A method and system for verifying the accuracy of triangulation

Abstract

The application discloses a kind of for verifying the method and system of triangulation accuracy, comprising: target A and target B are erected to arbitrary distance, actual distance between target A and target B is measured by interferometric distance meter using laser;Drive the camera camera of triangulation system to shoot the bearing diagram containing target A and target B, obtain target bearing diagram set;Extract the feature point of target A and target B in target bearing diagram set, respectively obtain first feature point set and second feature point set;According to the internal and external parameter data of camera camera, the Euclidean distance between first feature point set and second feature point set is estimated, and the Euclidean distance set is obtained;The Euclidean distance in Euclidean distance set is compared with actual distance respectively, if the error between the two is less than threshold value, then consider that the measurement accuracy of triangulation system is up to standard.The application is used to solve the technical problem that the existing calibration method is insufficient in calibration accuracy, so as to further improve the ranging accuracy of triangulation system.

Description

Technical Field

[0001] This invention relates to the field of triangulation technology, and more specifically to a method and system for verifying the accuracy of triangulation. Background Technology

[0002] Currently, non-contact measurement techniques for measuring distances between two points mainly include infrared ranging, laser ranging, interferometric ranging, ultrasonic ranging, and triangulation ranging. Infrared and laser ranging cannot yet achieve sub-millimeter accuracy and are easily affected by the external environment. Interferometric ranging can currently achieve micrometer-level measurement accuracy, but interferometric technology requires a demanding experimental environment and large equipment size, making it unsuitable for objectively limited conditions such as small target points and harsh environments. Ultrasonic ranging cannot yet achieve sub-millimeter accuracy, and its ranging accuracy is sensitive to air density and temperature.

[0003] For the reasons mentioned above, infrared ranging, laser ranging, and ultrasonic ranging methods cannot be used for sub-millimeter accuracy ranging due to insufficient accuracy. Interferometric ranging methods, which require the reflection of the measured light, are unsuitable for placing reflectors when the target is small. In this context, lens-based two-dimensional ranging systems based on triangulation show great promise. Currently, triangulation-based ranging systems are extensively studied. Compared to traditional ranging methods, triangulation has advantages such as clear principles and simple structure, and has been widely applied in fields such as robot vision, industrial automation control, and medicine.

[0004] To further improve the ranging accuracy and range of the triangulation-based ranging system, continuous calibration and debugging are required. However, due to insufficient calibration accuracy, existing calibration methods struggle to determine whether the ranging accuracy of the triangulation-based ranging system meets the standards when faced with higher precision calibration and adjustment, making it difficult to achieve high-precision calibration and debugging. Summary of the Invention

[0005] To overcome the shortcomings of existing technologies, this invention provides a method and system for verifying the accuracy of triangulation ranging, which solves the technical problem of insufficient verification accuracy in existing verification methods, thereby achieving the goal of further improving the ranging accuracy of triangulation ranging systems.

[0006] To solve the above problems, the technical solution adopted by the present invention is as follows:

[0007] A method for verifying the accuracy of triangulation includes the following steps:

[0008] Target A and target B are set up at an arbitrary distance, and the actual distance between target A and target B is measured using a laser with an interferometric rangefinder;

[0009] The camera of the triangulation system captures azimuth maps including target A and target B, thus obtaining a target azimuth map set;

[0010] Feature points of target A and target B are extracted from the target orientation map set to obtain the first feature point set and the second feature point set, respectively;

[0011] The Euclidean distance between the first feature point set and the second feature point set is estimated based on the camera's intrinsic and extrinsic parameters to obtain the Euclidean distance set.

[0012] The Euclidean distances in the Euclidean distance set are compared with the actual distances. If the errors between the two are both less than the threshold, the measurement accuracy of the triangulation system is considered to be up to standard.

[0013] In a preferred embodiment of the present invention, when driving the camera to capture a azimuth map including the target A and the target B, the following steps are included:

[0014] Drive the camera to the center of the target A and the target B, and take a picture of the orientation of the target A and the target B to obtain the orientation of the target at the center.

[0015] The target azimuth map set includes the target azimuth map at the center position.

