Ultraviolet camera optical axis calibration device and method based on multi-source error cooperative compensation

Abstract

The invention discloses an ultraviolet camera optical axis calibration device based on multi-source error cooperative compensation. Comprising a laser tracker, a contact measuring head, a collimator, a reflection ball base, a reflection target ball, a high-precision electric linear guide rail, a camera fixing tool, an optical two-dimensional rotary table, a data processing and control module, an optical vibration isolation air floating platform and a distributed temperature sensor array. And low-frequency noise is suppressed through the optical vibration isolation air floating platform. Position measurement is carried out through a laser tracker, the normal vector of the light emitting surface of the collimator, the actual motion axis of the high-precision electric linear guide rail and the optical axis of the camera are fitted, and the angle between the normal vector and the optical axis of the camera and the angle between the normal vector and the actual motion axis of the high-precision electric linear guide rail are calculated. The rotation angle of the optical two-dimensional rotary table and the position of a sliding block of the high-precision electric linear guide rail are adjusted, the optical axis of the camera is calibrated, and the requirement for the caliber of the collimator is lowered. Meanwhile, mechanical errors, environment errors and measurement errors are calculated by combining temperature data, and adjustment instruction correction is carried out.

Description

Technical Field

[0001] The present invention relates to the field of measurement and calibration technology, in particular to an optical axis calibration device and method for an ultraviolet camera based on collaborative compensation of multi-source errors. Background Art

[0002] The optical axis calibration of a UV camera plays a crucial role in UV detection performance. Existing UV camera technology has improved its measurement accuracy from angular binning to sub-arcsecond levels. Compared to traditional optical measurement, this technology offers significant advantages in cost, speed, convenience, and accuracy.

[0003] The optical axis of a UV camera is primarily determined by the mounting accuracy of the CMOS, the optical axis of the lens, and the overall optical axis. Measurement of the overall optical axis relies primarily on the parallelism of the collimator and the camera mounting fixture. Traditional calibration methods rely on theodolites or autocollimators, but their accuracy is limited by imaging resolution. These methods are complex to operate, offer limited measurement capabilities, and are unsuitable for high-precision coaxial multi-camera measurements. Using a large-aperture collimator is costly and uneconomical, and it also fails to achieve dynamic error separation and cannot correct for coupled mechanical errors, environmental thermal deformation, and measurement noise. Summary of the Invention

[0004] In response to the shortcomings of the existing technology, the present invention proposes a UV camera optical axis calibration device and method based on multi-source error collaborative compensation, which is suitable for optical calibration, space target positioning, multi-target coaxiality, and error compensation, and can improve measurement accuracy and efficiency and reduce equipment costs.

[0005] The ultraviolet camera optical axis calibration device based on multi-source error collaborative compensation includes a laser tracker, a contact probe, a parallel light tube, a reflective ball base, a reflective target ball, a high-precision electric linear guide, a camera fixing fixture, an optical two-dimensional turntable, a data processing and control module, an optical vibration isolation air floating platform and a distributed temperature sensor array.

[0006] The collimator is fixed on one side of the optical vibration isolation air floating platform. The laser tracker is set next to the optical vibration isolation air floating platform, the laser emission position is opposite to the light outlet of the collimator, and the contact probe is set near the light outlet of the collimator.

[0007] The high-precision electric linear guide is fixed on the optical vibration isolation air-floating platform, located between the collimator and the laser tracker. The optical two-dimensional turntable is used to fix the camera fixing fixture on the slider of the high-precision electric linear guide.

[0008] The air-floating vibration isolation platform is integrated with an active tilt adjustment module, which dynamically adjusts the platform levelness through the air film pressure to suppress low-frequency vibrations below 10 Hz.

[0009] The reflective ball base is used to fix the reflective target ball on the high-precision electric linear guide rail and the camera fixing tooling.

[0010] The distributed temperature sensor array includes multiple temperature sensors fixed on a parallel light tube, a high-precision electric linear guide rail and an optical two-dimensional turntable to measure the environment or equipment temperature.

[0011] The data processing and control module collects measurement data from the laser tracker and the distributed temperature sensor array, calculates multi-source error compensation values, and controls the movement of the high-precision electric linear guide rail and the optical two-dimensional turntable.

