Dynamic calibration method for sensitivity of linear angular vibration sensor based on multi-frequency sine excitation

Abstract

The invention discloses a dynamic calibration method for the sensitivity of a linear angular vibration sensor based on multi-frequency sine excitation. The method comprises the following steps: constructing a line angle vibration sensor sensitivity dynamic calibration device based on multi-frequency sine excitation; providing multi-frequency sine line angle excitation; collecting a motion sequence image and an output signal; calculating the line angle excitation displacement of the calibrated line angle vibration sensor; and calculating to obtain the sensitivity of the calibrated line angle vibration sensor. The method is efficient and accurate, and the calibration process is simplified; the method based on monocular vision is used for measuring the vibration displacement of the linear angle vibration sensor at different frequencies, and only a simple monocular vision measuring system is needed; the multi-frequency sine line angle excitation output by the line and angle vibration generating device can be controlled, and the influence of other extra excitation on the sensor can be effectively avoided; the method is suitable for different types of sensors, and the multi-frequency sine excitation method has a wide application range and can be used for calibrating different types of linear angle vibration sensors, such as gyroscopes and accelerometers.

Description

Technical Field

[0001] The invention belongs to the technical field of image processing and sensor calibration, and in particular relates to a dynamic calibration method for sensitivity of a linear angle vibration sensor based on multi-frequency sinusoidal excitation. Background Art

[0002] As a basic dynamic parameter measurement unit, linear angle vibration sensors have been widely used in precision manufacturing, navigation and guidance, building structure health monitoring and other fields. Linear angle vibration sensors with accurate sensitivity can improve the reliability of monitoring and measurement, help to respond to the dynamic changes of the system in a timely and accurate manner, thereby improving system performance and safety, reducing the risk of failure, and providing more reliable data support. Usually, the sensitivity of linear angle vibration sensors is considered to be of known accuracy in actual engineering applications, and its accuracy is crucial to ensure the high performance and reliability of the measurement system. However, existing calibration methods often cannot accurately calibrate the sensitivity of linear angle vibration sensors in a short period of time, and there are problems such as low efficiency, poor flexibility, and large errors caused by point-by-point operations. Therefore, there is an urgent need for an accurate and efficient calibration method to ensure the calibration accuracy and efficiency of linear angle vibration sensors. Summary of the invention

[0003] In order to solve the above problems, the purpose of the present invention is to provide an efficient, flexible and accurate method for dynamic calibration of the sensitivity of a linear angle vibration sensor, which can significantly improve the efficiency of vibration calibration.

[0004] In order to achieve the above object, the method for dynamic calibration of sensitivity of a linear angle vibration sensor based on multi-frequency sinusoidal excitation provided by the present invention comprises the following steps performed in sequence:

[0005] 1) Construct a dynamic calibration device for the sensitivity of a linear angle vibration sensor based on multi-frequency sinusoidal excitation;

[0006] 2) The high-contrast target and the calibrated line angle vibration sensor in the above-mentioned line angle vibration sensor sensitivity dynamic calibration device based on multi-frequency sinusoidal excitation are simultaneously fastened to the working table of the line and angle vibration generating device, and then a signal generator is used to generate a multi-frequency sinusoidal excitation signal with no mutual interference in the spectrum, so that the high-contrast target, the calibrated line angle vibration sensor and the working table of the line and angle vibration generating device have consistent motion characteristics, and the motion control of the line and angle vibration generating device is realized based on the excitation signal, so as to provide multi-frequency sinusoidal line angle excitation for the calibrated line angle vibration sensor;

[0007] 3) Using a camera to collect the motion sequence images of the high-contrast target, using a data acquisition card to collect the output signal of the calibrated angular vibration sensor, and transmitting the motion sequence images and output signals to the image and signal processing and display unit;

