuracy of the system instance is better than 0.2mm, and the rotational measurement accuracy is better than 0.2°. The measurement accuracy meets the actual demand.

[0119] The above is only a specific embodiment of the present application, but the protection scope of the present application is not limited thereto. Any skilled person in the art can easily think of various equivalent modifications or replacements within the technical range disclosed by the present application, and these modifications or replacements should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A monocular vision pose measurement method based on a circular feature target point, characterized in that, The method comprises the following steps: Step S1: fixing a target on a measured target surface and fixing a monocular camera on a stationary platform corresponding to the target, acquiring and recording monocular camera parameters, the target comprising five high-reflectivity circular target points, one of which is located at the center of the target, and the other four are located at the upper left, lower left, upper right and lower right of the center target point, the five high-reflectivity circular target points forming a completely central symmetric pattern, the other four high-reflectivity circular target points having the same diameter, the center high-reflectivity circular target point being No. 0, the upper left corner being No. 1, the upper right corner being No. 2, the lower right corner being No. 3, and the lower left corner being No. 4; Step S2: controlling the illumination unit to illuminate the target, and using the camera to acquire an image sequence of the target; Step S3: preprocessing the acquired image sequence; Step S4: detecting, matching and sorting the feature target points of the target based on the target information of the preprocessed image sequence, to obtain two-dimensional pixel coordinate information of all the feature target points, specifically comprising: acquiring the center two-dimensional pixel coordinates and radius information of all the circles in the image through the Hough circle detection algorithm, comparing the radius of all the circles, taking the circle with the largest radius as the center circular target point, i.e. No. 0, calculating the distance between the center of all the circles and the origin of the pixel coordinate system, taking the circle with the smallest distance as No. 1, and the circle with the largest distance as No. 3, calculating the slope of the line connecting the center of all the circles and the origin of the pixel coordinate system, taking the circle with the largest slope as No. 4, and the circle with the smallest slope as No. 2, to obtain the two-dimensional pixel coordinate information and corresponding number of all the feature target point circles; Step S5: obtaining three-dimensional coordinate information of the feature target points in the target coordinate system according to the target design parameters, and calculating the pose relationship of each image through the PnP algorithm according to the two-dimensional pixel coordinate information, the three-dimensional coordinate information and the monocular camera parameters; Step S6: calculating the relative pose results of adjacent time images through matrix multiplication and matrix inversion operation according to the pose relationship of adjacent time images; The step S5 specifically comprises: obtaining three-dimensional coordinates of all circle centers in a target coordinate system according to target design parameters, wherein the origin of the target coordinate system is the circle center of the central circular target point, the coordinate system is The axis passes through the midpoint of the line connecting the circle centers of No. 2 and No. 3, The plane is located on the upper surface of the target, The axis is perpendicular to the upper surface of the target, and the target design parameters are the circle center positions and radii of all circles. According to the two-dimensional pixel coordinates of all circles obtained in step S4, a matrix is formed The coordinates of the five characteristic target points in the target coordinate system are substituted into the formula: The coordinates of the five characteristic target points in the pixel coordinate system are substituted into the formula: The camera intrinsic parameter matrix K and the camera distortion coefficient D are substituted into the formula: Obtaining pose relationship of images: rotation matrix , translation vector .

2. The monocular vision pose measurement method based on a circular feature target according to claim 1, characterized in that, The diameter of the center target point is n times the diameter of the four high-reflectivity circular target points, where n is greater than 1.

3. The monocular vision pose measurement method based on a circular feature target according to claim 1, characterized in that, In step S1, the monocular camera is fixed on a stationary platform corresponding to the target, and the target is perpendicular to the optical axis of the monocular camera.

4. The monocular vision pose measurement method based on a circular feature target according to claim 1, characterized in that, In step S1, the monocular camera parameters include a camera intrinsic matrix K and a camera distortion coefficient D, wherein the camera intrinsic matrix K is: wherein, is the focal length of the camera, , is an intrinsic parameter of the camera sensor representing the actual size of each pixel, , is the principal point horizontal and vertical coordinates representing the offset of the image coordinate system origin with respect to the pixel coordinate system origin; The camera distortion coefficient D is: wherein , and are the first, second and third order coefficients of radial distortion, respectively, and are the first and second order coefficients of tangential distortion, respectively.

5. The monocular vision pose measurement method based on a circular feature target according to claim 1, characterized in that, The illumination unit in step S2 is two LED illumination lamps, which are installed on both sides of the monocular camera.

6. The monocular vision pose measurement method based on a circular feature target according to claim 1, characterized in that, The preprocessing in step S3 includes: using the camera intrinsic and distortion coefficients to correct the image, performing grayscale on the corrected image, and using Gaussian filtering to filter the grayscale image.

7. The monocular vision pose measurement method based on a circular feature target according to claim 1, characterized in that, The origin of the pixel coordinate system is the first pixel point at the upper left corner of the image. 8.The monocular vision pose measurement method based on a circular feature target according to claim 1, characterized in that, The step S6 calculation process includes: at time 1, the rotation matrix between the camera coordinate system and the target coordinate system can be solved by PnP algorithm and the translation vector , at time 2, the rotation matrix between the camera coordinate system and the target coordinate system can be solved by the same method and the translation vector , the transformation matrix of the target coordinate system at time 1 to the camera coordinate system and the transformation matrix of the target coordinate system at time 2 to the camera coordinate system : There is a relationship between the transformation matrices: According to the formula: The relative pose results can be obtained through matrix multiplication and matrix inversion operation, including rotation matrix and translation vector : 。 9. A monocular vision pose measurement system based on circular feature targets, characterized in that The system comprises a monocular camera, two LED illumination units, a target and a processor, the processor being connected with the monocular camera and used for executing the method in any one of claims 1-8.

Images