CN112792450A

CN112792450A - Optical fiber automatic focusing method and automatic focusing system for laser processing

Info

Publication number: CN112792450A
Application number: CN202110135634.8A
Authority: CN
Inventors: 何俊; 王义平; 肖珣周; 杜斌
Original assignee: Shenzhen University
Current assignee: Shenzhen University
Priority date: 2021-02-01
Filing date: 2021-02-01
Publication date: 2021-05-14
Anticipated expiration: 2041-02-01
Also published as: WO2022160452A1; CN112792450B

Abstract

The invention provides an optical fiber automatic focusing method and an automatic focusing system for laser processing, which perform automatic focusing and image acquisition through the following steps: acquiring an image of an optical fiber to be processed; processing the image to obtain an image edge; obtaining a core edge: carrying out non-maximum suppression processing and discrete edge connection processing on the image edge to obtain the upper edge and the lower edge of a fine fiber core; and (3) calculating an offset: calculating offset according to the upper edge and the lower edge of the fiber core; offset compensation: and compensating according to the information of the offset so as to focus the laser on the target position on the optical fiber. The optical fiber automatic focusing method and the automatic focusing system for laser processing provided by the invention can realize optical fiber automatic focusing and enable laser to be focused on a target position on an optical fiber. High positioning precision and high focusing speed, and can process different optical fibers.

Description

Optical fiber automatic focusing method and automatic focusing system for laser processing

Technical Field

The present invention relates to the field of laser processing, and in particular, to an optical fiber autofocus method and system for laser processing.

Background

The fiber grating or the microcavity can be prepared by processing the optical fiber through laser. The fiber grating is a diffraction grating formed by axially and periodically modulating the refractive index of a fiber core, and is a passive filter device. The fiber grating has the advantages of large reflection bandwidth range, small additional loss, small volume, easy coupling with optical fiber, compatibility with other optical devices, no influence of environmental dust and a series of excellent performances, and the resonance wavelength of the fiber grating is sensitive to the change of external environments such as temperature, strain, refractive index, concentration and the like, so the fiber grating is widely applied to the fields of manufacturing fiber lasers, fiber communication and sensing, and is an important device in the fields of fiber communication and sensing.

The fiber Bragg grating is manufactured by a method of writing by laser such as femtosecond laser, and the fiber Bragg grating can be written on all transparent optical fibers with various coatings, high-end special optical fibers and low-cost commercial optical fibers. Compared with the traditional method for manufacturing the fiber Bragg grating by the phase mask plate method, the fiber Bragg grating manufactured by femtosecond laser writing can work under the extreme temperature condition, the working temperature exceeds 1000 ℃, and the fiber Bragg grating has the characteristics of super-strong tensile strength, corrosion resistance, high-temperature and high-humidity environment resistance and the like.

In the existing method for preparing the fiber bragg grating by using laser, a section of optical fiber to be processed is manually arranged on a moving mechanism, and then six axial directions of the moving mechanism are manually adjusted, so that laser can be focused at the center of a fiber core. However, the existing manual focusing method not only has complicated preliminary preparation and leveling work, but also has poor focusing precision, is not beneficial to the preparation of large-scale fiber grating arrays, and prevents the use of fiber gratings or microcavity sensors in the field of distributed fiber sensing.

In addition, after one section of optical fiber is processed, the next section of optical fiber is moved to a processing area, and then the whole leveling work is repeated to prepare the next section of optical fiber; in addition, due to the performance of the motion mechanism and the vibration isolation capability of the optical platform, the optical fiber is difficult to keep absolutely parallel in the process of translation, slightly jittered laser light deviates from the center of the fiber core, loss is increased, and consistency is reduced.

Disclosure of Invention

In order to overcome the defects of the prior art, the invention provides the optical fiber automatic focusing method and the automatic focusing system for laser processing, which have the advantages of high positioning precision, high focusing speed, low hardware requirement, convenience in integration and the like, improve the preparation precision and efficiency, and can process different optical fibers. The method is particularly suitable for preparing large-scale fiber bragg grating or microcavity sensor arrays.