[0008] 4) The image and signal processing and display unit uses a template matching method to determine the ROI of the above-mentioned motion sequence image to eliminate the interference of background noise and similar edges, and then accurately extracts the motion feature edge in the ROI through the LSD method, and solves the line angle excitation displacement of the calibrated line angle vibration sensor based on the extracted motion feature edge;

[0009] 5) SAM is used to fit the linear angle excitation displacement calculated above and the corresponding output signal of the calibrated linear angle vibration sensor. The excitation displacement corresponding to each frequency and the output signal amplitude of the calibrated linear angle vibration sensor are obtained by solving the sparse matrix composed of the overdetermined equations. Finally, the sensitivity of the calibrated linear angle vibration sensor is calculated using the excitation displacement and the output signal amplitude.

[0010] In step 1), the dynamic calibration device for sensitivity of the linear angle vibration sensor based on multi-frequency sinusoidal excitation comprises a signal generator, a linear and angular vibration generating device, a high contrast target, a lighting device, a camera, a data acquisition card and an image and signal processing and display unit; wherein the high contrast target and the calibrated linear angle vibration sensor are fastened to the working table of the linear and angular vibration generating device; the lens of the camera faces the high contrast target; the lighting device is arranged on one side of the camera, and is used to provide lighting for the camera to ensure the quality of the acquired image; the image and signal processing and display unit are electrically connected to the signal generator, the camera and the data acquisition card respectively; the signal generator and the data acquisition card are also electrically connected to the linear and angular vibration generating device and the calibrated linear angle vibration sensor respectively; the signal generator generates a multi-frequency sinusoidal signal to control the linear and angular vibration generating device so that its working table outputs multi-frequency sinusoidal linear angle excitation;

[0011] In step 2), the frequency spectrum of the multi-frequency sinusoidal excitation signal x(t) is free of mutual interference. k Need to meet:

[0012]

[0013] Among them, f0 and p0 are the fundamental frequency and the fundamental prime number respectively, p i is a prime number greater than the base prime number p0, f k is the fundamental frequency f0 and the subsequent frequencies of the fundamental frequency f0 within its ten-fold frequency range, ω k and x k The frequencies are f k The corresponding angular frequency and amplitude, (-1) k Used to reduce the overshoot effect caused by the summation sign, t is the time.

[0014] In step 4), the image and signal processing and display unit uses a template matching method to determine the ROI of the motion sequence image to eliminate the interference of background noise and similar edges, and then accurately extracts the motion feature edge in the ROI by the LSD method, and solves the line angle excitation displacement of the calibrated line angle vibration sensor according to the extracted motion feature edge. The method is:

[0015] 4.1) The image and signal processing and display unit reads the motion sequence images of the high-contrast target captured by the camera {F j (x, y)}, using a nonlinear optimization algorithm based on the least squares principle, the distortion coefficient of the camera 7 is adjusted through continuous iteration to minimize the difference between the actual image and the theoretical image;

[0016] 4.2) Use Harris corner detection algorithm to preliminarily locate the feature points in the image, and then use sub-pixel interpolation fitting method to obtain the relative coordinates of each feature point;

[0017] 4.3) Establishing a coordinate transformation linear equation group by collecting the relationship between the feature points on the high-contrast target in the known world coordinate system and the pixel projection points in the captured image, and then using the coordinate transformation linear equation group to transform the relative coordinates of each feature point into an absolute coordinate in the world coordinate system;

[0018] 4.4) According to the absolute coordinates of the above feature points, the motion sequence image {F j (x,y)};

[0019] 4.5) Then find the position of the center of the circular area and determine the ROI;

[0020] 4.6) Based on the LSD method, the straight line edge in the ROI is extracted, and the translation change and direction vector change of the straight line edge are estimated by combining the feature point matching between adjacent frame images, and then the linear angle excitation displacement {s(t j )}.