[0012] The UV camera optical axis calibration method based on multi-source error collaborative compensation uses the above device to calibrate the UV camera optical axis. The specific steps are as follows:

[0013] Step 1: Establish a global coordinate system with the laser tracker as the origin, use the contact probe to measure the multi-point coordinates of the collimator's light-emitting surface, and calculate the normal vector of the collimator's light-emitting surface. The reflective target ball is driven by the slider of the high-precision electric linear guide to move, the coordinates of the reflective target ball are measured, and the actual motion axis of the high-precision electric linear guide is fitted. , calculate the angle θ between the high-precision electric linear guide and the parallel light tube, and adjust the installation position of the high-precision electric linear guide until .

[0014] Step 2: Fix the camera fixture on the optical two-dimensional turntable, and then fix the optical two-dimensional turntable on the slider of the high-precision electric linear guide. Start the high-precision electric linear guide to drive the camera fixture and the reflective target ball to move synchronously to multiple positions. At the kth position p k , fix the reflective target ball, measure the coordinates of the point at m different angles, and fit the coordinates at position p k The rotation axis vector of the camera fixture .

[0015] Step 3: Calculate the reference vector The rotation axis vector of the camera fixture Angle , and the base vector Actual motion axis with high-precision electric linear guides The angle between , adjust the angle of the optical two-dimensional turntable or the position of the slider of the high-precision electric linear guide so that 、 .

[0016] Step 4: The temperature data measured by the distributed temperature sensor array is collected through the data processing and control module, and the temperature difference is calculated based on the historical data. The vibration acceleration of the optical vibration isolation air flotation platform is collected through the acceleration sensor. . Calculate the mechanical error, environmental error and measurement error of the device and perform error compensation.

[0017] Step 5. Return to step 2, re-measure the coordinates and vector fit, and adjust the module posture until When the calibration is completed.

[0018] The present invention has the following beneficial effects:

[0019] 1. The proposed deflection angle calculation method and error compensation algorithm can ensure a measurement accuracy of 1". At the same time, the collimator used only needs to have an aperture larger than the camera lens to achieve coaxial measurement of multiple camera optical axes, which greatly reduces the cost of the collimator. Because the system only uses the laser tracker's coordinate system globally, manual modeling errors are reduced, the system bandwidth is improved, and the system operation is simple and low-cost.

[0020] 2. The multi-axis error chain is decoupled through the orthogonal error separation algorithm, and the coupled errors are decomposed into guide rail error, turntable error and measurement error, so that the errors are mapped to independent control channels and can be compensated independently.

[0021] 3. Air-flotation-laser collaborative control: the air-flotation platform is used to suppress low-frequency vibrations less than 10 Hz, and the laser tracker is used to correct high-frequency errors greater than 10 Hz in a closed-loop manner, thereby improving the system bandwidth.

[0022] 4. A nonlinear temperature compensation model is proposed. Through finite element thermal model calibration, the compensation strategy learning of temperature gradient, temperature change rate and historical data is integrated to reduce manual modeling errors and achieve advanced compensation.

[0023] 5. A multi-point dynamic measurement method is proposed. By calculating the angular error between the rotation axis vector and the reference vector at different positions when the camera fixture moves with the slider, high-precision coaxial calibration of multiple UV cameras can be achieved, reducing the requirements for the aperture size of the collimator. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 is a schematic diagram of the calibration device;

[0025] Figure 2 Error separation algorithm flow chart;

[0026] Figure 3 This is a schematic diagram of the laser compensation signal;

[0027] Figure 4 To control the vibration power spectrum density for air flotation-laser synergy;

[0028] Figure 5 This is the compensation effect of laser compensation on high-frequency residual.

[0029] 1. Laser tracker; 2. Contact probe; 3. Collimator; 4. Reflection ball base; 5. Reflection target ball; 6. High-precision electric linear guide; 7. Camera fixing fixture; 8. Optical two-dimensional turntable; 9. Data processing and control module; 10. Optical vibration isolation air flotation platform; 11. Temperature sensor. DETAILED DESCRIPTION

[0030] The present invention will be further explained below with reference to the accompanying drawings;

[0031] like Figure 1 As shown, the ultraviolet camera optical axis calibration device based on multi-source error collaborative compensation includes a laser tracker 1, a contact probe 2, a collimator 3, a reflective ball base 4, a reflective target sphere 5, a high-precision electric linear guide 6, a camera fixing fixture 7, an optical two-dimensional turntable 8, a data processing and control module 9, an optical vibration isolation air floating platform 10 and a distributed temperature sensor array 11.