[0016] In a preferred embodiment of the present invention, when driving the camera to capture a azimuth map including the target A and the target B, the following steps are included:

[0017] The camera is driven to capture azimuth maps containing target A and target B from different positions to obtain target azimuth maps from different positions.

[0018] The target orientation map set includes target orientation maps in different orientations.

[0019] In a preferred embodiment of the present invention, extracting feature points of target A and target B from the target orientation map set includes:

[0020] Extract the center points of target A and target B from the target orientation maps under different orientations to obtain the first center point set and the second center point set, respectively.

[0021] The first set of center points includes first center points in different orientations, and the second set of center points includes second center points in different orientations.

[0022] In a preferred embodiment of the present invention, the center dots of the target A and the target B are roughly circular after imaging, and are referred to as the first type of circle and the second type of circle.

[0023] When extracting the center point, the following is included:

[0024] Remove noise from the target orientation maps at different orientations;

[0025] Pixel-level coarse localization is performed on the edges of the first and second type of circles under different orientations to identify the center points of the target A and the target B under different orientations;

[0026] Subpixel-level fine positioning is performed on the edges of the first and second type of circles under different orientations;

[0027] Least square fitting is performed on the edge points of the first and second type of circles under different orientations to obtain the sub-pixel localization of the center points of the targets A and B under different orientations, thus completing the extraction of the center points.

[0028] In a preferred embodiment of the present invention, pixel-level coarse positioning includes:

[0029] The edges of the first and second types of circles under different orientations are extracted using the Canny operator, and the coordinates of the edge points of the first and second types of circles under different orientations are obtained.

[0030] When performing least squares fitting, the following are included:

[0031] Let m be the number of edge points obtained after sub-pixel-level fine positioning of the first and second type of circles under different orientations. Then the fitted circle curve is shown in Formula 1:

[0032]

[0033] In the formula, (x i ,y i ) represents the coordinates of the edge points of the first and second type of circles under different orientations, where F, R, and G are unknowns;

[0034] The error function is shown in Formula 2:

[0035] e i =x i F+y i R+G+a i (2);

[0036] In the formula, a i =x i +y i From m error functions, m error equations can be derived;

[0037] Based on the aforementioned F, R, and G, the third-order normal equations are shown in Equations 3 and 4:

[0038]

[0039] but

[0040] Equation 4 is solved by the adjoint matrix inversion method, as shown in Equations 5, 6, 7, and 8:

[0041] Δ=mPY+2WIQ-W 2 YI 2 P-mQ 2 (5);

[0042]

[0043]

[0044]

[0045] In the formula, P=[xx], Q=[xy], W=[x], T=[x(x 2 +y 2 )], Y=[yy], I=[y], B=[y(x 2 +y 2 )];

[0046] After obtaining F, R, and G, the coordinates of the center point and the radius are shown in Formula 9:

[0047]

[0048] In the formula, (x0, y0) are the coordinates of the center point, and r is the radius of the center point.

[0049] In a preferred embodiment of the present invention, estimating the Euclidean distance between the first feature point set and the second feature point set includes:

[0050] Based on the two-dimensional coordinates of the feature points contained in the first feature point set and the second feature point set, and according to the camera's intrinsic and extrinsic parameters, the three-dimensional coordinates of the feature points contained in the first feature point set and the second feature point set are obtained;

[0051] Based on the three-dimensional coordinates of feature points in one orientation in the first feature point set and the three-dimensional coordinates of feature points in the same orientation in the second feature point set, the Euclidean distance is estimated until the Euclidean distance of all feature points is estimated, thus obtaining the Euclidean distance set.

[0052] In a preferred embodiment of the present invention, obtaining the three-dimensional coordinates of the feature points contained in the first feature point set and the second feature point set includes:

[0053] Based on the two-dimensional coordinates of the feature points and the magnification of the camera in the x and y directions, the three-dimensional coordinates of the feature points are obtained, as shown in Formula 10:

[0054]

[0055] In the formula, Z c Let (u,v) be the straight-line distance from the camera to the target, and (f) be the coordinates of a corner point on the azimuth map. x ,f y ) represents the magnification of the camera in the x, y directions, R is the rotation matrix of the camera's extrinsic parameters, and T is the translation matrix of the camera's extrinsic parameters. w ,y w ,z w ) represents the coordinates of a corner point in a three-dimensional coordinate system.