[0021] In step 5), the SAM method is used to fit the calculated line angle excitation displacement and the corresponding output signal of the calibrated line angle vibration sensor, and the excitation displacement corresponding to each frequency and the output signal amplitude of the calibrated line angle vibration sensor are obtained by solving the sparse matrix composed of the overdetermined equations. Finally, the sensitivity of the calibrated line angle vibration sensor is calculated by using the excitation displacement and the output signal amplitude. The method is: through the motion sequence image {F j (x,y)} to obtain the linear angle excitation displacement {s(t j )} is for each frequency f k The displacement amplitude is fitted using the following SAM method:

[0022] s(t j )=Acos(ω k t j )-Bsin(ω k t j )+C (2)

[0023] The equation in matrix form is:

[0024]

[0025] Among them, t j is the sampling time of the corresponding j-th frame image, A and B are the corresponding sinusoidal components, C is the offset component, and D is the corresponding matrix. By solving the overdetermined equations composed of m equations shown in formula (3), we can obtain the different frequencies f k The excitation displacement at the position and the output signal amplitude V(t) of the calibrated angular vibration sensor are calculated; finally, the sensitivity of the calibrated angular vibration sensor at different frequencies is obtained by using the above excitation displacement and output signal amplitude V(t).

[0026] The method for dynamic calibration of sensitivity of a linear angle vibration sensor based on multi-frequency sinusoidal excitation provided by the present invention has the following advantages:

[0027] 1. The method of the present invention is efficient, accurate and simplifies the calibration process. Compared with the traditional calibration method of testing each frequency one by one, this method can calibrate the sensitivity of different frequencies in the range of base frequency to ten times the frequency band in one operation, saving at least one order of magnitude in calibration time and avoiding the large amount of time consumption caused by repeated point-by-point operations in the traditional calibration method.

[0028] 2. The method of the present invention utilizes a monocular vision-based method to measure the vibration displacement of the linear angle vibration sensor at different frequencies, and only requires a simple monocular vision measurement system.

[0029] 3. The present invention can control the multi-frequency sinusoidal line-angle excitation output by the line and angular vibration generating device by designing a multi-frequency sinusoidal signal with no mutual interference in the generated spectrum, and can also effectively avoid the influence of other additional excitations on the sensor.

[0030] 4. The method of the present invention is applicable to different types of sensors. The multi-frequency sinusoidal excitation method has a wide range of applications and can be used to calibrate different types of linear angular vibration sensors, such as gyroscopes and accelerometers. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 It is a schematic diagram of the structure of the linear angle vibration sensor sensitivity dynamic calibration device based on multi-frequency sinusoidal excitation constructed in the present invention when used for linear vibration sensor calibration;

[0032] Figure 2 It is a schematic diagram of the structure of the linear angular vibration sensor sensitivity dynamic calibration device based on multi-frequency sinusoidal excitation constructed in the present invention when used for angular vibration sensor calibration;

[0033] Figure 3 A flow chart of a method for dynamic calibration of sensitivity of a linear angle vibration sensor based on multi-frequency sinusoidal excitation provided by the present invention;

[0034] Figure 4 This is a flow chart of multi-frequency sinusoidal excitation displacement solution based on monocular vision;

[0035] Figure 5 (a) and (b) are comparisons of the sensitivity calibration results and relative standard deviation data of the linear vibration sensor using the method of the present invention, the laser interferometry method (LI), and the traditional monocular vision method (MV);

[0036] FIG6 (a) and (b) are comparisons of the sensitivity calibration results and relative standard deviation data of the angular vibration sensor using the method of the present invention, the laser interferometry method (LI), and the traditional monocular vision method (MV). DETAILED DESCRIPTION

[0037] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0038] like Figure 3 As shown, the method for dynamic calibration of sensitivity of a linear angle vibration sensor based on multi-frequency sinusoidal excitation provided by the present invention comprises the following steps performed in sequence:

[0039] 1) Construct a dynamic calibration device for the sensitivity of a linear angle vibration sensor based on multi-frequency sinusoidal excitation;

