Wide-temperature-band anti-vibration type decoupling adjustable laser and compensation method thereof

Abstract

The application provides a wide-temperature-range anti-vibration decoupling adjustable laser and a compensation method thereof, and belongs to the technical field of laser optical path compensation; the wide-temperature-range anti-vibration decoupling adjustable laser comprises a laser light source, which is fixedly arranged at one end of a hollow cylinder; an anti-vibration ceramic pin unit, which has a pigtail end and a connector end, the pigtail end is fused to the laser light source, and the pigtail end is hingedly connected to the cylinder; a first calibration unit is arranged at the connector end of the anti-vibration ceramic pin unit and is used for adjusting the posture of the connector end; a collimating lens unit is arranged inside the cylinder; a second calibration unit is arranged between the collimating lens unit and the inside of the cylinder and is used for adjusting and locking the position of the collimating lens unit in the radial direction or the axial direction; the cylinder drives the anti-vibration ceramic pin unit and the collimating lens unit to synchronously rotate relative to the central axis of the cylinder; an image analysis unit is arranged outside the cylinder and on the light output side of the collimating lens unit and is used for evaluating the acquired image; and the first calibration unit or the second calibration unit is driven to act according to the evaluation result.

Description

Technical Field

[0001] This invention relates to the field of laser optical path compensation technology, and in particular to a wide-temperature-band vibration-resistant decoupled adjustable laser and its compensation method. Background Technology

[0002] Optical collimators are core optical components that convert diverging beams into parallel beams, and are widely used in many fields such as laser communication, precision measurement, and optical inspection. With the expansion of application scenarios, aerospace, automotive electronics, and industrial environments have placed higher demands on the environmental adaptability of optical collimators. They not only need to have high-precision collimation performance, but also need to be able to withstand wide-range temperature fluctuations and continuous vibration interference.

[0003] Existing optical collimators are mostly designed for conventional environments and lack specialized wide-temperature-range adaptation and vibration-resistant structural optimization, resulting in the following core technical defects: First, poor temperature adaptability. When the ambient temperature fluctuates over a wide range, the components are prone to thermal deformation and thermal stress due to differences in the thermal expansion coefficients of their materials, leading to optical coaxiality shift, a significant decrease in collimation accuracy, or even failure to function properly, making it difficult to adapt to extreme temperature change scenarios. Second, weak vibration resistance. In vibration environments such as industrial vibration and vehicle bumps, core components such as lenses and ceramic pins are prone to slight displacement and loosening of connections, and the adjustment structure is prone to accuracy drift, making it impossible to maintain a stable collimation state. This defect is particularly prominent in mid-to-high frequency vibration scenarios of 10~200Hz.

[0004] Therefore, it is essential to provide a wide-temperature-range vibration-resistant decoupled adjustable laser and its compensation method. By rationally setting the adjustment mechanism of the optical path and using visual monitoring combined with a piezoelectric micro-motion stage for attitude adjustment and position maintenance, it is necessary to improve the stability and reliability of the laser output. Summary of the Invention

[0005] In view of this, the present invention proposes a wide-temperature-range vibration-resistant decoupled adjustable laser and its compensation method, which can adapt to a wide temperature operating range, has the function of fine-tuning and maintaining the position of the core optical path device, and can achieve stable laser output.

[0006] On one hand, the present invention provides a wide-temperature-band vibration-resistant decoupled tunable laser, comprising: A laser light source, used to output laser light, is fixedly installed at one end of the hollow cylinder. The vibration-resistant ceramic pin unit has a fiber optic end and a connector end. The fiber optic end is fused to the laser source and is hinged to the cylinder. The connector end is used to transmit the laser. The first calibration unit is located at the connector end of the anti-vibration ceramic pin unit and is used to adjust the attitude of the connector end. The collimating lens unit is located inside the cylinder and is positioned on the light-emitting side of the vibration-resistant ceramic pin unit; The second calibration unit is located between the collimating lens unit and the inside of the cylinder, and is used to adjust and lock the position of the collimating lens unit in the radial or axial direction; the cylinder drives the anti-vibration ceramic pin unit and the collimating lens unit to rotate synchronously relative to the central axis of the cylinder; An image analysis unit, located outside the tube and on the light-emitting side of the collimating lens unit, is used to evaluate the acquired image and drive the first calibration unit or the second calibration unit to operate based on the evaluation result.

[0007] Based on the above technical solutions, preferably, the first calibration unit includes a two-dimensional piezoelectric micro-motion stage and a first locking device; the two-dimensional piezoelectric micro-motion stage is also equipped with an orthogonally arranged strain sensor and a first position sensor; the two-dimensional piezoelectric micro-motion stage is used to adjust the attitude of the connector end, the strain sensor is used to sense vibration or structural deformation, and the first locking device is used to lock the current real-time position of the connector end.

[0008] Preferably, the second calibration unit includes a three-axis moving device, a second position sensor, and a second position locking device. The three-axis moving device is used to adjust the radial or axial position of the collimating lens unit, the second position sensor is used to acquire the real-time position of the collimating lens unit, and the second position locking device is used to lock the current position of the collimating lens unit.

[0009] Preferably, the image analysis unit includes an optical path switching module, which includes a collimating lens, a dichroic mirror, a focusing lens, a 4f relay lens module, and a camera. The collimating lens, dichroic mirror, focusing lens, and camera are sequentially spaced and coaxially arranged to obtain a far-field optical path for far-field imaging. The collimating lens, dichroic mirror, 4f relay lens module, and camera are also sequentially spaced and coaxially arranged to obtain a near-field optical path for near-field imaging. Far-field and near-field images are obtained based on the far-field and near-field optical paths, respectively, and the image analysis unit evaluates the far-field and near-field images.

[0010] Preferably, the 4f relay lens module includes an achromatic cemented doublet lens and a microscope objective, wherein the achromatic cemented doublet lens and the microscope objective are coaxially arranged, and the microscope objective is located on the light-inlet side or the light-outlet side of the achromatic cemented doublet lens.