[0032] The collimator 3 is fixed to one side of the optical vibration isolation air floating platform. The laser tracker 1 is set next to the optical vibration isolation air floating platform 10, with the laser emission position opposite to the light outlet of the collimator 3, and the contact probe 2 is set near the light outlet of the collimator 3.

[0033] The high-precision electric linear guide 6 is fixed on the optical vibration isolation air-floating platform and is located between the collimator 3 and the laser tracker 1. The optical two-dimensional turntable 8 is used to fix the camera fixing fixture 7 on the slider of the high-precision electric linear guide 6.

[0034] The optical vibration isolation air floating platform 10 is integrated with an active tilt adjustment module, which dynamically adjusts the surface levelness through the air film pressure to suppress low-frequency vibrations below 10 Hz.

[0035] The reflective ball base 4 is used to fix the reflective target ball 5 on the high-precision electric linear guide rail 6 and the camera fixing tool 7.

[0036] The distributed temperature sensor array includes a plurality of temperature sensors 11 fixed on the collimator 3, the high-precision electric linear guide 6 and the optical two-dimensional turntable 8 to measure the environment or equipment temperature.

[0037] The data processing and control module 9 collects measurement data from the laser tracker 1 and the temperature sensor 11 , calculates multi-source error compensation values, and controls the movement of the high-precision electric linear guide 6 and the optical two-dimensional turntable 8 .

[0038] like Figure 2As shown in FIG, a UV camera optical axis calibration method based on multi-source error collaborative compensation is used to calibrate the UV camera optical axis using the above device. The specific steps are as follows:

[0039] Step 1: Preheat the laser tracker to ensure its stability. Establish a global coordinate system with the laser tracker as the origin and calculate the normal vector of the collimator's light-emitting surface. , the actual motion axis of high-precision electric linear guide and the rotation axis vector of the camera fixture , the specific steps are as follows:

[0040] s1.1. Use the contact probe to measure the three-dimensional coordinates of 9 non-collinear points on the light-emitting surface of the collimator , , by fitting the spatial plane equation using the least squares method, calculate the normal vector of the collimator's light-emitting surface , as the base vector.

[0041] For the 3D coordinates of 9 non-collinear points , first calculate the centroid coordinates :

[0042]

[0043] Then construct the covariance matrix And expand:

[0044]

[0045] Where T represents the matrix transpose, 、 、 .

[0046] Perform eigenvalue decomposition on the covariance matrix M:

[0047]

[0048] Then the reference vector is the eigenvector corresponding to the minimum eigenvalue.

[0049] According to the measurement error of the laser tracker and the number of points selected, you can control the fitting residuals .

[0050] s1.2. Start the high-precision electric linear guide, drive the reflective target ball to move through the slider, and use the laser tracker to measure the coordinates of the reflective target ball when it moves to different positions. Similarly, perform eigenvalue decomposition based on the covariance matrix, where the eigenvector corresponding to the maximum eigenvalue represents the main direction of the data point distribution, that is, the actual motion axis of the high-precision electric linear guide. s.

[0051] Calculate the actual motion axis of the high-precision electric linear guide according to the vector angle calculation formula With the reference vector The angle between Angle, when When adjusting the installation position of the high-precision electric linear guide, .

[0052]

[0053] s1.3. Fix the camera fixture to the optical 2D turntable, and then fix the optical 2D turntable to the slider of the high-precision electric linear guide. Fix the reflective target ball to the camera fixture through the reflective ball base. Start the high-precision electric linear guide to drive the camera fixture and the reflective target ball to move synchronously. Each time it moves to a position p k , the rotation angle and pitch angle are changed by rotating the optical two-dimensional turntable, so that the turntable rotates around the rotation axis of the camera fixed fixture, and the coordinates L of the reflective target ball at different rotation angles are recorded by the laser tracker m , m=1,2,3...M, in order to fit the rotation axis vector of the camera fixture , it is necessary to solve the main direction axis of the reflective target sphere coordinate point set, even if all coordinate points to The sum of squared vertical distances is minimized:

[0054]

[0055] Where c represents the coordinate of the rotation center, and V represents the rotation axis vector of the camera fixture. The unit vector of .