[0040] like Figure 1 , 2 As shown, the dynamic calibration device for sensitivity of the linear angle vibration sensor based on multi-frequency sinusoidal excitation comprises a signal generator 1, a linear and angular vibration generating device 2, a high contrast target 4, a lighting device 6, a camera 7, a data acquisition card 8 and an image and signal processing and display unit 9; wherein the high contrast target 4 and the calibrated linear angle vibration sensor 5 are fastened to the working table 3 of the linear and angular vibration generating device 2; the lens of the camera 7 faces the high contrast target 4; the lighting device 6 is arranged on one side of the camera 7, and is used to provide lighting for the camera 7 to ensure the quality of the acquired image; the image and signal processing and display unit 9 is electrically connected to the signal generator 1, the camera 7 and the data acquisition card 8 respectively; the signal generator 1 and the data acquisition card 8 are also electrically connected to the linear and angular vibration generating device 2 and the calibrated linear angle vibration sensor 5 respectively; the signal generator 1 generates a multi-frequency sinusoidal signal to control the linear and angular vibration generating device 2 so that its working table 3 outputs a multi-frequency sinusoidal linear angle excitation;

[0041] 2) The high-contrast target 4 and the calibrated line angle vibration sensor 5 in the above-mentioned line angle vibration sensor sensitivity dynamic calibration device based on multi-frequency sinusoidal excitation are simultaneously fastened to the working table 3 of the line and angle vibration generating device 2, and then a multi-frequency sinusoidal excitation signal with no mutual interference in the spectrum is generated by the signal generator 1, so that the high-contrast target 4, the calibrated line angle vibration sensor 5 and the working table 3 of the line and angle vibration generating device 2 have consistent motion characteristics, and the motion control of the line and angle vibration generating device 2 is realized based on the excitation signal, so as to provide multi-frequency sinusoidal line angle excitation for the calibrated line angle vibration sensor 5;

[0042] The frequency spectrum has no mutual interference among the multiple frequencies f in the multi-frequency sinusoidal excitation signal x(t) k Need to meet:

[0043]

[0044] Among them, f0 and p0 are the fundamental frequency and the fundamental prime number respectively, p i is a prime number greater than the base prime number p0, f k is the fundamental frequency f0 and the subsequent frequencies of the fundamental frequency f0 within its ten-fold frequency range, ω k and x k The frequencies are f k The corresponding angular frequency and amplitude, (-1) k Used to reduce the overshoot effect caused by the summation sign, t is the time.

[0045] 3) Using the camera 7 to collect the motion sequence images of the high contrast target 4 j (x, y)}, use the data acquisition card 8 to collect the output signal of the calibrated angular vibration sensor 5, and convert the motion sequence image {F j (x, y)} and the output signal is transmitted to the image and signal processing and display unit 9;

[0046] 4) The image and signal processing and display unit 9 uses the template matching method to determine the above motion sequence image {F j (x, y)} ROI to eliminate the interference of background noise and similar edges, and then accurately extract the motion feature edges in the ROI through the LSD method, and calculate the line angle excitation displacement of the calibrated line angle vibration sensor 5 according to the extracted motion feature edges;

[0047] The method is:

[0048] 4.1) Figure 4 As shown, the image and signal processing and display unit 9 reads the motion sequence images of the high contrast target 4 taken by the camera 7 {F j(x, y)}, using a nonlinear optimization algorithm based on the least squares principle, the distortion coefficient of the camera 7 is adjusted through continuous iteration to minimize the difference between the actual image and the theoretical image;

[0049] 4.2) Use Harris corner detection algorithm to preliminarily locate the feature points in the image, and then use sub-pixel interpolation fitting method to obtain the relative coordinates of each feature point;