[0011] On the other hand, the present invention also provides a compensation method for a wide-temperature-band vibration-resistant decoupled adjustable laser, comprising the following steps: S1: Configure the aforementioned wide-temperature-range vibration-resistant decoupled adjustable laser; configure a first calibration unit at the connector end of the vibration-resistant ceramic ferrule unit. The first calibration unit includes a two-dimensional piezoelectric micro-stage, a first position sensor, and a first locking device. The two-dimensional piezoelectric micro-stage is used to drive the connector end to move radially relative to the pigtail end. The first position sensor is used to acquire the real-time position of the connector end, and the first locking device is used to lock the current position of the connector end. Configure a second calibration unit between the collimating lens unit and the cylinder. The second calibration unit includes a three-axis moving device, a second position sensor, and a second locking device. The three-axis moving device is used to drive... The collimating lens unit moves radially or axially relative to the cylinder. A second position sensor is used to acquire the real-time position of the collimating lens unit, and a second locking device is used to lock the current position of the collimating lens unit. A temperature sensor is also installed inside the cylinder. A rotation device is installed on the outside of the cylinder, which drives the cylinder to rotate synchronously with the anti-vibration ceramic pin unit and the collimating lens unit. The connector end has two degrees of freedom in the X and Y directions, and the collimating lens unit has three degrees of freedom in the X, Y, and Z directions. The X and Y axes are two orthogonal directions of the radial section of the cylinder, and the Z axis is the axial direction of the cylinder, which coincides with the virtual optical axis of the laser. S2: Unlock the two-dimensional piezoelectric micro-motion stage, the first locking device, the three-axis moving device and the second locking device, and return the connector end and the collimating lens unit to their initial positions and maintain them; configure the image analysis unit on the light-emitting side of the collimating lens unit, start the rotation device, and use the near-field optical path and the far-field optical path to collect near-field images and far-field images respectively for every 10° rotation of the cylinder; S3: Preprocess the acquired near-field image to obtain a preprocessed near-field image; preprocess the acquired far-field image to obtain a preprocessed far-field image. S4: Based on the preprocessed near-field and far-field images, perform near-field trajectory analysis and far-field trajectory analysis respectively; obtain the centroid coordinates of the light spot in each near-field image and fit it to a first virtual circle to obtain the pixel coordinates of the center of the first virtual circle; obtain the centroid coordinates of the light spot in each far-field image and fit it to a second virtual circle to obtain the pixel coordinates of the center of the second virtual circle; then determine the defocus modulation depth according to whether the radius of the light spot in the far-field image changes with the rotation angle of the cylinder; finally, determine the ellipticity modulation amplitude according to whether the ellipticity of the light spot in the far-field image changes with the rotation angle of the cylinder. A decision matrix is ​​constructed by combining the pixel coordinates of the center of the first virtual circle, the pixel coordinates of the center of the second virtual circle, the defocus modulation depth, and the ellipticity modulation amplitude. S5: Based on the content of the decision matrix, adjust the position of the connector end or collimating lens unit, and then use the first locking device and the second locking device to fix the connector end and the collimating lens unit respectively. Start the rotation device again, and re-acquire near-field and far-field images every 10° rotation of the cylinder, and execute steps S3 and S4 again to verify the result of the adjustment of the position of the connector end or collimating lens unit.

[0012] Preferably, the preprocessing of the acquired near-field image described in step S3 to obtain a preprocessed near-field image specifically includes the following: constructing a theoretical point spread function containing astigmatism terms; introducing temperature sensor data as a priori astigmatism coefficients in Richardson-Lucy deconvolution; performing iterative deconvolution of the asymmetric PSF to restore internal structural details lost due to saturation; eliminating spot elliptic distortion caused by thermal effects; and obtaining a restored near-field image. Then, a physical model is established for the background image, and the physical model is refined through a neural network. Through multiple forward propagations, the background image is output, and the background image is subtracted from the restored near-field image to restore the true spot size.

[0013] Preferably, the preprocessing of the acquired far-field image in step S3 to obtain the preprocessed far-field image specifically includes the following: acquiring data from the strain sensor at the connector end, estimating the vibration trajectory, constructing a theoretical point spread function (PSF) based on the vibration trajectory, and obtaining the restored far-field image through a deconvolution algorithm; then, referring to the step of removing the background image from the near-field image, removing the background image from the restored far-field image; then, using energy weighted phase consistency, introducing pixel intensity as a weight into phase calculation, extracting the contour corresponding to the energy threshold, fitting the contour, and extracting the centroid coordinates of the far-field image.

[0014] Preferably, the decision matrix includes the following: 1) When the pixel coordinates of the center of the first virtual circle The static eccentricity with respect to the cylinder's rotation axis is not less than the first threshold and the pixel coordinates of the center of the second virtual circle. The static eccentricity with respect to the rotation axis is not less than the second threshold, and the defocus modulation depth and ellipticity modulation amplitude When all values ​​are less than the set threshold, it is determined that the connector end is significantly off-center. The position of the connector end is adjusted only through the first calibration unit, and the current position is locked by the first locking device after adjustment. 2) When the pixel coordinates of the center of the first virtual circle The pixel coordinates of the center of the second virtual circle whose static eccentricity with respect to the cylinder's rotation axis is less than the first threshold. The static eccentricity with respect to the rotation axis is not less than the second threshold, and the defocus modulation depth and ellipticity modulation amplitude When all values ​​are less than the set threshold, it is determined that the connector end is tilted. The position of the connector end is adjusted only through the first calibration unit, and the current position is locked using the first locking device after adjustment. 3) When the pixel coordinates of the center of the first virtual circle The pixel coordinates of the center of the second virtual circle whose static eccentricity with respect to the cylinder's rotation axis is less than the first threshold. The static eccentricity with respect to the rotation axis is less than the second threshold, and the ellipticity modulation amplitude is... Not less than the set threshold, defocus modulation depth When the value is less than the set threshold, it is determined that the collimating lens unit is eccentric. The X-axis and Y-axis positions of the collimating lens unit are adjusted only by the second calibration unit, and the current position is locked by the second locking device after adjustment. 4) When the pixel coordinates of the center of the first virtual circle The pixel coordinates of the center of the second virtual circle whose static eccentricity with respect to the cylinder's rotation axis is less than the first threshold. The static eccentricity with respect to the rotation axis is less than the second threshold, and the ellipticity modulation amplitude is... Defocus modulation depth is less than the set threshold. If the value is not less than the set threshold, it is considered that the collimating lens unit is axially defocused. The Z-axis position of the collimating lens unit is adjusted only by the second calibration unit, and the current position is locked by the second locking device after adjustment. 5) When the pixel coordinates of the center of the first virtual circle The static eccentricity with respect to the cylinder's rotation axis is not less than the first threshold and the pixel coordinates of the center of the second virtual circle. The static eccentricity with respect to the rotation axis is not less than the second threshold, and the ellipticity modulation amplitude is... Not less than the set threshold, but the defocus modulation depth When the value is less than the set threshold, it is determined that the connector end is eccentric and the collimating lens unit is eccentric. Then, the first calibration unit is used to adjust the position of the connector end, and the first locking device is used to lock the current position after adjustment. Then, the X-axis and Y-axis positions of the collimating lens unit are adjusted by the second calibration unit, and the second locking device is used to lock the current position after adjustment. 6) When the pixel coordinates of the center of the first virtual circle The pixel coordinates of the center of the second virtual circle whose static eccentricity with respect to the cylinder's rotation axis is less than the first threshold. The static eccentricity with respect to the rotation axis is not less than the second threshold, and the ellipticity modulation amplitude is... Not less than the set threshold, defocus modulation depth When the value is less than the set threshold, it is determined that the connector end is tilted and the collimating lens unit is eccentric. Then, the first calibration unit is used to adjust the position of the connector end, and the first locking device is used to lock the current position after adjustment. Then, the X-axis and Y-axis positions of the collimating lens unit are adjusted by the second calibration unit, and the second locking device is used to lock the current position after adjustment. 7) When the pixel coordinates of the center of the first virtual circle The pixel coordinates of the center of the second virtual circle whose static eccentricity with respect to the cylinder's rotation axis is less than the first threshold. The static eccentricity with respect to the rotation axis is not less than the second threshold, and the ellipticity modulation amplitude is... Defocus modulation depth is less than the set threshold. If the value is not less than the set threshold, and it is determined that the connector end is tilted and the collimating lens unit is defocused axially, the first calibration unit is used to adjust the position of the connector end, and the first locking device is used to lock the current position after adjustment. Then, the Z-axis position of the collimating lens unit is adjusted by the second calibration unit, and the second locking device is used to lock the current position after adjustment. 8) When the pixel coordinates of the center of the first virtual circle The pixel coordinates of the center of the second virtual circle whose static eccentricity with respect to the cylinder's rotation axis is less than the first threshold. The static eccentricity with respect to the rotation axis is less than the second threshold, and the ellipticity modulation amplitude is... Not less than the set threshold, defocus modulation depth If the value is not less than the set threshold, it is determined that the collimating lens unit is eccentric and out of focus. The X-axis and Y-axis positions of the collimating lens unit are first adjusted by the second calibration unit, and then the Z-axis position is adjusted. After adjustment, the current position is locked by the second locking device. 9) When the pixel coordinates of the center of the first virtual circle The static eccentricity with respect to the cylinder's rotation axis is not less than the first threshold and the pixel coordinates of the center of the second virtual circle. The static eccentricity with respect to the rotation axis is not less than the second threshold, and the ellipticity modulation amplitude is... Not less than the set threshold, defocus modulation depth If the value is not less than the set threshold, it is determined that the connector end and the collimating lens unit are completely misaligned. Then, the first calibration unit is used to adjust the position of the connector end, and the first locking device is used to lock the current position after adjustment. The second calibration unit is used to adjust the X-axis and Y-axis positions of the collimating lens unit first, and then adjust the Z-axis position. After adjustment, the second locking device is used to lock the current position.