The invention aims to provide an optical fiber automatic focusing method and system for laser processing.

The application provides an optical fiber automatic focusing method for laser processing, which comprises the following steps:

s1, image acquisition: acquiring an image of an optical fiber to be processed;

s2, processing the image to obtain the image edge;

s3, positioning the fiber core: carrying out non-maximum suppression processing and discrete edge connection processing on the image edge to obtain the upper edge and the lower edge of a fine fiber core;

s4, offset calculation: calculating offset according to the upper edge and the lower edge of the fiber core;

s5, offset compensation: and compensating according to the information of the offset so as to focus the laser on the target position on the optical fiber.

As an improvement of the optical fiber auto-focusing method provided by the present invention, in step S3, the method of the non-maximum suppression process includes S31:

firstly, screening out points larger than a threshold value;

then, the screened points are judged one by one, whether the gradient is a local maximum value in the horizontal or vertical direction is judged, and if so, the gradient is retained; if not, the next point is continuously judged until the maximum value is found.

As an improvement of the optical fiber auto-focusing method provided by the present invention, in step S3, the discrete edge connection method includes S32:

and detecting the amplitude and angle difference between each point and the adjacent point retained after the non-maximum suppression treatment one by one, and determining that the points are positioned on the same edge and connecting the points within a certain range.

As an improvement of the optical fiber auto-focusing method provided by the present invention, in step S3, the method further includes the following steps:

summing the horizontal pixel gray scales of the image containing the edge extracted in the step S32 and obtaining a gray scale average value;

the upper and lower edges of the fiber core of the optical fiber are positioned in a symmetrically distributed relationship according to the gray scale in the radial direction of the optical fiber.

As an improvement of the optical fiber automatic focusing method provided by the present invention, in step S2, the method for obtaining the image edge for the image processing includes the following steps:

s21, converting the acquired image into a gray-scale image, and performing gray-scale equalization processing on the gray-scale image;

s22, filtering the image after gray level equalization processing through convolution operation to remove noise of a specific frequency band;

and S23, after the filtering is finished, obtaining a preliminary image edge through difference processing and gradient solving processing.

As an improvement of the optical fiber auto-focusing method provided by the present invention, in step S22, the method of filtering by convolution operation includes: constructing a kernel function of an image filter, and performing convolution operation on the kernel function and the two gray level images respectively to remove noise of a specific frequency band;

in step S23, the method of difference processing is: performing subtraction operation on the two images obtained through convolution operation pixel by pixel to extract difference information of the two images to obtain a two-dimensional matrix;

in step S23, the gradient solving process is performed by: and solving the gradient direction of the two-dimensional matrix obtained after the difference processing, and solving the gradient strength of the two-dimensional matrix obtained after the difference processing.

As an improvement of the optical fiber auto-focusing method provided by the present invention, in step S21, the acquired image is converted into a gray-scale image to obtain a gray-scale histogram, and the mean value and standard deviation of the histogram are changed by gray-scale equalization processing.

As an improvement of the optical fiber automatic focusing method provided by the invention, the illumination when the image is acquired is evaluated according to the gray level histogram of the image, and the illumination compensation is carried out according to the evaluation result.

As an improvement of the optical fiber auto-focusing method provided by the present invention, in step S4:

according to the upper and lower edges of the fiber core extracted in the step S3, converting and calculating the position offset and the focus offset under a pixel coordinate system through function fitting;

and converting the position offset and the focus offset in the pixel coordinate system into the position offset and the focus offset in the world coordinate system.

As an improvement of the optical fiber auto-focusing method provided by the present invention, in step S5, the offset compensation includes:

before processing, adjusting the position of the optical fiber relative to a laser focusing point according to the offset;

and in the processing process, the position of the optical fiber relative to the laser focusing point is adjusted according to the offset.