[0056] In a preferred embodiment of the present invention, the Euclidean distance is estimated as shown in Formula 11:

[0057] D = Sqrt(power(X1-X2,2)+power(Y1-Y2,2)+power(Z1-Z2,2)) (11);

[0058] In the formula, (X1,Y1,Z1) are the three-dimensional coordinates of the feature points in the first feature point set under one orientation, and (X2,Y2,Z2) are the three-dimensional coordinates of the feature points in the second feature point set under the same orientation.

[0059] A system for verifying the accuracy of triangulation includes:

[0060] Actual distance acquisition unit: After target A and target B are set up at any distance, it is used to acquire the actual distance between target A and target B measured by laser using an interferometric rangefinder;

[0061] Azimuth map acquisition unit: After the camera of the driving triangulation system captures an azimuth map containing the target A and the target B, it is used to acquire the target azimuth map to obtain a target azimuth map set.

[0062] Feature point acquisition unit: used to extract feature points of target A and target B in the target orientation map set, and obtain the first feature point set and the second feature point set respectively;

[0063] Euclidean distance acquisition unit: used to estimate the Euclidean distance between the first feature point set and the second feature point set based on the camera's intrinsic and extrinsic parameters, and obtain the Euclidean distance set;

[0064] Accuracy judgment unit: used to compare the Euclidean distances in the Euclidean distance set with the actual distances respectively. If the error between the two is less than the threshold, the measurement accuracy of the triangulation system is considered to be up to standard.

[0065] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0066] (1) The method of the present invention provides a target A and a target B, acquires an image containing the target A and the target B through a triangulation ranging system, extracts the feature points of the target A and the target B from the image, estimates the distance between the feature points based on the intrinsic and extrinsic parameters of the camera, thereby fully testing the ranging capability of the triangulation ranging system, and finally compares its ranging result with the ranging result of the interferometric ranging instrument. With the help of the micron-level measurement accuracy of the interferometric ranging instrument, it accurately judges whether the ranging accuracy of the triangulation ranging system meets the standard, and realizes high-precision correction and debugging.

[0067] (2) The method provided by the present invention can acquire images of target A and target B in different orientations through a triangulation system, extract feature points of target A and target B from the images in different orientations, estimate the distance between feature points in different orientations based on the intrinsic and extrinsic parameters of the camera, thereby fully testing the ranging capability of the camera in different orientations of the triangulation system, and finally compare the ranging results in different orientations with the ranging results of the interferometric rangefinder, thereby accurately judging whether the ranging accuracy of the triangulation system in different orientations meets the standard by means of the micron-level measurement accuracy of the interferometric rangefinder, and realizing high-precision correction and debugging in different orientations.

[0068] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments. Attached Figure Description

[0069] Figure 1 - This is a flowchart illustrating the method steps for verifying the accuracy of triangulation according to an embodiment of the present invention;

[0070] Figure 2 - This is a schematic diagram showing the installation positions of targets A and B in an embodiment of the present invention;

[0071] Figure 3 - is a schematic diagram showing the arrangement position of the rotary cylinder in an embodiment of the present invention;

[0072] Figure 4 - is a schematic diagram of the composition structure of the triangulation system according to an embodiment of the present invention;

[0073] Figure 5 - This is a schematic diagram showing the placement of the reflector and interferometric rangefinder in an embodiment of the present invention.

[0074] Figure 6- This is a schematic diagram of the calibration chart according to an embodiment of the present invention.

[0075] The following are the reference numerals: 1. Target A; 2. Target B; 3. Test host; 4. Camera; 5. Monitor; 6. Keyboard; 7. Tri-color LED; 8. Control panel; 9. Scanner module; 10. ESD module; 11. Rotary cylinder; 12. Reflector; 13. Interferometer. Detailed Implementation

[0076] The method for verifying the accuracy of triangulation provided by this invention, such as Figure 1 As shown, it includes the following steps:

[0077] Step S1: Set up targets A1 and B2 at any distance, and use the interferometric rangefinder 13 to measure the actual distance between targets A1 and B2 using laser.