[0015] Preferably, the first threshold is 10% of the spot diameter in the near-field image; the second threshold is 20% of the reference radius of the spot in the far-field image; and the ellipticity modulation amplitude... The set threshold is 0.1; defocus modulation depth The threshold is set to 5% of the reference radius of the spot in the far-field image.

[0016] The present invention provides a wide-temperature-band vibration-resistant decoupled adjustable laser and its compensation method, which have the following advantages compared with the prior art: 1. This invention, by configuring a two-degree-of-freedom angle adjustment of the anti-vibration ceramic pin unit and combining it with a three-axis position adjustment of the collimating lens unit, constitutes a five-degree-of-freedom adjustment mechanism. The cylinder synchronously drives the anti-vibration ceramic pin unit and the collimating lens unit to rotate, and combined with near-field and far-field optical path imaging, it can achieve precise positioning and independent compensation of the error source; with the locking device, the compensation result can be locked, taking into account both adjustment accuracy and long-term stability.

[0017] 2. The near-field and far-field optical paths use only one camera. Switching between the near-field and far-field optical paths is very convenient, requiring only the replacement of some components in the optical path.

[0018] 3. The compensation strategy is based on image analysis of the rotation process, distinguishing various errors such as static eccentricity, dynamic tilt, defocus, etc. and their combinations. By formulating a decision matrix, it can deal with all possible fault combinations and propose corresponding correction measures.

[0019] 4. The preprocessing process driven by the physical model improves the accuracy of rotated image analysis and lays the foundation for the rationality of the decision matrix. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a schematic diagram of the overall structure of a wide-temperature-band vibration-resistant decoupled adjustable laser and its compensation method according to the present invention. Figure 2 This is a schematic diagram of the near-field optical path of a wide-temperature-band vibration-resistant decoupled adjustable laser and its compensation method according to the present invention. Figure 3 This is a schematic diagram of the far-field optical path of a wide-temperature-band vibration-resistant decoupled adjustable laser and its compensation method according to the present invention. Figure 4 This is a flowchart illustrating the steps of a wide-temperature-band vibration-resistant decoupled adjustable laser and its compensation method according to the present invention. Detailed Implementation

[0022] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0023] Existing optical collimators are mostly designed for conventional environments and lack specialized wide-temperature-range adaptation and vibration-resistant structural optimization, resulting in the following core technical defects: First, poor temperature adaptability. When the ambient temperature fluctuates over a wide range, the components are prone to thermal deformation and thermal stress due to differences in the thermal expansion coefficients of their materials, leading to optical coaxiality shift, a significant decrease in collimation accuracy, or even failure to function properly, making it difficult to adapt to extreme temperature change scenarios. Second, weak vibration resistance. In vibration environments such as industrial vibration and vehicle bumps, core components such as lenses and ceramic pins are prone to slight displacement and loosening of connections, and the adjustment structure is prone to accuracy drift.

[0024] In view of this, such as Figure 1 Combination Figure 2 and Figure 3 As shown, in one aspect, the present invention provides a wide-temperature-band vibration-resistant decoupled tunable laser, comprising: The laser source, used to output laser light, is fixedly installed at one end of the hollow cylinder. In order to maintain stable operation, the laser source requires an additional temperature stabilization component, which will not be described in detail here.

[0025] The vibration-resistant ceramic pin unit has a fiber optic end and a connector end. The fiber optic end is fused to the laser source and is hinged to the cylinder. The connector end is used to transmit the laser. Here, the fiber optic end is set as a hinged connection, strictly coaxial with the central axis of the cylinder. The connector end at the other end is a free end with two degrees of freedom in two directions, but it cannot move axially.

[0026] The first calibration unit is located at the connector end of the anti-vibration ceramic pin unit and is used to adjust the attitude of the connector end. The collimating lens unit is located inside the cylinder and is positioned on the light-emitting side of the anti-vibration ceramic pin unit; the cylinder drives the anti-vibration ceramic pin unit and the collimating lens unit to rotate synchronously relative to the central axis of the cylinder.

[0027] The second calibration unit is located between the collimating lens unit and the inside of the cylinder, and is used to adjust and lock the position of the collimating lens unit in the radial or axial direction. An image analysis unit, located outside the tube and on the light-emitting side of the collimating lens unit, is used to evaluate the acquired image and drive the first calibration unit or the second calibration unit to operate based on the evaluation result.

[0028] In one embodiment, the first calibration unit includes a two-dimensional piezoelectric micro-stage and a first locking device. The two-dimensional piezoelectric micro-stage is further equipped with an orthogonally arranged strain sensor and a first position sensor. The two-dimensional piezoelectric micro-stage is used to adjust the orientation of the connector end, such as pitch and yaw directions. The strain sensor is used to sense vibration or dynamic structural deformation, while the first position sensor is used to acquire the static position of the connector end for real-time positioning. The first locking device is used to lock the current real-time position of the connector end. The two-dimensional piezoelectric micro-stage can achieve nanometer-level fine-tuning and has a fast response speed.