As an improvement of the automatic focusing method of the optical fiber provided by the invention, the offset compensation in the processing process comprises the following steps:

compensation for short range motion and compensation for long range motion;

the compensation for the short-range motion is used for compensating when the position of the optical fiber is moved in the processing process of the optical fiber; the compensation of the long-range motion is used for compensating when a next section of optical fiber is moved to a processing area after one section of optical fiber is processed;

the compensation speed for the short-range motion is slow but the precision is high, and the compensation speed for the long-range motion is fast but the precision is low.

As an improvement of the optical fiber automatic focusing method provided by the invention, when compensating for short-range motion,

sampling the motion path of the short-range motion in a segmented manner, calculating the offset of each segmented point, and calculating the offset in the whole process through interpolation; and finally, correcting the motion trail according to the offset in the whole process and planning a new motion trail.

The present application also provides an optical fiber auto-focusing system for laser processing, which performs focusing by the optical fiber auto-focusing method of the above claims.

The invention has the following beneficial effects:

the automatic focusing method and the automatic focusing system for the optical fiber for laser processing provided by the invention can realize the automatic focusing of the laser on the optical fiber and focus the laser on the target position on the optical fiber. High positioning precision and high focusing speed, and can process different optical fibers.

Drawings

FIG. 1 is a schematic diagram of an optical fiber auto-focusing method for laser processing according to an embodiment of the present invention;

fig. 2 is a schematic structural diagram of a laser processing system according to an embodiment of the present invention;

FIG. 3 is a schematic structural diagram of an optical fiber auto-focusing system according to an embodiment of the present invention;

fig. 4 is a schematic diagram of an image processing module according to an embodiment of the present invention;

FIG. 5 is a schematic illustration of an image recognition algorithm locating the fiber core;

fig. 6 is a schematic diagram illustrating a working procedure of a path planning module of an optical fiber auto-focusing system according to an embodiment of the present invention;

fig. 7 is a schematic structural diagram of a motion control module according to an embodiment of the present invention;

fig. 8 is a schematic structural diagram of a motion control module according to an embodiment of the present invention;

fig. 9 is a schematic structural diagram of a motion control module according to an embodiment of the present invention.

The attached drawings are marked as follows:

the system comprises an image acquisition module 10, an image processing module 20, a motion control module 30, a fiber clamping module 40, an illumination control module 50, a gray scale equalization module 21, a filtering module 22, an edge extraction module 23, an offset calculation module 24, a short-range motion control module 32, a long-range motion control module 31, a position correction module 33, a rotary motion control module 34, a laser 101, an electronic control shutter 102, a reflector 103, an objective lens 104, a CCD camera 105, a light source 106, a fiber clamp 107, a displacement platform 108, a first controller 109, a fiber spool 110, a second controller 111 and a computer 112.

Detailed Description

The present invention will be described in detail below with reference to the accompanying drawings and examples.

The present application provides a laser processing system for laser processing an optical fiber. Comprises a laser optical path system and an optical fiber automatic focusing system.

As shown in fig. 2, the laser optical path system includes a laser 101, a mirror 103, a diaphragm, an electrically controlled shutter 102, and an objective lens 104. The laser beam provided by the laser 101 is used for fiber micromachining; the laser beam passes through the reflector 103 and then is focused by the objective lens, so that the focusing point of the laser beam is positioned at the position to be processed on the optical fiber.

As shown in fig. 3, the optical fiber auto-focusing system includes:

the image acquisition module 10 is used for acquiring an image of an optical fiber to be processed;

the image processing module 20 is used for processing the image and calculating the offset according to the upper edge and the lower edge of the fiber core;

the optical fiber clamping module 40 is used for fixing the optical fiber and driving the optical fiber to move;

and the motion control module 30 drives the optical fiber to move according to the information of the offset.

The fiber holding module 40 is used to hold the fiber during processing and movement. The fiber clamping module 40 includes a fiber clamp 107.