[0078] Step S2: Drive the camera 4 of the triangulation system to capture an orientation map containing targets A1 and B2, and obtain the target orientation map set;

[0079] Step S3: Extract feature points of target A1 and target B2 from the target orientation map set to obtain the first feature point set and the second feature point set respectively;

[0080] Step S4: Estimate the Euclidean distance between the first feature point set and the second feature point set based on the intrinsic and extrinsic parameter data of camera 4, and obtain the Euclidean distance set;

[0081] Step S5: Compare the Euclidean distances in the Euclidean distance set with the actual distances. If the error between the two is less than the threshold, the measurement accuracy of the triangulation system is considered to be up to standard.

[0082] Specifically, in step S1 above, the main component of the triangulation system is the test host 3, which is equipped with a camera 4 for acquiring target images. The test host 3 also includes a rotary cylinder 11 to drive the camera 4 to acquire target images from different orientations, such as... Figure 2 and Figure 3 As shown. In addition, the triangulation system also includes some auxiliary components, such as a display (5), keyboard (6), tri-color LEDs (7), control panel (8), scanning gun module (9), and ESD module (10), etc. Figure 4 As shown.

[0083] Specifically, in step S1 above, a reflector 12 is placed at the position corresponding to the center point of target A1, and an interferometric rangefinder 13 is placed at the position corresponding to the center point of target B2. The reflector 12 and the interferometric rangefinder 13 are located on the same horizontal plane. The center distance between targets A1 and B2 is measured using the interferometric rangefinder 13 and the reflector 12. Figure 5As shown.

[0084] In step S2 above, when driving camera 4 to capture a azimuth map including target A1 and target B2, the following steps are included:

[0085] Drive camera 4 to the center of target A1 and target B2, and take a picture containing the orientation of target A1 and target B2 to obtain the orientation of the target at the center.

[0086] The target orientation map set includes the orientation map of the target at the very center.

[0087] In step S2 above, when driving camera 4 to capture a azimuth map including target A1 and target B2, the following steps are included:

[0088] The camera 4 is driven to capture orientation maps containing targets A1 and B2 from different positions, thus obtaining target orientation maps from different positions.

[0089] The target orientation map set includes target orientation maps in different orientations.

[0090] Specifically, the purpose of taking target orientation maps from different angles is to test whether camera 4, after being calibrated in different positions, reaches the specified level.

[0091] In step S3 above, when extracting feature points of targets A1 and B2 from the target orientation map set, the following steps are included:

[0092] Extract the center points of targets A1 and B2 from the target orientation maps under different orientations to obtain the first center point set and the second center point set, respectively;

[0093] The first set of center points includes the first center points in different orientations, and the second set of center points includes the second center points in different orientations.

[0094] Specifically, calibration charts are attached to targets A1 and B2, and the design specifications for the calibration charts are as follows:

[0095] (1) There are specified feature points, including: corner points, round points and QR codes;

[0096] (2) The calibration chart has a reference actual distance, which can be used to calculate external parameters, such as a grid size of 300x300mm;

[0097] (3) Please refer to the implementation case. Figure 6 .

[0098] Furthermore, the central dots of targets A1 and B2 are roughly circular after imaging, and are referred to as the first type of circle and the second type of circle;

[0099] When extracting the center point, the following is included:

[0100] Remove noise from target orientation maps at different azimuth angles;

[0101] Perform pixel-level coarse localization of the edges of the first and second type of circles under different orientations, and identify the center points of targets A1 and B2 under different orientations;

[0102] Perform sub-pixel-level fine positioning of the edges of the first and second type of circles under different orientations;

[0103] Least square fitting is performed on the edge points of the first and second type circles under different orientations to obtain the sub-pixel localization of the center points of targets A1 and B2 under different orientations, thus completing the extraction of the center points.

[0104] Furthermore, when performing pixel-level coarse positioning, it includes:

[0105] The edges of the first and second types of circles under different orientations are extracted using the Canny operator, and the coordinates of the edge points of the first and second types of circles under different orientations are obtained.