[0029] To isolate the effects of vibration, low-frequency vibration isolators can be installed at the interface between the laser and the support base of the cylinder to attenuate the transmission path of vibrations above 10Hz. If strain sensors detect the influence of mid-to-high frequency vibrations of 10~200Hz during normal use, the vibration isolation measures should be checked to see if they have failed.

[0030] In one embodiment, the second calibration unit includes a three-axis moving device, a second position sensor, and a second position locking device. The three-axis moving device is used to adjust the radial or axial position of the collimating lens unit, the second position sensor is used to acquire the real-time position of the collimating lens unit, and the second position locking device is used to lock the current position of the collimating lens unit. The three-axis moving device can be a piezoelectric ceramic actuator, which can be guided by a pre-configured guide rail during axial movement, and combined with position feedback means such as a grating ruler as the second position sensor to achieve position feedback, which will not be elaborated further here. The first and second position locking devices can use piezoelectric ceramic locks, which utilize the inverse piezoelectric effect of piezoelectric ceramics to generate huge locking force, resulting in extremely high rigidity after locking.

[0031] The image analysis unit includes an optical path switching module, which comprises a collimating lens, a dichroic mirror, a focusing lens, a 4f relay lens module, and a camera. The collimating lens, dichroic mirror, focusing lens, and camera are sequentially spaced and coaxially arranged to obtain a far-field optical path for far-field imaging. Similarly, the collimating lens, dichroic mirror, 4f relay lens module, and camera are sequentially spaced and coaxially arranged to obtain a near-field optical path for near-field imaging. Far-field and near-field images are obtained based on the far-field and near-field optical paths, respectively, and the image analysis unit evaluates these images. The dichroic mirror is not used for color separation here but rather as a fixed 45° reflector. It is crucial to ensure that the image plane of the focusing lens strictly coincides with the output image plane of the 4f system. This means that after switching between the near-field and far-field optical paths, the position of the camera target surface remains absolutely fixed; only the current position of the focusing lens or the 4f relay lens module needs to be changed. Since only one camera is used, the switching process is greatly simplified, eliminating the need for frequent calibration.

[0032] The 4f relay lens module includes an achromatic cemented doublet and a microscope objective. The achromatic cemented doublet and the microscope objective are coaxially positioned, with the microscope objective located on either the light-inlet or light-outlet side of the achromatic cemented doublet. A standard 4f relay lens module consists of two objective lenses L1 and an image lens L2 with the same focal length, arranged at a specific interval. Objective lens L1 receives parallel light from the collimating lens, forming a Fourier spectrum plane on its rear focal plane. Image lens L2, located behind objective lens L1, reconstructs the spectrum plane into a real image and projects it onto the camera target surface. The distance between objective lens L1 and image lens L2 is 2f, where f is the lens focal length, and the total distance from the object plane to the image plane is 4f. To avoid an excessively small image size, a microscope objective is added. The microscope objective can be positioned on either the light-inlet or light-outlet side of the achromatic cemented doublet.

[0033] In addition, such as Figure 4 As shown, the present invention also provides a compensation method for a wide-temperature-band vibration-resistant decoupled adjustable laser, comprising the following steps: S1: Configure the aforementioned wide-temperature-range vibration-resistant decoupled adjustable laser; configure a first calibration unit at the connector end of the vibration-resistant ceramic ferrule unit. The first calibration unit includes a two-dimensional piezoelectric micro-stage, a first position sensor, and a first locking device. The two-dimensional piezoelectric micro-stage is used to drive the connector end to move radially relative to the pigtail end. The first position sensor is used to acquire the real-time position of the connector end, and the first locking device is used to lock the current position of the connector end. Configure a second calibration unit between the collimating lens unit and the cylinder. The second calibration unit includes a three-axis moving device, a second position sensor, and a second locking device. The three-axis moving device is used to drive... The collimating lens unit moves radially or axially relative to the cylinder. The second position sensor is used to obtain the real-time position of the collimating lens unit, and the second locking device is used to lock the current position of the collimating lens unit. A temperature sensor is also installed inside the cylinder. A rotation device is installed on the outside of the cylinder, which is used to drive the cylinder to rotate synchronously with the anti-vibration ceramic pin unit and the collimating lens unit. The connector end has two degrees of freedom in the X and Y directions, and the collimating lens unit has three degrees of freedom in the X, Y and Z directions. The X and Y axes are two orthogonal directions of the radial section of the cylinder, and the Z axis is the axial direction of the cylinder, which coincides with the virtual optical axis of the laser.

[0034] S2: Unlock the two-dimensional piezoelectric micro-motion stage, the first locking device, the three-axis moving device and the second locking device, and return the connector end and the collimating lens unit to their initial positions and maintain them; configure the image analysis unit on the light-emitting side of the collimating lens unit, start the rotation device, and use the near-field optical path and the far-field optical path to collect near-field images and far-field images respectively for every 10° rotation of the cylinder.

[0035] S3: Preprocess the acquired near-field image to obtain the preprocessed near-field image; preprocess the acquired far-field image to obtain the preprocessed far-field image.

[0036] The preprocessing of the near-field image specifically includes the following: Constructing a theoretical point spread function (PSF) containing astigmatism terms; introducing temperature sensor data as a priori for astigmatism coefficients in the Richardson-Lucy deconvolution; performing iterative deconvolution of the asymmetric PSF to recover internal structural details lost due to saturation; eliminating the elliptic distortion of the light spot caused by thermal effects; and obtaining the restored near-field image. Then, a physical model is established for the background image, and refined using a neural network. Through multiple forward propagations, the background image is output. The background image is subtracted from the restored near-field image to restore the true light spot size. Specific details are as follows.

[0037] First, a theoretical point spread function (PSF) including astigmatism coefficients is modeled. ,in For pixel coordinates, T For temperature sensor readings, The temperature-dependent astigmatism coefficient, , T 0 is the reference temperature, usually 25 degrees Celsius. The intrinsic astigmatism coefficient at room temperature, The temperature sensitivity coefficient is calculated by measuring the temperature at different temperature points. Acquire standard point light source images and fit them using least squares. Obtain the coefficient , Temperature point Ellipticity fitting of the point light source image; These are the width parameters in the X and Y directions. This represents the cross term coefficient.

[0038] Introducing temperature sensor data as a priori for the astigmatism coefficient in the Richardson-Lucy deconvolution is used to construct a temperature-adaptive theoretical point spread function (PSF) that includes the astigmatism coefficient: , The current temperature measured by the temperature sensor. This is a linear mapping function based on temperature changes, meaning the current temperature... Not equal to calibration temperature The optical parameters at the current temperature are derived from the optical parameters at the calibration temperature using interpolation. This indicates parametric interpolation operations, such as cubic spline interpolation; the iterative update content for a regularized image is: ,in , For the firstk Second and third k The image estimated in +1 iterations, The image represents the acquired near-field image, where the forward slash / indicates element-wise division and * indicates convolution. It is a temperature-adaptive theoretical point spread function that includes astigmatism coefficients. The flip, It is a regularization term. , For gradient smoothing weights, The temperature is the prior weight, and all values ​​are real numbers. The gradient of the image describes the spatial rate of change of light intensity distribution within the image. Historical images related to temperature; It is the theoretical point spread function including astigmatism coefficients at the current temperature; the convergence criterion for the regularized image is... ,in It is a second-order norm. The basic convergence threshold has a range of 10. -4 -10 -6 , This is the temperature change sensitivity coefficient, with a value ranging from 0.05 to 0.2. The image represents the temperature change. Iterative updates of the image, using regularization, continue until the convergence criterion is met or the maximum number of iterations is reached, at which point the restored image is obtained. By iteratively restoring the internal structural details lost due to saturation, the elliptic distortion of the light spot caused by thermal effects is eliminated.