The specific embodiment of the application provides an optical fiber automatic focusing method for laser processing, which comprises the following steps:

s1, image acquisition: acquiring an image of an optical fiber to be processed;

s2, processing the image to obtain the image edge;

s3, core positioning, obtaining core edge: carrying out non-maximum suppression processing and discrete edge connection processing on the image edge to obtain the upper edge and the lower edge of a fine fiber core;

The target position at which the laser is focused on the optical fiber may be a position in the center of the core, a position on the core that is offset from the center of the core by a specific distance, or a target position on the cladding on the optical fiber.

The description of step S1 is as follows:

in step S1, an image of the optical fiber to be processed is acquired by the image acquisition module 10, and the acquired image is subjected to subsequent image processing and further processing. The image acquisition module 10 includes an image sensor, and the image acquisition module 10 is, for example, a CCD camera 105.

The image acquisition step includes:

converting the transmitted analog image signal into a digital signal through an analog-to-digital converter A/D, transmitting the digital signal to an image processor, processing the digital signal to obtain an image, and finally completing the conversion from a world coordinate system to an image coordinate system;

according to the digital signal obtained by the conversion of the analog-to-digital converter, parameters such as exposure time, image brightness, contrast, saturation, resolution and the like obtained by processing are adjusted.

Steps S2-S4 are all performed by the image processor.

The image processor includes: a gray level equalization module 21, a filtering module 22, an edge extraction module 23, and an offset calculation module 24, as shown in fig. 4. The offset calculation module 24 also includes a coordinate conversion module.

The description of step S2 is as follows:

in step S2, the method for obtaining the image edge for image processing includes the steps of:

s21, converting the acquired digital image into a gray image through the gray balance module 21, and carrying out gray balance processing on the gray image;

s22, filtering the image after gray level equalization processing through convolution operation by the filtering module 22 to remove noise of a specific frequency band;

The image processor includes a DSP microprocessor/FPGA integrated circuit.

The DSP microprocessor/FPGA integrated circuit comprises the gray scale equalization module 21 and the filtering module 22, the step S21 and the step S22 are realized through the DSP microprocessor/FPGA integrated circuit, and the high-frequency signals are amplified after low-frequency clutter signals are filtered.

The DSP microprocessor/FPGA integrated circuit further includes the edge extraction module 23, and the step S23 is implemented by the edge extraction module 23 of the DSP microprocessor/FPGA integrated circuit.

In step S21, firstly, the gray level equalization module 21 performs gray level processing on the digital image acquired in step S1, completes the conversion of the color image into the gray level image through the color space conversion, converts the acquired image into the gray level image, and obtains the gray level histogram of the image;

formula (1)

Equation (1) represents the calculation process for converting a color image into a grayscale image, where R, G and B represent the three channels of the color image, respectively.

Next, the gradation histogram of the image is operated, and the mean value and the standard deviation of the histogram are changed by the gradation equalization processing.

Formula (2)

Wherein

Represents an input image,

Representing the output image, m is the mean of the input image gray levels, and s is the standard deviation of the gray levels. The average luminance value of the image can be removed by averaging the input image and dividing by the standard deviation, and the overall brightness of the image does not affect what objects are present in the image. It makes sense to remove the mean of the pixels for each data point at this time.

The result of the gradation equalization process will proceed to step S22 for further processing.

In step S22, the filter module 22 is used to perform convolution operation to remove noise in a specific frequency band. The filtering method comprises the following steps: and constructing a kernel function of the image filter, and performing convolution operation on the kernel function and the two gray level images respectively.

Specifically, a filter kernel function is constructed first, and the filter kernel function can be expressed as formula (3):

formula (3)

Formula (3) shows a two-dimensional gaussian filter function, which is the product of one-dimensional gaussian functions in x and y directions.