[0106] When performing least squares fitting, the following are included:

[0107] Let m be the number of edge points obtained after sub-pixel-level fine positioning of the first and second type of circles under different orientations. Then the fitted circle curve is as shown in Formula 1:

[0108]

[0109] In the formula, (x i ,y i () represents the coordinates of the edge points of the first and second type circles under different orientations, where F, R, and G are unknowns;

[0110] The error function is shown in Formula 2:

[0111] e i =x i F+y i R+G+a i (2);

[0112] In the formula, a i =x i +y i From m error functions, m error equations can be derived;

[0113] The third-order normal equations are based on F, R, and G, as shown in Equations 3 and 4:

[0114]

[0115] but

[0116] Equation 4 is solved using the adjoint matrix inversion method, as shown in Equations 5, 6, 7, and 8:

[0117] Δ=mPY+2WIQ-W 2 YI 2 P-mQ 2 (5);

[0118]

[0119]

[0120]

[0121] In the formula, P=[xx], Q=[xy], W=[x], T=[x(x 2 +y 2 )], Y=[yy], I=[y], B=[y(x 2 +y 2 )];

[0122] After obtaining F, R, and G, the coordinates of the center point and the radius are shown in Formula 9:

[0123]

[0124] In the formula, (x0, y0) are the coordinates of the center point, and r is the radius of the center point.

[0125] In step S4 above, estimating the Euclidean distance between the first feature point set and the second feature point set includes:

[0126] Based on the two-dimensional coordinates of the feature points contained in the first and second feature point sets, and according to the intrinsic and extrinsic parameter data of the camera 4, the three-dimensional coordinates of the feature points contained in the first and second feature point sets are obtained.

[0127] Based on the three-dimensional coordinates of feature points in one orientation in the first feature point set and the three-dimensional coordinates of feature points in the same orientation in the second feature point set, the Euclidean distance is estimated until the Euclidean distance of all feature points is estimated, thus obtaining the Euclidean distance set.

[0128] Furthermore, when obtaining the three-dimensional coordinates of the feature points contained in the first and second feature point sets, the process includes:

[0129] Based on the two-dimensional coordinates of the feature points and the magnification of the camera 4 in the x and y directions, the three-dimensional coordinates of the feature points are obtained, as shown in Formula 10:

[0130]

[0131] In the formula, Z c Let (u,v) be the straight-line distance from camera 4 to the target, and (f) be the coordinates of a corner point on the azimuth map. x ,f y Let R be the magnification of camera 4 in the x, y directions, R be the rotation matrix of camera 4's extrinsic parameters, and T be the translation matrix of camera 4's extrinsic parameters. w ,y w ,z w ) represents the coordinates of a corner point in a three-dimensional coordinate system.

[0132] Furthermore, when estimating Euclidean distance, as shown in Formula 11:

[0133] D = Sqrt(power(X1-X2,2)+power(Y1-Y2,2)+power(Z1-Z2,2)) (11);

[0134] In the formula, (X1,Y1,Z1) are the three-dimensional coordinates of the feature points in the first feature point set under one orientation, and (X2,Y2,Z2) are the three-dimensional coordinates of the feature points in the second feature point set under the same orientation.

[0135] The system for verifying the accuracy of triangulation provided by the present invention includes: an actual distance acquisition unit, a azimuth map acquisition unit, a feature point acquisition unit, an Euclidean distance acquisition unit, and an accuracy judgment unit.

[0136] Actual distance acquisition unit: After the targets A1 and B2 are set up at any distance, it is used to acquire the actual distance between the targets A1 and B2 measured by laser using the interferometric rangefinder 13.

[0137] Azimuth map acquisition unit: After the camera 4 driving the triangulation system captures the azimuth map containing target A1 and target B2, it is used to acquire the target azimuth map to obtain the target azimuth map set.

[0138] Feature point acquisition unit: used to extract feature points of target A1 and target B2 in the target orientation map set, and obtain the first feature point set and the second feature point set respectively.

[0139] Euclidean distance acquisition unit: used to estimate the Euclidean distance between the first feature point set and the second feature point set based on the intrinsic and extrinsic parameter data of the camera 4, and obtain the Euclidean distance set.