[0039] Then, physical modeling is performed on the background image: , The initial physical model for the background image, the thermal diffusion background model. ,in T For temperature sensor readings, For laser driving current, This is the thermal response coefficient, expressed in degrees Celsius per second, and takes the value of a positive real number. t For time, is the thermal time constant, in seconds, and takes the value of a positive real number; It is a thermal response spatial distribution template, preset in the camera's factory settings; halo thermal diffusion model. , Preliminary light spot intensity estimation is performed for the unsaturated region; then, the U-net neural network is used for refinement, based on the restored image. Initial physical model of background image Temperature sensor readings T Laser driving current saturation maskM sat Constructing input vectors saturation mask M sat It's a binary image; if the current pixel is saturated, the mask value is 1; otherwise, it's 0. This is used to read 1D temperature sensor readings. T and laser drive current The corresponding fully connected layers are each input and mapped to the feature space, and then compared with the image features to obtain the high-dimensional reconstructed image. Initial physical model of background image and saturation mask M sat The data is then concatenated and fused, and the fused result is fed into a convolutional neural network (CNN). The CNN outputs the residual. The initial physical model is corrected by residuals. .

[0040] Obtained through multiple sampling and processing N Zhang's final background image sequence , i This is the final background image sequence number. i =1, 2, ..., N The final background estimate is Uncertainty diagram Establish an adaptive filtering model for the background image. , This is a smoothed background image after online filtering. This is the background image of the final output from the previous time step. This is the instantaneously estimated background image at the current moment; the final background estimate can usually be used. , This is an adaptive factor, with a value range of [0, 1]. , It is the basic balance coefficient, and its value is a positive real number. For sensitivity, The rate of temperature change is represented by a real number.

[0041] During the training of the U-net neural network, the training objective is to correct the initial physical model using residuals. To get as close as possible to a realistic background image , making the loss function Minimum, of which M Correcting the initial physical model for residuals Or a real background image The total number of pixels in Correcting the initial physical model for residuals Or a real background image The pixel coordinates of each point in the dataset. Additionally, the training batch size of the U-net neural network can be further configured to 50-100 epochs, with a batch size of 8 and a learning rate of 10. -4 .

[0042] From the restored image Smoothed background image after online filtering After that, the preprocessed near-field image is obtained.

[0043] The preprocessing of the far-field image specifically includes the following steps: acquiring data from the strain sensor at the connector end, estimating the vibration trajectory, constructing a theoretical point spread function (PSF) based on the vibration trajectory, and obtaining the restored far-field image through a deconvolution algorithm; then, referring to the steps of removing background images from the near-field image, removing the background image from the restored far-field image; finally, using energy-weighted phase consistency, incorporating pixel intensity as a weight into phase calculation, extracting the contour corresponding to the energy threshold, fitting the contour, and extracting the centroid coordinates of the far-field image. Specific details include the following.

[0044] Based on the motion fuzzy point spread function of micro-vibration trajectories, a theoretical point spread function (PSF) is constructed: ,in For the Dirac function, Indicates in t The center of the light spot at that moment is exactly located at point. Place, This is the normalization coefficient.

[0045] The iterative formula for reconstructing the far-field image by independently using Richardson-Lucy deconvolution is: The acquired far-field image is The forward slash / in square brackets indicates pixel-wise division, and * indicates convolution operation; The theoretical point spread function The flipping; For the first l Second and third l Image estimated at +1 iteration; far-field image obtained by Richardson-Lucy deconvolution algorithm. Iterative restoration was performed to obtain a clear far-field spot image.

[0046] Using a method similar to that used for near-field images, the background image is removed from the clear far-field spot image. Then, pixel intensity is used as a weight in the current far-field spot image to enhance the contour response of high-energy regions. The expression for phase consistency is rewritten to obtain the formula for calculating phase consistency with energy weights: ,in For local energy, For amplitude, To prevent positive real numbers with a denominator of zero, the weighting term... ,in This is the gain factor, with a value range of 10-15; The intensity threshold is set, and the normalization value ranges from 0.15 to 0.25. The pixel intensity value is used; thresholding is performed based on the phase consistency calculation result of the energy weight to obtain a binary contour map. , The threshold for phase consistency results is recommended to be 0.3-0.5; This represents the lower limit of pixel intensity, with a normalization value ranging from 0.01 to 0.05.

[0047] For binary contour maps In the connected components, the centroid is calculated using pixel intensity as the weight, and the centroid coordinates are: , ,in and The pixel coordinates of each pixel in the connected component. x Values ​​and pixel coordinates y value, Let be the intensity of each pixel in the connected component. .

[0048] S4: Based on the preprocessed near-field and far-field images, perform near-field trajectory analysis and far-field trajectory analysis respectively; obtain the centroid coordinates of the light spot in each near-field image and fit it to a first virtual circle to obtain the pixel coordinates of the center of the first virtual circle; obtain the centroid coordinates of the light spot in each far-field image and fit it to a second virtual circle to obtain the pixel coordinates of the center of the second virtual circle; then determine the defocus modulation depth according to whether the radius of the light spot in the far-field image changes with the rotation angle of the cylinder; finally, determine the ellipticity modulation amplitude according to whether the ellipticity of the light spot in the far-field image changes with the rotation angle of the cylinder. For near-field trajectory analysis, the centroid coordinates of the light spot in each near-field image are obtained. The rotation angle of the cylinder is , If the phase is given, then the centroid coordinates of the light spot in the near-field image are... The calculation formula is: , To determine the static eccentricity of the connector end relative to the cylinder's rotation axis, a first virtual circle is fitted to the centroid coordinates of all light spots to obtain the pixel coordinates of the center of the first virtual circle. The position compensation amount of the connector end is determined based on the center coordinates of the first virtual circle. For the far-field trajectory, the centroid coordinates of the light spot in the far-field image are extracted. Similarly, based on the centroid coordinates of the light spot in the far-field image... Perform a second virtual circle fitting to obtain the pixel coordinates of the center of the second virtual circle. The tilt angle of the connector end is calculated according to the following formula. , ,in L The equivalent optical distance from near field to far field is calculated. The radius of the light spot in each far-field image is obtained, and the relationship between the radius and the rotation angle of the cylinder is analyzed. If the relationship between the radius of the light spot in the far-field image and the rotation angle of the cylinder is constant, it indicates no defocus. If the radius of the light spot in the far-field image is a function of the rotation angle of the cylinder... This indicates that there is defocusing, among which r 0 represents the reference radius of the light spot in the far-field image; The defocus modulation depth is half the amplitude of the spot radius transformation; The defocus phase is used; the ellipticity of the light spot in each far-field image is calculated. If the ellipticity of the light spot in the far-field image is constant, it indicates no astigmatism. If the ellipticity of the light spot in the far-field image is a function of the rotation angle of the cylinder, then... , indicating the existence of astigmatism, where Mean ellipticity; This is the ellipticity modulation amplitude, used to describe the severity of astigmatism; The astigmatic phase angle indicates the direction of the principal axis of astigmatism.