The type and order of the kernel function is controlled by the gray scale equalization module 21. The idea of filtering is as follows: discretizing the function kernel function, taking the function value on the discrete point as a weight, and performing weighted average in a certain range of neighborhood on each pixel point of the image matrix to effectively eliminate noise.

The convolution of the two images with the kernel function can be expressed as formula (4) and formula (5):

formula (4)

Formula (5)

The formulas (4) and (5) represent the image

Respectively with two filtering kernels

And

and performing convolution operation, wherein the convolution operation represents the result of summation after the two functions are multiplied in a certain range.

The convolution operation of step S22 may be implemented by a convolution operator, an adder, or the like.

The edge extraction module 23 further includes a difference processing module and a gradient processing module.

In step S23, the method of performing the difference processing by the difference processing module is: and performing subtraction operation on the two images obtained through convolution operation pixel by pixel to extract difference information of the two images, so as to obtain a two-dimensional matrix. The difference processing is performed by the following formula (6):

formula (6)

Since the convolution operation has a combination law, performing subtraction on two images obtained by the two convolution operations pixel by pixel is equivalent to performing subtraction on two filter kernels and then performing convolution operation on the two filter kernels and the image, so that the computational complexity can be greatly reduced, as shown in formula (6).

The two images are subtracted pixel by pixel to extract spatial information contained in a frequency band common to the two images. The difference processing here thus corresponds to a band-pass filter that removes all frequency information except those that remain in the original image.

In step S23, the gradient solving process by the gradient processing module is performed by: and (3) solving the gradient direction of the two-dimensional matrix obtained after the difference processing by using a formula (7), and solving the gradient strength of the two-dimensional matrix obtained after the difference processing by using a formula (8) so as to strengthen the edge information.

The solving formula of the gradient direction is as follows:

formula (7)

Wherein G is_xAnd G_yFirst derivative values of the image in the horizontal and vertical directions respectively;

it is indicated that the direction of the gradient,

is an arctangent function. The solving formula of the gradient strength is as follows:

formula (8)

Wherein

And

the first derivative values in the horizontal direction and the vertical direction are respectively, and the gradient strength can be obtained by solving the gradient mode.

Illumination compensation:

it should be added that after step S21, the illumination at the time of image acquisition may be evaluated according to the gray histogram of the image, and illumination compensation may be performed according to the evaluation result, so as to acquire an image with better quality.

The illumination compensation mainly comprises: any one of the light emission intensity, color, light emission surface width, light emission surface length, and the distance between the light source and the optical fiber of the illumination light source is adjusted.

As shown in fig. 3, the optical fiber automatic focusing system for laser processing according to the embodiment of the present application further includes a light source 106, where the light source 106 is configured to provide illumination when the image sensor acquires an image of an optical fiber to be processed.

The fiber optic autofocus system also includes an illumination control module 50. The illumination control module 50 includes: the illumination evaluation module and the illumination light source control module.

The illumination evaluation module may determine the imaging quality of the image obtained under the current illumination condition according to the gray histogram transmitted by the gray balancing module 21, determine whether the current illumination effect meets the condition of normal work of image processing in step S2, obtain a quantifiable index, and output the evaluation result (processing result) to the illumination light source control module.

The illumination source control module is used for adjusting relevant parameters of the illumination source according to the evaluation index calculated by the illumination evaluation module so as to compensate illumination.

The description of step S3 is as follows:

in step S3, the non-maximum suppression processing and the connection of the discrete edges are performed by the edge extraction module 23 to obtain a fine edge.

The non-maximum suppression processing is performed by the edge extraction module 23, and the specific method includes step S31:

firstly, screening out points larger than a threshold value;

In step S3, discrete edge connection is performed by the edge extraction module 23, and the specific method includes step S32:

and detecting the amplitude and angle difference between each point and the adjacent point after the non-maximum value inhibition processing one by one, and determining that the points are positioned at the same edge and connecting the points within a certain range to obtain an image containing the extracted edge.