[0050] 4.3) Establishing a coordinate transformation linear equation group by collecting the relationship between the feature points on the high-contrast target 4 in the known world coordinate system and the pixel projection points in the captured image, and then using the coordinate transformation linear equation group to transform the relative coordinates of each feature point into an absolute coordinate in the world coordinate system;

[0051] 4.4) According to the absolute coordinates of the above feature points, the motion sequence image {F j (x,y)};

[0052] 4.5) Then find the position of the center of the circular area and determine the ROI;

[0053] 4.6) Based on the LSD method, the straight line edge in the ROI is extracted, and the translation change and direction vector change of the straight line edge are estimated by combining the feature point matching between adjacent frame images, and then the line angle excitation displacement {s(t j )}.

[0054] 5) SAM is used to fit the linear angle excitation displacement calculated above and the corresponding output signal of the calibrated linear angle vibration sensor. The excitation displacement corresponding to each frequency and the output signal amplitude of the calibrated linear angle vibration sensor are obtained by solving the sparse matrix composed of the overdetermined equations. Finally, the sensitivity of the calibrated linear angle vibration sensor is calculated using the excitation displacement and the output signal amplitude.

[0055] The method is:

[0056] Through the motion sequence image {F j (x,y)} to obtain the linear angle excitation displacement {s(t j )} is for each frequency f k The displacement amplitude is fitted using the following SAM method:

[0057] s(t j )=Acos(ω k t j )-Bsin(ω k t j )+C (2)

[0058] The equation in matrix form is:

[0059]

[0060] Among them, t j is the sampling time of the corresponding j-th frame image, A and B are the corresponding sinusoidal components, C is the offset component, and D is the corresponding matrix. By solving the overdetermined equations composed of m equations shown in formula (3), we can obtain the different frequencies f k The excitation displacement at and the output signal amplitude V(t) of the calibrated angular vibration sensor 5 are calculated; finally, the sensitivity of the calibrated angular vibration sensor 5 at different frequencies is obtained by using the above excitation displacement and output signal amplitude V(t).

[0061] Example

[0062] The specific parameters of the dynamic calibration device for the sensitivity of the linear angle vibration sensor based on multi-frequency sinusoidal excitation in this embodiment are: FSG (RIGOL DG 4202) is used as the signal generator 1 to generate a multi-frequency sinusoidal signal in the range of 0.01 to 10 Hz, the calibrated linear angle vibration sensor 5 (MSV 3000 and BW-VG 527), the long-stroke exciter (ESZ185-400) and the turntable (NIM-001) are the linear and angular vibration generating device 2 mentioned in the present invention, which is used to provide multi-frequency linear angle excitation for the calibrated linear angle vibration sensor 5, the lighting equipment 6 uses a 60W incandescent lamp, and a CMOS camera 7 (OS10-V3-4K) with an effective frame rate of 200fps and a DAQ card (NI USB6210) with an effective sampling rate of 1000hz are used as the data acquisition card 8 to collect the motion sequence image and output signal of the calibrated linear angle vibration sensor 5.

[0063] In order to verify the accuracy and speed of the dynamic calibration method of the sensitivity of the linear angular vibration sensor based on multi-frequency sinusoidal excitation provided by the present invention, the sensitivity calibration of the linear vibration sensor in the frequency range of 0.01 to 10 Hz and the angular vibration sensor in the frequency range of 0.01 to 5 Hz is realized by using the method of the present invention. FIG5 shows the sensitivity calibration results of the linear vibration sensor of the method of the present invention, the laser interferometer method (LI), and the traditional monocular vision method (MV); FIG6 shows the sensitivity calibration results of the angular vibration sensor of the method of the present invention, the laser interferometer method (LI), and the traditional monocular vision method (MV). From the results shown in FIG5 and FIG6, it can be seen that the sensitivity average values ​​of the method of the present invention are highly consistent with those of LI and MV, and the relative absolute deviations of the calibration results of the linear vibration sensor are less than 0.30% and 0.16%, respectively. The maximum relative standard deviation (RStd) of the method of the present invention is about 0.16%, which is less than 0.51% of LI and slightly greater than 0.14% of MV. In the calibration results of the angular vibration sensor, the maximum relative absolute deviations of the method of the present invention and LI and MV are 0.51% and 0.23% respectively, and the RStd of the method of the present invention is about 0.26%, while the RStd of LI and MV are 0.22% and 0.20% respectively. Although they have similar accuracy, as the frequency decreases, the calibration time cost of the method of the present invention drops sharply, and the efficiency can be improved by an order of magnitude.