[0056] Perform eigenvalue decomposition based on the covariance matrix and use the eigenvector corresponding to the minimum eigenvalue as the rotation axis vector of the camera fixture .

[0057] Step 2: Calculate the camera fixture position p k Lower reference vector The rotation axis vector of the camera fixture Angle ,when When the turntable inverse kinematics model is used to output the correction instruction , perform linear error compensation:

[0058]

[0059]

[0060] in, Indicates the angle to which the optical two-dimensional turntable is to be adjusted; represents the cumulative error of the camera fixture moving from the first position to the kth position, is the proportional gain coefficient, with a value of 0.5 and a unit of μrad / ″, used in the proportional control link; It is the integral gain coefficient, with a value of 0.1 and a unit of μrad / (″·s), and is used in the integral control link.

[0061] Calculate the rotation axis vector of the camera fixture Actual motion axis with high-precision electric linear guides The angle between ,when When the turntable inverse kinematic model is used to generate straightness compensation pulses , further adjust the slider position of the high-precision electric linear guide until :

[0062]

[0063]

[0064] in, Indicates the length of the guide rail in mm; Indicates the distance the slider moves, in mm.

[0065] Calculating the thermal expansion deformation of high-precision electric linear guides , perform pre-compensation of the slider motion instruction and calculate the theoretical target position after compensation :

[0066]

[0067]

[0068] in, is the initial length of the high-precision electric linear guide, in mm. β is the thermal expansion coefficient of the high-precision electric linear guide material, in °C -1 Starget is the actual target position of the high-precision electric linear guide after pre-compensation, in mm. ΔT is the difference between the actual temperature and the initial temperature, in °C. Zero is the zero point of the high-precision guide coordinate system.

[0069] Step 3: The data processing and control module collects the temperature data measured by the distributed temperature sensor array to obtain the ambient temperature T and the temperature gradient ΔT of the high-precision electric linear guide and the optical two-dimensional turntable. The acceleration sensor collects the vibration acceleration of the optical vibration isolation air flotation platform. Calculate the mechanical error, environmental error and measurement error of the device and perform error compensation. The specific steps are as follows:

[0070] s3.1. Calculate the center drift of the optical 2D turntable :

[0071]

[0072] The unit of ΔO is mm, γ1 represents the linear error coefficient, and γ2 represents the nonlinear error coefficient. The rotation command of the optical 2D turntable Dynamic correction to :

[0073]

[0074] Wherein, R is the rotation radius of the optical two-dimensional turntable, in mm.

[0075] s3.2. Calculate the temperature drift of the laser tracker measurement value based on the temperature data T measured by the distributed temperature sensor array, and calculate the coordinate measurement value P of the reflective target sphere. 测 Make corrections:

[0076]

[0077]

[0078] in, is the measurement coordinate correction amount, the unit is mm; is the temperature change rate, in °C / min; α1 and α2 are the temperature compensation coefficients calibrated by experiments. 真 Indicates the corrected coordinate value.

[0079] s3.3, start the optical vibration isolation air flotation platform to suppress low-frequency vibrations below 10Hz, according to the vibration acceleration of the optical vibration isolation air flotation platform , calculate the inclination of the platform :

[0080]

[0081] in, is the pressure feedback coefficient, in kPa·s 2 / m. Dynamically adjust the platform level by air film pressure:

[0082]

[0083] Among them, P 气is the air film pressure after real-time adjustment, P0 is the reference air film pressure, is the pressure feedback coefficient.

[0084] During the laser tracker measurement process, there is multipath reflection interference in the data. The frequency response function Performing real-time frequency domain filtering to attenuate signals with frequencies above 10 Hz can effectively correct high-frequency errors and improve measurement data accuracy:

[0085]

[0086] Where h is the imaginary unit and f is the frequency of the input signal. , represents the cutoff frequency.