For further optimization, step S3 further includes step S33: finally, the image including the extracted edge is summed in the transverse direction, i.e. the transverse pixel gray scale sum, and the average value of the matrix in the transverse direction can be obtained through step S33. The gray level average value is obtained through transverse summation, then optical fiber positioning is carried out, and the average value of the optical fiber position in the image is obtained, so that the robustness is good, and noise interference is not easy to occur. If only the gray values on one line are extracted, it is easily disturbed by noise or foreign matter (e.g., dust and air bubbles in the immersion oil) to cause positioning errors.

Step S34 is further included next in step S3: the position of the optical fiber is located according to the existence of four gray scale maximum points (peaks) and two gray scale minimum points (valleys) in the radial direction of the optical fiber, as shown in fig. 5; and after transverse summation, the optical fiber positioning is equivalent to taking the average value of the optical fiber positions in the image, so that the robustness is good, and the noise interference is not easy to occur.

And finally, the upper and lower edges of the fiber core of the optical fiber are positioned according to the relationship that the gray scales in the radial direction of the optical fiber are symmetrically distributed.

The description of step S4 is as follows:

in step S4, the offset amount calculation method specifically includes:

through the offset calculation module 24, according to the upper and lower edges of the fiber core extracted in step S3, the position offset and the focus offset in the pixel coordinate system are converted through function fitting;

and then, converting the position offset and the focus offset in the pixel coordinate system into the position offset and the focus offset in the world coordinate system through a coordinate conversion module.

It is understood that, in step S4, the central position of the core is calculated according to the upper and lower edges of the core obtained in step S3, and then the offset is calculated according to the relationship between the central position of the core and the target position on the optical fiber on which the laser needs to be focused. Alternatively, the offset may be directly calculated according to the target position at which the laser needs to be focused and the upper and lower edges of the core obtained in step S3.

The offset calculation module 24 divides the extracted gray scale maximum value point (peak) and gray scale minimum value point (valley)The z-axis and the z-axis of the pixel coordinate system are calculated through a fitting function,

Shaft and

the offset of the axis pixel. The fitting function is obtained by acquiring images under different deviation conditions and recording the distribution relation fitting of the gray scale maximum value points (wave crests) and the gray scale minimum value points (wave troughs).

Finally, the coordinate conversion module converts the z-axis and the z-axis of the pixel coordinate system,

Shaft and

the offset of the axis pixel is respectively converted into the offset of the optical fiber focusing in the preparation process and the offset of the up-and-down fluctuation in the world coordinate system, namely the offset of the vertical direction and the graphic plane direction and the offset of the vertical direction and the graphic plane direction parallel to the graphic plane direction in the graph 5. The conversion relation from the pixel coordinate system to the world coordinate system is obtained by calibrating the camera by utilizing the etalon.

The description of step S5 is as follows:

in step S5, the motion control module 30 compensates the position of the optical fiber and the laser focusing position according to the calculated optical fiber offset in the world coordinate system, thereby ensuring that the target position on the optical fiber can be accurately aligned every processing.

The motion control module 30 includes: a short-range motion control module 32, a long-range motion control module 31, and a position correction module 33, as shown in fig. 7.

As shown in fig. 2. The short-range motion control module 32 includes a displacement stage 108, and a first controller 109 for controlling the displacement stage, and the displacement stage 108 is controlled by the first controller 109 to move the optical fiber in a translational manner. The optical fiber is fixed on the displacement platform 108 through the optical fiber clamp 107, and the displacement platform 108 drives the optical fiber clamp 107 and the optical fiber to move together when moving. The displacement platform 108 is a precision three-dimensional displacement platform, and the short-range motion speed is slow but the precision is high.

The long-range motion control module 31 includes an optical fiber spool 110, a driving motor, and a second controller 111, and moves the optical fiber by dragging (winding) the optical fiber spool 110, and the long-range motion control module 31 can drive the optical fiber to move at a faster speed, but with a lower precision than the short-range motion.