[0064] The above description is a detailed description of the implementation examples of the present invention, which is not intended to limit the present invention in any form. Those skilled in the art can make a series of optimizations, improvements and modifications based on the present invention. Therefore, the protection scope of the present invention should be defined by the attached claims.

Claims

1. A method for dynamic calibration of sensitivity of a linear angle vibration sensor based on multi-frequency sinusoidal excitation, characterized in that: The calibration method comprises the following steps performed in sequence: 1) Construct a dynamic calibration device for the sensitivity of a linear angle vibration sensor based on multi-frequency sinusoidal excitation; 2) The high-contrast target and the calibrated line angle vibration sensor in the above-mentioned line angle vibration sensor sensitivity dynamic calibration device based on multi-frequency sinusoidal excitation are simultaneously fastened to the working table of the line and angle vibration generating device, and then a signal generator is used to generate a multi-frequency sinusoidal excitation signal with no mutual interference in the spectrum, so that the high-contrast target, the calibrated line angle vibration sensor and the working table of the line and angle vibration generating device have consistent motion characteristics, and the motion control of the line and angle vibration generating device is realized based on the excitation signal, so as to provide multi-frequency sinusoidal line angle excitation for the calibrated line angle vibration sensor; 3) Using a camera to collect the motion sequence images of the high-contrast target, using a data acquisition card to collect the output signal of the calibrated angular vibration sensor, and transmitting the motion sequence images and output signals to the image and signal processing and display unit; 4) The image and signal processing and display unit uses a template matching method to determine the ROI of the above-mentioned motion sequence image to eliminate the interference of background noise and similar edges, and then accurately extracts the motion feature edge in the ROI through the LSD method, and solves the line angle excitation displacement of the calibrated line angle vibration sensor based on the extracted motion feature edge; 5) SAM is used to fit the linear angle excitation displacement calculated above and the corresponding output signal of the calibrated linear angle vibration sensor. The excitation displacement corresponding to each frequency and the output signal amplitude of the calibrated linear angle vibration sensor are obtained by solving the sparse matrix composed of the overdetermined equations. Finally, the sensitivity of the calibrated linear angle vibration sensor is calculated using the excitation displacement and the output signal amplitude.

2. The method for dynamic calibration of sensitivity of a linear angle vibration sensor based on multi-frequency sinusoidal excitation according to claim 1 is characterized in that: In step 1), the dynamic calibration device for sensitivity of a linear angular vibration sensor based on multi-frequency sinusoidal excitation comprises a signal generator (1), a linear and angular vibration generating device (2), a high contrast target (4), a lighting device (6), a camera (7), a data acquisition card (8), and an image and signal processing and display unit (9); wherein the high contrast target (4) and the calibrated linear angular vibration sensor (5) are fastened to a working table (3) of the linear and angular vibration generating device (2); the lens of the camera (7) faces the high contrast target (4); the lighting device (6) ) is arranged on one side of the camera (7) and is used to provide lighting for the camera (7) to ensure the quality of the acquired image; the image and signal processing and display unit (9) is electrically connected to the signal generator (1), the camera (7) and the data acquisition card (8) respectively; the signal generator (1) and the data acquisition card (8) are also electrically connected to the linear and angular vibration generating device (2) and the calibrated linear angular vibration sensor (5) respectively; the signal generator (1) generates a multi-frequency sinusoidal signal to control the linear and angular vibration generating device (2) so that its working table (3) outputs multi-frequency sinusoidal linear angular excitation.