[0140] Accuracy judgment unit: It is used to compare the Euclidean distances in the Euclidean distance set with the actual distances. If the error between the two is less than the threshold, the measurement accuracy of the triangulation system is considered to be up to standard.

[0141] The above embodiments are merely preferred embodiments of the present invention and should not be construed as limiting the scope of protection of the present invention. Any non-substantial changes and substitutions made by those skilled in the art based on the present invention shall fall within the scope of protection claimed by the present invention.

Claims

1. A method for verifying the accuracy of triangulation, characterized in that, Includes the following steps: Target A and target B are set up at an arbitrary distance, and the actual distance between target A and target B is measured using a laser with an interferometric rangefinder; The camera of the triangulation system captures azimuth maps including target A and target B, thus obtaining a target azimuth map set; Feature points of target A and target B are extracted from the target orientation map set to obtain the first feature point set and the second feature point set, respectively; The Euclidean distance between the first feature point set and the second feature point set is estimated based on the camera's intrinsic and extrinsic parameters to obtain the Euclidean distance set. The Euclidean distances in the Euclidean distance set are compared with the actual distances. If the error between the two is less than the threshold, the measurement accuracy of the triangulation system is considered to be up to standard. Wherein, the center dots of the targets A and B are roughly circular after imaging, and are referred to as the first type of circle and the second type of circle; When extracting the center point, the following is included: Remove noise from target orientation maps at different azimuth angles; Pixel-level coarse localization is performed on the edges of the first and second type of circles under different orientations to identify the center points of the target A and the target B under different orientations; Subpixel-level fine positioning is performed on the edges of the first and second type of circles under different orientations; Least square fitting is performed on the edge points of the first and second type of circles under different orientations to obtain the sub-pixel localization of the center circles of the targets A and B under different orientations, thus completing the extraction of the center circles; When performing pixel-level coarse positioning, the following is included: The edges of the first and second types of circles under different orientations are extracted using the Canny operator, and the coordinates of the edge points of the first and second types of circles under different orientations are obtained. When performing least squares fitting, the following are included: Suppose that the number of edge points obtained after sub-pixel level fine positioning of the first and second type of circles under different orientations is . The fitted circular curve is shown in Formula 1: （1）； In the formula, Let these be the coordinates of the edge points of the first and second type of circles under different orientations. , , It is an unknown; The error function is shown in Formula 2: （2）； In the formula, ,Depend on List of error functions One error equation; Based on the above , , The third-order normal equations are formed, as shown in Equations 3 and 4: （3）； but (4); Equation 4 is solved by inverting the adjoint matrix, as shown in Equations 5, 6, 7, and 8: （5）； （6）； （7）； （8）； In the formula, , , , , , , ; The above is obtained , , Then, the coordinates of the center point and the radius are as shown in Formula 9: （9）； In the formula, The coordinates of the center point are: The radius of the center point.

2. The method for verifying the accuracy of triangulation according to claim 1, characterized in that, When driving the camera to capture a azimuth map including target A and target B, the following steps are included: Drive the camera to the center of the target A and the target B, and take a picture of the orientation of the target A and the target B to obtain the orientation of the target at the center. The target azimuth map set includes the target azimuth map at the center position.

3. The method for verifying the accuracy of triangulation according to claim 1, characterized in that, When driving the camera to capture a azimuth map including target A and target B, the following steps are included: The camera is driven to capture azimuth maps containing target A and target B from different positions to obtain target azimuth maps from different positions. The target orientation map set includes target orientation maps in different orientations.

4. The method for verifying the accuracy of triangulation according to claim 3, characterized in that, When extracting feature points of target A and target B from the target orientation map set, the following steps are included: Extract the center points of target A and target B from the target orientation maps under different orientations to obtain the first center point set and the second center point set, respectively. The first set of center points includes first center points in different orientations, and the second set of center points includes second center points in different orientations.