[0049] A decision matrix is ​​constructed by combining the pixel coordinates of the center of the first virtual circle, the pixel coordinates of the center of the second virtual circle, the defocus modulation depth, and the ellipticity modulation amplitude.

[0050] The decision matrix includes the following: 1) When the pixel coordinates of the center of the first virtual circle The static eccentricity with respect to the cylinder's rotation axis is not less than the first threshold and the pixel coordinates of the center of the second virtual circle. The static eccentricity with respect to the rotation axis is not less than the second threshold, and the defocus modulation depth and ellipticity modulation amplitude When all values ​​are less than the set threshold, it is determined that the connector end is significantly off-center. The position of the connector end is adjusted only through the first calibration unit, and the current position is locked by the first locking device after adjustment. 2) When the pixel coordinates of the center of the first virtual circle The pixel coordinates of the center of the second virtual circle whose static eccentricity with respect to the cylinder's rotation axis is less than the first threshold. The static eccentricity with respect to the rotation axis is not less than the second threshold, and the defocus modulation depth and ellipticity modulation amplitude When all values ​​are less than the set threshold, it is determined that the connector end is tilted. The position of the connector end is adjusted only through the first calibration unit, and the current position is locked using the first locking device after adjustment. 3) When the pixel coordinates of the center of the first virtual circle The pixel coordinates of the center of the second virtual circle whose static eccentricity with respect to the cylinder's rotation axis is less than the first threshold. The static eccentricity with respect to the rotation axis is less than the second threshold, and the ellipticity modulation amplitude is... Not less than the set threshold, defocus modulation depth When the value is less than the set threshold, it is determined that the collimating lens unit is eccentric. The X-axis and Y-axis positions of the collimating lens unit are adjusted only by the second calibration unit, and the current position is locked by the second locking device after adjustment. 4) When the pixel coordinates of the center of the first virtual circle The pixel coordinates of the center of the second virtual circle whose static eccentricity with respect to the cylinder's rotation axis is less than the first threshold. The static eccentricity with respect to the rotation axis is less than the second threshold, and the ellipticity modulation amplitude is... Defocus modulation depth is less than the set threshold. If the value is not less than the set threshold, it is considered that the collimating lens unit is axially defocused. The Z-axis position of the collimating lens unit is adjusted only by the second calibration unit, and the current position is locked by the second locking device after adjustment. 5) When the pixel coordinates of the center of the first virtual circle The static eccentricity with respect to the cylinder's rotation axis is not less than the first threshold and the pixel coordinates of the center of the second virtual circle. The static eccentricity with respect to the rotation axis is not less than the second threshold, and the ellipticity modulation amplitude is... Not less than the set threshold, but the defocus modulation depth When the value is less than the set threshold, it is determined that the connector end is eccentric and the collimating lens unit is eccentric. Then, the first calibration unit is used to adjust the position of the connector end, and the first locking device is used to lock the current position after adjustment. Then, the X-axis and Y-axis positions of the collimating lens unit are adjusted by the second calibration unit, and the second locking device is used to lock the current position after adjustment. 6) When the pixel coordinates of the center of the first virtual circle The pixel coordinates of the center of the second virtual circle whose static eccentricity with respect to the cylinder's rotation axis is less than the first threshold. The static eccentricity with respect to the rotation axis is not less than the second threshold, and the ellipticity modulation amplitude is... Not less than the set threshold, defocus modulation depth When the value is less than the set threshold, it is determined that the connector end is tilted and the collimating lens unit is eccentric. Then, the first calibration unit is used to adjust the position of the connector end, and the first locking device is used to lock the current position after adjustment. Then, the X-axis and Y-axis positions of the collimating lens unit are adjusted by the second calibration unit, and the second locking device is used to lock the current position after adjustment. 7) When the pixel coordinates of the center of the first virtual circle The pixel coordinates of the center of the second virtual circle whose static eccentricity with respect to the cylinder's rotation axis is less than the first threshold. The static eccentricity with respect to the rotation axis is not less than the second threshold, and the ellipticity modulation amplitude is... Defocus modulation depth is less than the set threshold. If the value is not less than the set threshold, and it is determined that the connector end is tilted and the collimating lens unit is defocused axially, the first calibration unit is used to adjust the position of the connector end, and the first locking device is used to lock the current position after adjustment. Then, the Z-axis position of the collimating lens unit is adjusted by the second calibration unit, and the second locking device is used to lock the current position after adjustment. 8) When the pixel coordinates of the center of the first virtual circle The pixel coordinates of the center of the second virtual circle whose static eccentricity with respect to the cylinder's rotation axis is less than the first threshold. The static eccentricity with respect to the rotation axis is less than the second threshold, and the ellipticity modulation amplitude is... Not less than the set threshold, defocus modulation depth If the value is not less than the set threshold, it is determined that the collimating lens unit is off-center and out of focus. The X-axis and Y-axis positions of the collimating lens unit are first adjusted by the second calibration unit, and then the Z-axis position is adjusted. After adjustment, the current position is locked by the second locking device. 9) When the pixel coordinates of the center of the first virtual circle The static eccentricity with respect to the cylinder's rotation axis is not less than the first threshold and the pixel coordinates of the center of the second virtual circle. The static eccentricity with respect to the rotation axis is not less than the second threshold, and the ellipticity modulation amplitude is... Not less than the set threshold, defocus modulation depth If the value is not less than the set threshold, it is determined that the connector end and the collimating lens unit are completely misaligned. Then, the first calibration unit is used to adjust the position of the connector end, and after adjustment, the first locking device is used to lock the current position. The second calibration unit is used to adjust the X-axis and Y-axis positions of the collimating lens unit first, and then adjust the Z-axis position. After adjustment, the second locking device is used to lock the current position.

[0051] The first threshold is 10% of the spot diameter in the near-field image; the second threshold is 20% of the reference radius of the spot in the far-field image; and the ellipticity modulation amplitude... The set threshold is 0.1; defocus modulation depth The threshold is set to 5% of the reference radius of the spot in the far-field image.

[0052] This involves four variables: the pixel coordinates of the center of the first virtual circle. The pixel coordinates of the center of the second virtual circle Ellipticity modulation amplitude and defocus modulation depth Theoretically, there are 16 combinations, but cases that do not conform to the laws of physics need to be excluded, leaving only 9 possible cases. After determining that adjustment is needed, if both the connector end and the collimating lens unit need to be adjusted simultaneously, the order should be: connector end first, then collimating lens unit. If all three axes of the collimating lens unit need to be adjusted, the order should be: X-axis and Y-axis first, then Z-axis.