3. The method for dynamic calibration of sensitivity of a linear angle vibration sensor based on multi-frequency sinusoidal excitation according to claim 1 is characterized in that: In step 2), the frequency spectrum of the multi-frequency sinusoidal excitation signal x(t) is free of mutual interference. k Need to meet: Among them, f0 and p0 are the fundamental frequency and the fundamental prime number respectively, p i is a prime number greater than the base prime number p0, f k is the fundamental frequency f0 and the subsequent frequencies of the fundamental frequency f0 within the ten-fold frequency range, ω k and x k The frequencies are f k The corresponding angular frequency and amplitude, (-1) k Used to reduce the overshoot effect caused by the summation sign, t is the time.

4. The method for dynamic calibration of sensitivity of a linear angle vibration sensor based on multi-frequency sinusoidal excitation according to claim 2 is characterized in that: In step 4), the image and signal processing and display unit uses a template matching method to determine the ROI of the motion sequence image to eliminate the interference of background noise and similar edges, and then accurately extracts the motion feature edge in the ROI by the LSD method, and solves the line angle excitation displacement of the calibrated line angle vibration sensor according to the extracted motion feature edge. The method is: 4.1) The signal processing and display unit (9) reads the motion sequence images of the high contrast target (4) captured by the camera (7) {F j (x, y)}, using a nonlinear optimization algorithm based on the least squares principle, the distortion coefficient of the camera (7) is adjusted through continuous iteration to minimize the difference between the actual image and the theoretical image; 4.2) Use Harris corner detection algorithm to preliminarily locate the feature points in the image, and then use sub-pixel interpolation fitting method to obtain the relative coordinates of each feature point; 4.3) establishing a coordinate transformation linear equation group by collecting the relationship between the feature points on the high-contrast target (4) in the known world coordinate system and the pixel projection points thereof in the captured image, and then using the coordinate transformation linear equation group to transform the relative coordinates of each feature point into an absolute coordinate in the world coordinate system; 4.4) According to the absolute coordinates of the above feature points, the motion sequence image {F j (x,y)}; 4.5) Then find the position of the center of the circular area and determine the ROI; 4.6) Based on the LSD method, the straight line edge in the ROI is extracted, and the translation change and direction vector change of the straight line edge are estimated by combining the feature point matching between adjacent frame images, and then the line angle excitation displacement {s(t j )}.

5. The method for dynamic calibration of sensitivity of a linear angle vibration sensor based on multi-frequency sinusoidal excitation according to claim 4 is characterized in that: In step 5), the SAM method is used to fit the calculated line angle excitation displacement and the corresponding output signal of the calibrated line angle vibration sensor, and the excitation displacement corresponding to each frequency and the output signal amplitude of the calibrated line angle vibration sensor are obtained by solving the sparse matrix composed of the overdetermined equations. Finally, the sensitivity of the calibrated line angle vibration sensor is calculated using the excitation displacement and the output signal amplitude. The method is: Through the motion sequence image {F j (x,y)} to obtain the linear angle excitation displacement {s(t j )} is for each frequency f k The displacement amplitude is fitted using the following SAM method: s(t j )=Acos(ω k t j )-Bsin(ω k t j )+C (2) The equation in matrix form is: Among them, t j is the sampling time of the corresponding j-th frame image, A and B are the corresponding sinusoidal components, C is the offset component, and D is the corresponding matrix. By solving the overdetermined equations composed of m equations shown in formula (3), we can obtain the different frequencies f k The excitation displacement at the position and the output signal amplitude V(t) of the calibrated angular vibration sensor (5) are calculated; finally, the sensitivity of the calibrated angular vibration sensor (5) at different frequencies is obtained by using the excitation displacement and the output signal amplitude V(t).

Images