5. The method for verifying the accuracy of triangulation according to claim 1, characterized in that, Estimating the Euclidean distance between the first feature point set and the second feature point set includes: Based on the two-dimensional coordinates of the feature points contained in the first feature point set and the second feature point set, and according to the camera's intrinsic and extrinsic parameter data, the three-dimensional coordinates of the feature points contained in the first feature point set and the second feature point set are obtained; Based on the three-dimensional coordinates of feature points in one orientation in the first feature point set and the three-dimensional coordinates of feature points in the same orientation in the second feature point set, the Euclidean distance is estimated until the Euclidean distance of all feature points is estimated, thus obtaining the Euclidean distance set.

6. The method for verifying the accuracy of triangulation according to claim 5, characterized in that, When obtaining the three-dimensional coordinates of the feature points contained in the first feature point set and the second feature point set, the following steps are included: Based on the two-dimensional coordinates of the feature points and the magnification of the camera in the x and y directions, the three-dimensional coordinates of the feature points are obtained, as shown in Formula 10: （10）； In the formula, The straight-line distance from the camera to the target is [missing information]. These are the coordinates of the corner points on the orientation map. Let be the magnification of the camera in the x, y directions. Let be the rotation matrix of the camera's extrinsic parameters. Let be the translation matrix of the camera's extrinsic parameters. These are the coordinates of the corner points in a three-dimensional coordinate system.

7. The method for verifying the accuracy of triangulation according to claim 5, characterized in that, When estimating Euclidean distance, as shown in Formula 11: （11）； In the formula, Let be the three-dimensional coordinates of the feature points in one orientation within the first feature point set. The coordinates are the three-dimensional coordinates of the feature points in the second feature point set that are located in the same orientation.

8. A system for verifying the accuracy of triangulation, characterized in that, include: Actual distance acquisition unit: After target A and target B are set up at any distance, it is used to acquire the actual distance between target A and target B measured by laser using an interferometric rangefinder; Azimuth map acquisition unit: After the camera of the driving triangulation system captures an azimuth map containing the target A and the target B, it is used to acquire the target azimuth map to obtain a target azimuth map set. Feature point acquisition unit: used to extract feature points of target A and target B in the target orientation map set, and obtain the first feature point set and the second feature point set respectively; Euclidean distance acquisition unit: used to estimate the Euclidean distance between the first feature point set and the second feature point set based on the camera's intrinsic and extrinsic parameters, and obtain the Euclidean distance set; Accuracy judgment unit: used to compare the Euclidean distances in the Euclidean distance set with the actual distances respectively. If the error between the two is less than the threshold, the measurement accuracy of the triangulation system is considered to be up to standard. Wherein, the center dots of the targets A and B are roughly circular after imaging, and are referred to as the first type of circle and the second type of circle; When extracting the center point, the following is included: Remove noise from target orientation maps at different azimuth angles; Pixel-level coarse localization is performed on the edges of the first and second type of circles under different orientations to identify the center points of the target A and the target B under different orientations; Subpixel-level fine positioning is performed on the edges of the first and second type of circles under different orientations; Least square fitting is performed on the edge points of the first and second type of circles under different orientations to obtain the sub-pixel localization of the center circles of the targets A and B under different orientations, thus completing the extraction of the center circles; When performing pixel-level coarse positioning, the following is included: The edges of the first and second types of circles under different orientations are extracted using the Canny operator, and the coordinates of the edge points of the first and second types of circles under different orientations are obtained. When performing least squares fitting, the following are included: Suppose that the number of edge points obtained after sub-pixel level fine positioning of the first and second type of circles under different orientations is . The fitted circular curve is shown in Formula 1: （1）； In the formula, Let these be the coordinates of the edge points of the first and second type of circles under different orientations. , , It is an unknown; The error function is shown in Formula 2: （2）； In the formula, ,Depend on List of error functions One error equation; Based on the above , , The third-order normal equations are formed, as shown in Equations 3 and 4: （3）； but (4); Equation 4 is solved by inverting the adjoint matrix, as shown in Equations 5, 6, 7, and 8: （5）； （6）； （7）； （8）； In the formula, , , , , , , ; The above is obtained , , Then, the coordinates of the center point and the radius are as shown in Formula 9: （9）； In the formula, The coordinates of the center point are: The radius of the center point.

Images