The first controller 109 and the second controller 111 are both connected to the computer 112. The computer 112 also simultaneously connects and controls the laser 101 and the CCD camera 105.

In step S5, the offset compensation method includes:

before processing, the position of the optical fiber relative to the laser focusing point is adjusted according to the offset, so that the laser focusing position is aligned with the target position on the optical fiber.

The offset compensation during the machining process includes compensation for short range motion and compensation for long range motion.

The compensation of the short-range motion is performed by the short-range motion control module 32 when the position of the optical fiber is moved in the processing process of the optical fiber, so that the processing position is always located at the target position on the optical fiber.

The compensation of the long-range motion is performed by the long-range motion control module 31 in the process of moving the next section of optical fiber to the processing area after the processing of the section of optical fiber is finished.

In contrast, the compensation speed for short-range motion is slow but the accuracy is high, and the compensation speed for long-range motion is fast but the accuracy is low. Until the preparation of the entire array is complete.

According to the application, the optical fiber is moved by the short-range motion control module 32 in the processing process, so that the speed is low, but the precision is high; and when the optical fiber moves to the area to be processed of the next section of optical fiber after the processing is finished, the optical fiber moves through the long-range motion control module 31, so that the speed is high, but the precision is low. Therefore, the processing speed is ensured, and the processing precision is also ensured.

And a position correction module 33, configured to perform a compensation on the optical fiber processing process according to the result (offset in the spatial coordinate system) calculated by the image processor during the whole motion process. In the existing preparation method, a position correction module is not used, but in the actual processing process, the laser position may deviate from the target position on the optical fiber, such as a fiber core, before the preparation starts, and if the position correction is not carried out, the finally processed structural performance is reduced, and even the final processed structural performance cannot be processed into the target position on the optical fiber; since the offset is accumulated, the array cannot be processed without introducing the position correction.

The offset compensation method has the advantages that the processing precision is higher: before each processing, the offset of the optical fiber relative to the laser focusing point can be automatically corrected according to an image recognition algorithm, and the positioning precision of the image recognition algorithm reaches the submicron level. And simultaneously, in the processing process, the position of the optical fiber relative to the laser focus point is adjusted according to the offset. The prior art rarely corrects the position during processing.

The position correction module also includes a path planning module 35, as shown in fig. 9. The path planning module 35 is specifically configured to perform segmented sampling on the motion path given by the short-range motion control module.

Setting an initial point and an end point during compensation of short-range motion, sampling along a motion track in a segmented manner, and calculating the offset of each segmented point to obtain the offset of a sampling position; calculating the offset of each section by piecewise interpolation fitting; and finally, correcting the motion trail according to the offset in the whole process and planning a new motion trail. After the subsequent processing is started, the position correction module compensates the position of the optical fiber according to the new motion trajectory, compensates the offset of each segment, aligns the target position (for example, the fiber core) on the optical fiber to the region to be processed in the whole motion trajectory, and finally completes the processing, as shown in fig. 6. The introduction of the path planning module in this scenario can be used to fabricate longer fiber microstructures as well as waveguide structures.

In the existing preparation method, segmented sampling and subsequent offset compensation are not used, so that the possibility that the laser position deviates from the target position on the optical fiber exists in the preparation process, and the processed structural performance is reduced; meanwhile, the consistency of each structure is difficult to guarantee in the array processing process, and even the automatic preparation of the array cannot be carried out.

Further, the motion control module 30 also includes a rotational motion control module 34, as shown in FIG. 8. The rotational motion control module 34 is specifically configured to rotate the optical fiber to align target positions, such as cores, on the optical fiber at different positions of the multicore fiber to the region to be processed. In this scenario, the rotational motion control module 34 may be configured to control the rotation of the optical fiber clamp, for example, the optical fiber clamp may be driven by a motor to rotate; the laser beam to be processed may be controlled to rotate. The rotary control module is introduced, so that the problem that the non-processed fiber core can cause loss to the energy of the passing laser when the laser focusing position is changed to process in different fiber cores of the multi-core fiber can be avoided.