[0087] The high-frequency error with a frequency higher than 10Hz is corrected in a closed loop by the real-time measurement data of the laser tracker to achieve air-floatation-laser collaborative control, thereby improving the system bandwidth. Figures 3-5 The effect of gain is shown. Figure 3 It can be seen from the figure that this method can separate high-frequency errors in real time, and the laser compensation signal can effectively compensate for high-frequency residuals. Figure 4 It can be seen from the figure that after starting the coordinated control, the nonlinear harmonics at different frequencies can be effectively suppressed. Figure 5 It can be seen that, especially in the high-frequency region with a frequency greater than 10 Hz, after laser compensation, the structural function of 60 Hz is suppressed by 45 dB, while the high-frequency noise of 120 Hz is suppressed by 32 dB, with an average suppression of more than 30 dB.

[0088] Step 4: Dynamic Temperature Compensation Model (DTCM), integrated into the data processing and control module, is based on the temperature data T measured by the distributed temperature sensor array at different times t. t , calculating the deformation of the device :

[0089]

[0090] Where n represents the number of components; Indicates the thermal expansion coefficient of the qth component, in °C -1 ; represents the actual temperature of the qth component measured at time t; Represents the initial geometric length of the qth component, in meters.

[0091] Step 5: According to the deformation obtained by the dynamic temperature compensation model, the laser tracker measurement data is corrected in real time, and the pre-compensation of the motion control instruction is calculated. Return to step 2 and re-perform vector fitting and module posture adjustment until When the calibration is completed.

Claims

1. A UV camera optical axis calibration device based on multi-source error collaborative compensation includes a laser tracker, a contact probe, a collimator, a camera fixture, a reflective sphere base, a reflective target sphere, and a data processing and control module. The device is characterized by: It also includes high-precision electric linear guides, camera fixtures, optical 2D turntables, optical vibration isolation air-floating platforms, and distributed temperature sensor arrays; The collimator is fixed on one side of the optical vibration isolation air floating platform; the laser tracker is arranged next to the optical vibration isolation air floating platform, the laser emission position is opposite to the light outlet of the collimator, and the contact probe is arranged near the light outlet of the collimator; The high-precision electric linear guide is fixed on the optical vibration isolation air-floating platform and is located between the collimator and the laser tracker; the optical two-dimensional turntable is used to fix the camera fixing fixture on the slider of the high-precision electric linear guide; The air-floating vibration isolation platform dynamically adjusts the platform levelness through the air film pressure to suppress low-frequency vibrations; The reflective ball base is used to fix the reflective target ball on the high-precision electric linear guide rail and the camera fixing tooling; The distributed temperature sensor array includes multiple temperature sensors fixed on a collimator, a high-precision electric linear guide and an optical two-dimensional turntable to measure the ambient or device temperature; The data processing and control module collects measurement data from the laser tracker and the distributed temperature sensor array, calculates mechanical errors, environmental errors and measurement errors, controls the movement of high-precision electric linear guides and optical two-dimensional turntables, and calibrates the camera optical axis.

2. A UV camera optical axis calibration method based on multi-source error collaborative compensation is characterized by: The device as claimed in claim 1 is used to calibrate the optical axis of the UV camera, and the specific steps are as follows: Step 1: Establish a global coordinate system with the laser tracker as the origin, and fit the normal vector of the collimator's light-emitting surface. Actual motion axis with high-precision electric linear guides , calculate the angle between the high-precision electric linear guide and the parallel light tube , adjust the installation position of the high-precision electric linear guide until ; Step 2: Fix the camera fixture to the slider of the high-precision electric linear guide through the optical two-dimensional turntable, start the high-precision electric linear guide, and drive the camera fixture and the reflective target ball to move synchronously to multiple positions. At the kth position p k Fix the reflective target ball and rotate the optical two-dimensional turntable to different angles to fit the position p k The rotation axis vector of the camera fixture ; Step 3: Calculate the normal vector The rotation axis vector of the camera fixture Angle ,as well as Actual motion axis with high-precision electric linear guides The angle error between , adjust the angle of the optical two-dimensional turntable or the position of the slider of the high-precision electric linear guide so that 、 ; Step 4: Obtain temperature data from the distributed temperature sensor array and calculate the temperature difference based on historical data; collect the vibration acceleration of the optical vibration isolation air flotation platform through the acceleration sensor ; Mechanical errors, environmental errors and measurement errors of computing devices; Step 5. Return to step 2, re-measure the coordinates and re-fit the vectors according to the error value obtained in step 4, and adjust the module posture until When the calibration is completed.