[0053] S5: Based on the content of the decision matrix, adjust the position of the connector end or collimating lens unit, and then use the first locking device and the second locking device to fix the connector end and the collimating lens unit respectively. Start the rotation device again, and re-acquire near-field and far-field images every 10° rotation of the cylinder, and execute steps S3 and S4 again to verify the result of the adjustment of the position of the connector end or collimating lens unit.

[0054] This step is to verify whether the state of the first or second calibration unit after its operation meets the output requirements of the laser. If it is not properly adjusted, further adjustments are required.

[0055] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A wide-temperature-band vibration-resistant, decoupled, adjustable laser, characterized in that, include: A laser light source, used to output laser light, is fixedly installed at one end of the hollow cylinder. The vibration-resistant ceramic pin unit has a fiber optic end and a connector end. The fiber optic end is fused to the laser source and is hinged to the cylinder. The connector end is used to transmit the laser. The first calibration unit is located at the connector end of the anti-vibration ceramic pin unit and is used to adjust the attitude of the connector end. The first calibration unit includes a two-dimensional piezoelectric micro-motion stage and a first locking device; the two-dimensional piezoelectric micro-motion stage is also equipped with an orthogonally arranged strain sensor and a first position sensor; the two-dimensional piezoelectric micro-motion stage is used to adjust the attitude of the connector end, the strain sensor is used to sense vibration or structural deformation, and the first locking device is used to lock the current real-time position of the connector end. The collimating lens unit is located inside the cylinder and is positioned on the light-emitting side of the vibration-resistant ceramic pin unit; The second calibration unit is located between the collimating lens unit and the inside of the cylinder, and is used to adjust and lock the position of the collimating lens unit in the radial or axial direction; the cylinder drives the anti-vibration ceramic pin unit and the collimating lens unit to rotate synchronously relative to the central axis of the cylinder; The second calibration unit includes a three-axis moving device, a second position sensor, and a second position locking device. The three-axis moving device is used to adjust the radial or axial position of the collimating lens unit, the second position sensor is used to acquire the real-time position of the collimating lens unit, and the second position locking device is used to lock the current position of the collimating lens unit. An image analysis unit is located outside the cylinder and on the light-emitting side of the collimating lens unit. It is used to evaluate the acquired image and drive the first calibration unit or the second calibration unit to operate based on the evaluation result. The image analysis unit includes an optical path switching module, which comprises a collimating lens, a dichroic mirror, a focusing lens, a 4f relay lens module, and a camera. The collimating lens, dichroic mirror, focusing lens, and camera are sequentially spaced and coaxially arranged to obtain a far-field optical path for far-field imaging. Similarly, the collimating lens, dichroic mirror, 4f relay lens module, and camera are sequentially spaced and coaxially arranged to obtain a near-field optical path for near-field imaging. Far-field and near-field images are obtained based on the far-field and near-field optical paths, respectively, and the image analysis unit evaluates the far-field and near-field images.

2. The wide-temperature-band vibration-resistant decoupled adjustable laser according to claim 1, characterized in that, The 4f relay lens module includes an achromatic cemented doublet lens and a microscope objective. The achromatic cemented doublet lens and the microscope objective are coaxially arranged, and the microscope objective is located on the light-inlet side or the light-outlet side of the achromatic cemented doublet lens.

3. A compensation method for a wide-temperature-band vibration-resistant, decoupled, adjustable laser, characterized in that, Includes the following steps: S1: Configure a wide-temperature-range vibration-resistant decoupled adjustable laser as described in claim 1 or 2; configure a first calibration unit at the connector end of the vibration-resistant ceramic ferrule unit. The first calibration unit includes a two-dimensional piezoelectric micro-motion stage, a first position sensor, and a first locking device. The two-dimensional piezoelectric micro-motion stage is used to drive the connector end to move along the radial direction of the cylinder. The first position sensor is used to obtain the real-time position of the connector end. The first locking device is used to lock the current position of the connector end. A second calibration unit is configured between the collimating lens unit and the cylinder. The second calibration unit includes a three-axis moving device, a second position sensor, and a second locking device. The three-axis moving device is used to drive the collimating lens unit to move radially or axially relative to the cylinder. The second position sensor is used to obtain the real-time position of the collimating lens unit. The second locking device is used to lock the current position of the collimating lens unit. A temperature sensor is also installed inside the cylinder. A rotation device is installed on the outside of the cylinder. The rotation device is used to drive the cylinder to rotate synchronously with the vibration-damping ceramic pin unit and the collimating lens unit. The connector end has two degrees of freedom in the X and Y directions. The collimating lens unit has three degrees of freedom in the X, Y, and Z directions. The X and Y axes are two orthogonal directions of the radial section of the cylinder. The Z axis is the axial direction of the cylinder and coincides with the virtual optical axis of the laser. S2: Unlock the two-dimensional piezoelectric micro-motion stage, the first locking device, the three-axis moving device and the second locking device, and return the connector end and the collimating lens unit to their initial positions and maintain them; configure the image analysis unit on the light-emitting side of the collimating lens unit, start the rotation device, and use the near-field optical path and the far-field optical path to collect near-field images and far-field images respectively for every 10° rotation of the cylinder; S3: Preprocess the acquired near-field image to obtain the preprocessed near-field image; The acquired far-field image is preprocessed to obtain the preprocessed far-field image; S4: Based on the preprocessed near-field and far-field images, perform near-field trajectory analysis and far-field trajectory analysis respectively; obtain the centroid coordinates of the light spot in each near-field image and fit it to a first virtual circle to obtain the pixel coordinates of the center of the first virtual circle; obtain the centroid coordinates of the light spot in each far-field image and fit it to a second virtual circle to obtain the pixel coordinates of the center of the second virtual circle; then determine the defocus modulation depth according to whether the radius of the light spot in the far-field image changes with the rotation angle of the cylinder; finally, determine the ellipticity modulation amplitude according to whether the ellipticity of the light spot in the far-field image changes with the rotation angle of the cylinder. A decision matrix is ​​constructed by combining the pixel coordinates of the center of the first virtual circle, the pixel coordinates of the center of the second virtual circle, the defocus modulation depth, and the ellipticity modulation amplitude. S5: Based on the content of the decision matrix, adjust the position of the connector end or collimating lens unit, and then use the first locking device and the second locking device to fix the connector end and the collimating lens unit respectively. Start the rotation device again, and re-acquire near-field and far-field images every 10° rotation of the cylinder, and execute steps S3 and S4 again to verify the result of the adjustment of the position of the connector end or collimating lens unit.

4. The compensation method for a wide-temperature-band vibration-resistant decoupled adjustable laser according to claim 3, characterized in that, The preprocessing of the acquired near-field image described in step S3 to obtain the preprocessed near-field image specifically includes the following: constructing a theoretical point spread function containing astigmatism terms; introducing temperature sensor data as a priori astigmatism coefficients in the Richardson-Lucy deconvolution; performing iterative deconvolution of the asymmetric PSF to restore the internal structural details lost due to saturation; eliminating the elliptic distortion of the light spot caused by thermal effects; and obtaining the restored near-field image. Then, a physical model is established for the background image, and the physical model is refined through a neural network. Through multiple forward propagations, the background image is output, and the background image is subtracted from the restored near-field image to restore the true light spot size.