The invention has the following beneficial effects:

the invention provides an optical fiber automatic focusing method and system for laser processing, which have the following advantages compared with the existing optical fiber grating array processing system:

the processing precision is high: before each processing, the offset between the optical fiber position and the laser focus position can be automatically corrected according to an image recognition algorithm, the positioning precision of the image recognition algorithm reaches a submicron level, and the position after processing is rarely corrected in the prior art.

The processing speed is high: the optical fiber is moved by the short-range motion control module in the processing process, so that the speed is low, but the precision is high; and when the optical fiber moves to the area to be processed of the next section of optical fiber after the processing is finished, the optical fiber moves through the long-range motion control module, so that the speed is high, but the precision is low.

The consistency among all microstructures is good: the same focal length and position can be ensured in each processing, so that the consistency among all the microstructures is improved, and the automatic preparation of the array becomes possible.

The insertion loss is low: the entire process ensures that the target location on the fiber is accurately processed without damaging the cladding, thereby reducing insertion loss.

The operation is convenient: the system realizes automatic positioning and automatic focusing in the processing process, and the position and the level of the optical fiber are not required to be manually adjusted before each processing.

Different types of optical fibers can be processed: the optical fiber may be a quartz fiber, a plastic fiber, a photonic crystal fiber, a multicore fiber, or other kind of optical fiber.

It is to be understood that the above-described embodiments are only some of the embodiments of the present application, and not all embodiments of the present application. This application is capable of embodiments in many different forms and is provided for the purpose of enabling a thorough understanding of the disclosure of the application. Although the present application has been described in detail with reference to the foregoing embodiments, it will be apparent to one skilled in the art that the present application may be practiced without modification or with equivalents of some of the features described in the foregoing embodiments. All equivalent structures made by using the contents of the specification and the drawings of the present application are directly or indirectly applied to other related technical fields and are within the protection scope of the present application.

Claims

1. A method of optical fiber auto-focusing for laser machining, comprising the steps of:

s1, image acquisition: acquiring an image of an optical fiber to be processed;

s2, processing the image to obtain the image edge;

2. The optical fiber auto-focusing method according to claim 1, wherein in the step S3, the method of non-maximum suppression processing includes S31:

firstly, screening out points larger than a threshold value;

3. The optical fiber auto-focusing method according to claim 2, wherein in the step S3, the discrete edge connecting method includes S32:

4. The optical fiber auto-focusing method according to claim 3, further comprising, in step S3, the steps of:

5. The optical fiber automatic focusing method according to claim 1, wherein in step S2, the method for processing the image to obtain the image edge comprises the following steps:

6. The optical fiber auto-focusing method according to claim 5,

in step S22, the method of filtering by convolution operation includes: constructing a kernel function of an image filter, and performing convolution operation on the kernel function and the two gray level images respectively to remove noise of a specific frequency band;

7. The optical fiber auto-focusing method according to claim 5,

in step S21, the acquired image is converted into a grayscale map to obtain a grayscale histogram, and the mean value and standard deviation of the histogram are changed by grayscale equalization processing.

8. The optical fiber auto-focusing method according to claim 7, wherein illumination at the time of image acquisition is evaluated based on a gray histogram of the image, and illumination compensation is performed based on the evaluation result.

9. The optical fiber auto-focusing method according to claim 1, wherein in step S4:

10. The optical fiber auto-focusing method according to claim 1, wherein the offset compensation comprises, in step S5:

11. The method of claim 10, wherein the offset compensation during processing comprises:

compensation for short range motion and compensation for long range motion;

12. The optical fiber auto-focusing method according to claim 11, wherein, in the compensation for the short-range motion,

13. An optical fiber autofocus system for laser processing, wherein focusing is performed by the optical fiber autofocus method according to any one of claims 1 to 12.