3. The method for calibrating the optical axis of a UV camera based on coordinated compensation of multi-source errors according to claim 2, wherein: Use a contact probe to measure the three-dimensional coordinates of multiple non-collinear points on the light-emitting surface of the parallel light tube, and then calculate the centroid coordinates of multiple non-collinear points , construct the covariance matrix And expand: ； Where T represents the matrix transpose, 、 、 , Represents the three-dimensional coordinates of the i-th non-collinear point; perform eigenvalue decomposition on the covariance matrix M and select the eigenvector corresponding to the minimum eigenvalue as the normal vector of the collimator light output surface .

4. The method for calibrating the optical axis of a UV camera based on coordinated compensation of multi-source errors according to claim 2, wherein: Fix the reflective target ball on the slider of the high-precision electric linear guide, start the high-precision electric linear guide, and drive the reflective target ball to move through the slider. Use a laser tracker to measure the coordinates of the reflective target ball when it moves to different positions. Perform eigenvalue decomposition based on the covariance matrix, and use the eigenvector corresponding to the maximum eigenvalue of the rotation as the actual motion axis of the high-precision electric linear guide. .

5. The method for calibrating the optical axis of a UV camera based on coordinated compensation of multi-source errors according to claim 2, wherein: At the kth position p k , with the goal of minimizing the sum of the squares of the vertical distances from the coordinate points of the optical two-dimensional turntable to the rotation center at different angles, the k The rotation axis vector of the camera fixture .

6. The method for calibrating the optical axis of a UV camera based on coordinated compensation of multi-source errors according to claim 2, wherein: when When the adjustment angle of the output optical two-dimensional turntable is : ； in, represents the cumulative error of the camera fixture moving from the first position to the kth position, is the proportional gain coefficient, is the integral gain coefficient; when When the straightness compensation pulse is output , adjust the slider position of the high-precision electric linear guide: ； in, Indicates the length of the guide rail. Indicates the distance the slider moves; calculates the thermal expansion deformation of high-precision electric linear guides , and get the actual target position Starget of the slider after adjustment: ； ； in, is the initial length of the high-precision electric linear guide, β is the thermal expansion coefficient of the high-precision electric linear guide material, and ΔT is the difference between the actual temperature and the initial temperature.

7. The method for calibrating the optical axis of a UV camera based on coordinated compensation of multi-source errors according to claim 2 or 6, characterized in that: Calculating the Rotation Center Drift of an Optical 2D Turntable : ； Among them, γ1 represents the linear error coefficient, γ2 represents the nonlinear error coefficient; according to the rotation center drift The rotation command of the optical 2D turntable Dynamic correction to : ； Wherein, R is the rotation radius of the optical two-dimensional turntable.

8. The method for calibrating the optical axis of a UV camera based on collaborative compensation of multi-source errors according to any one of claims 2 to 5, characterized in that: Calculating Temperature Drift of Laser Tracker Measurements , the coordinate measurement value P of the reflective target sphere 测 Make corrections: ； ； in, is the temperature change rate, T represents the temperature data measured by the distributed temperature sensor array, t represents time, ΔT is the difference between the current temperature and the initial temperature; α1 and α2 are the temperature compensation coefficients calibrated by experiments; P 真 Indicates the corrected coordinate value.

9. The method for calibrating the optical axis of a UV camera based on coordinated compensation of multi-source errors according to claim 2, wherein: Collect the vibration acceleration α of the optical vibration isolation air flotation platform 振动 , through the frequency response function Suppress signals with frequencies above 10Hz and calculate the inclination of the platform , adjust the air film pressure: ； ； in, is the pressure feedback coefficient, P 气 is the air film pressure after real-time adjustment, P0 is the reference air film pressure, is the pressure feedback coefficient.

10. The method for calibrating the optical axis of a UV camera based on collaborative compensation of multi-source errors according to any one of claims 2 to 6, characterized in that: According to the temperature data T measured by the distributed temperature sensor array at different times t t , calculating the deformation of the device , used to perform temperature compensation on laser tracker measurement data and adjustment instructions: ； Where n represents the number of components; represents the thermal expansion coefficient of the qth component; represents the actual temperature of the qth component measured at time t; represents the initial geometric length of the qth component.

Images