5. The compensation method for a wide-temperature-band vibration-resistant decoupled adjustable laser according to claim 4, characterized in that, The preprocessing of the acquired far-field image described in step S3 to obtain the preprocessed far-field image specifically includes the following: acquiring data from the strain sensor at the connector end, estimating the vibration trajectory, constructing a theoretical point spread function (PSF) based on the vibration trajectory, and obtaining the restored far-field image through a deconvolution algorithm; then, referring to the step of removing the background image from the near-field image, removing the background image from the restored far-field image; then, using energy weighted phase consistency, introducing pixel intensity as a weight into phase calculation, extracting the contour corresponding to the energy threshold, fitting the contour, and extracting the centroid coordinates of the far-field image.

6. The compensation method for a wide-temperature-band vibration-resistant decoupled adjustable laser according to claim 3, characterized in that, The decision matrix includes the following: 1) When the pixel coordinates of the center of the first virtual circle The static eccentricity with respect to the cylinder's rotation axis is not less than the first threshold and the pixel coordinates of the center of the second virtual circle. The static eccentricity with respect to the rotation axis is not less than the second threshold, and the defocus modulation depth and ellipticity modulation amplitude When all values ​​are less than the set threshold, it is determined that the connector end is significantly off-center. The position of the connector end is adjusted only through the first calibration unit, and the current position is locked by the first locking device after adjustment. 2) When the pixel coordinates of the center of the first virtual circle The pixel coordinates of the center of the second virtual circle whose static eccentricity with respect to the cylinder's rotation axis is less than the first threshold. The static eccentricity with respect to the rotation axis is not less than the second threshold, and the defocus modulation depth and ellipticity modulation amplitude When all values ​​are less than the set threshold, it is determined that the connector end is tilted. The position of the connector end is adjusted only through the first calibration unit, and the current position is locked using the first locking device after adjustment. 3) When the pixel coordinates of the center of the first virtual circle The pixel coordinates of the center of the second virtual circle whose static eccentricity with respect to the cylinder's rotation axis is less than the first threshold. The static eccentricity with respect to the rotation axis is less than the second threshold, and the ellipticity modulation amplitude is... Not less than the set threshold, defocus modulation depth When the value is less than the set threshold, it is determined that the collimating lens unit is eccentric. The X-axis and Y-axis positions of the collimating lens unit are adjusted only by the second calibration unit, and the current position is locked by the second locking device after adjustment. 4) When the pixel coordinates of the center of the first virtual circle The pixel coordinates of the center of the second virtual circle whose static eccentricity with respect to the cylinder's rotation axis is less than the first threshold. The static eccentricity with respect to the rotation axis is less than the second threshold, and the ellipticity modulation amplitude is... Defocus modulation depth is less than the set threshold. If the value is not less than the set threshold, it is considered that the collimating lens unit is axially defocused. The Z-axis position of the collimating lens unit is adjusted only by the second calibration unit, and the current position is locked by the second locking device after adjustment. 5) When the pixel coordinates of the center of the first virtual circle The static eccentricity with respect to the cylinder's rotation axis is not less than the first threshold and the pixel coordinates of the center of the second virtual circle. The static eccentricity with respect to the rotation axis is not less than the second threshold, and the ellipticity modulation amplitude is... Not less than the set threshold, but the defocus modulation depth When the value is less than the set threshold, it is determined that the connector end is eccentric and the collimating lens unit is eccentric. Then, the first calibration unit is used to adjust the position of the connector end, and the first locking device is used to lock the current position after adjustment. Then, the X-axis and Y-axis positions of the collimating lens unit are adjusted by the second calibration unit, and the second locking device is used to lock the current position after adjustment. 6) When the pixel coordinates of the center of the first virtual circle The pixel coordinates of the center of the second virtual circle whose static eccentricity with respect to the cylinder's rotation axis is less than the first threshold. The static eccentricity with respect to the rotation axis is not less than the second threshold, and the ellipticity modulation amplitude is... Not less than the set threshold, defocus modulation depth When the value is less than the set threshold, it is determined that the connector end is tilted and the collimating lens unit is eccentric. Then, the first calibration unit is used to adjust the position of the connector end, and the first locking device is used to lock the current position after adjustment. Then, the X-axis and Y-axis positions of the collimating lens unit are adjusted by the second calibration unit, and the second locking device is used to lock the current position after adjustment. 7) When the pixel coordinates of the center of the first virtual circle The pixel coordinates of the center of the second virtual circle whose static eccentricity with respect to the cylinder's rotation axis is less than the first threshold. The static eccentricity with respect to the rotation axis is not less than the second threshold, and the ellipticity modulation amplitude is... Defocus modulation depth is less than the set threshold. If the value is not less than the set threshold, and it is determined that the connector end is tilted and the collimating lens unit is defocused axially, the first calibration unit is used to adjust the position of the connector end, and the first locking device is used to lock the current position after adjustment. Then, the Z-axis position of the collimating lens unit is adjusted by the second calibration unit, and the second locking device is used to lock the current position after adjustment. 8) When the pixel coordinates of the center of the first virtual circle The pixel coordinates of the center of the second virtual circle whose static eccentricity with respect to the cylinder's rotation axis is less than the first threshold. The static eccentricity with respect to the rotation axis is less than the second threshold, and the ellipticity modulation amplitude is... Not less than the set threshold, defocus modulation depth If the value is not less than the set threshold, it is determined that the collimating lens unit is off-center and out of focus. The X-axis and Y-axis positions of the collimating lens unit are first adjusted by the second calibration unit, and then the Z-axis position is adjusted. After adjustment, the current position is locked by the second locking device. 9) When the pixel coordinates of the center of the first virtual circle The static eccentricity with respect to the cylinder's rotation axis is not less than the first threshold and the pixel coordinates of the center of the second virtual circle. The static eccentricity with respect to the rotation axis is not less than the second threshold, and the ellipticity modulation amplitude is... Not less than the set threshold, defocus modulation depth If the value is not less than the set threshold, it is determined that the connector end and the collimating lens unit are completely misaligned. Then, the first calibration unit is used to adjust the position of the connector end, and after adjustment, the first locking device is used to lock the current position. The second calibration unit is used to adjust the X-axis and Y-axis positions of the collimating lens unit first, and then adjust the Z-axis position. After adjustment, the second locking device is used to lock the current position.

7. The compensation method for a wide-temperature-band vibration-resistant decoupled adjustable laser according to claim 6, characterized in that, The first threshold is 10% of the spot diameter in the near-field image; the second threshold is 20% of the reference radius of the spot in the far-field image; ellipticity modulation amplitude. The set threshold is 0.1; defocus modulation depth The threshold is set to 5% of the reference radius of the spot in the far-field image.

Images