Optical fiber end face positioning method and its application

Abstract

The present invention discloses a method for locating an optical fiber end face and an application thereof. The optical fiber end face locating method provided by the present invention not only cleverly reduces the difficulty of image recognition by acquiring an optical fiber end face image and coordinating the transformation and selection of a lighting method, but also can realize efficient and accurate docking of polarization-maintaining optical fiber end faces, thereby reducing the fusion loss of the optical fiber and maintaining a stable extinction ratio. The optical fiber end face locating method provided by the present invention is not only suitable for the end face alignment and fusion splicing of different polarization-maintaining optical fibers, but can also be applied to the production of optical fiber quarter-wave plates.

Description

Technical Field

[0001] The invention relates to the technical field of optical fiber processing, and in particular to an optical fiber end face positioning method and application thereof. Background Art

[0002] As a special type of optical fiber, polarization-maintaining optical fiber can transmit linearly polarized light, ensure that the linear polarization direction remains unchanged, and improve the coherent signal-to-noise ratio. It is widely used in various fields of the national economy, such as aerospace, aviation, navigation, industrial manufacturing technology, and communications. Compared with the structure of ordinary optical fibers, the cladding of polarization-maintaining optical fiber contains not only the core, but also a stress zone symmetrical about the core. According to the shape of the stress zone, polarization-maintaining optical fiber is generally divided into panda type, bow tie type, and elliptical type. On the one hand, for the fusion splicing of ordinary optical fibers, it is only necessary to align the core part, while for polarization-maintaining optical fiber, in addition to aligning the core, it is also necessary to align the stress zone part, which further increases the difficulty of fusion splicing. On the other hand, in the field of optical fiber quarter-wave plate preparation, it is a very critical step to 90° fuse the polarization-maintaining optical fiber pigtail that outputs linearly polarized light with the input end of the polarization-maintaining optical fiber, and how to ensure that the optical fibers on both sides are 90° aligned has become a difficult problem. At present, there is little research on the end-face docking technology of polarization-maintaining optical fiber in China, and most of the polarization-maintaining optical fiber fusion splicers on the market only include the core alignment function. If the end-face alignment of the polarization-maintaining optical fiber and the quarter-wave plate preparation method can be combined to ensure the normal completion of the core-alignment fusion splicing function of the optical fiber, it can also support the quarter-wave plate preparation to a certain extent, which is of great significance to the manufacturing of precision instruments in my country.

[0003] The patent with publication number CN105676356B discloses a method for positioning a fiber core and a method for calibrating the core alignment of fiber fusion splicing. The fiber outline is identified by collecting a side view image of the fiber to be tested, and the grayscale image is obtained and analyzed to determine the core center positions of the two optical fibers. In this way, the relative positions of the two optical fibers are adjusted and the core positions are calibrated. However, the positioning and calibration method disclosed in the invention can only identify the core, but cannot identify the complete fiber end face. Therefore, it is not suitable for the docking of polarization-maintaining optical fibers, and the positioning and calibration method is also not suitable for the field of quarter-wave plate preparation. Summary of the invention

[0004] Those skilled in the art are well aware that the smaller the splicing loss of the polarization-maintaining optical fiber after fusion splicing and the larger the extinction ratio, the better the fusion splicing effect. The key factor affecting the fusion splicing performance lies in the accuracy of the core, especially the accuracy of the end face connection of the polarization-maintaining optical fiber. Once the end face alignment effect is not good, it will lead to poor fusion splicing effect, and even the optical fiber after fusion splicing cannot be used. In order to overcome the deficiencies of the above-mentioned prior art, the present invention provides a method for locating an optical fiber end face, which obtains an optimal end face grayscale image by lighting with a supplementary light source on the basis of natural lighting, i.e., correcting the light, thereby accurately determining the center position of a target area of ​​the optical fiber end face. For ordinary optical fibers, the target area is the core, and other positions are background areas; for polarization-maintaining optical fibers, the target area includes the core and the center of the stress zone, and the corresponding other positions are background areas; after obtaining the center position coordinates of the corresponding target area, it can be used to accurately adjust the position of the optical fiber end face according to different purposes.

[0005] In a first aspect, an embodiment of the present invention provides a method for locating an optical fiber end face, comprising:

[0006] Set different lighting conditions and obtain initial images of the optical fiber end face under different lighting conditions; perform color gamut conversion on the initial images of the optical fiber end face to obtain corresponding optical fiber end face grayscale images; obtain the grayscale value difference between the target area and the background area in each optical fiber end face grayscale image, and compare them. The optical fiber end face grayscale image with the largest grayscale value difference is the optimal optical fiber end face grayscale image;

[0007] The best optical fiber end face grayscale image is subjected to image salt and pepper noise filtering by using median filtering to obtain a higher quality optical fiber end face grayscale image marked as a high-definition grayscale image;

[0008] The maximum inter-class variance algorithm is used to iterate the high-definition grayscale image to achieve adaptive binarization, so as to obtain the center position coordinates of the target area in the image, which are the core center position coordinates and the stress area center position coordinates in the polarization-maintaining optical fiber.

[0009] It should be noted that the different lighting conditions described in the present invention include: the color temperature of the supplementary light source, the power of the supplementary light source, the lighting angle of the supplementary light source and the bending degree of the optical fiber, but are not limited to the above-mentioned lighting conditions. Any changes in external unnatural lighting conditions that affect the grayscale value difference between the target area and the background area in the obtained optical fiber end face grayscale image belong to the lighting conditions described in the present invention, wherein the range of the color temperature of the light source includes but is not limited to white light, warm white and red light; the power selection range of the light source is preferably within 0.5-1.5W. If the power is too high or too low, the obtained optical fiber image will lose details, which is not conducive to obtaining a larger grayscale value difference; such as Figure 4As shown in the figure, the lighting angle of the supplementary light source refers to the horizontal plane where the optical fiber is located as the reference, and the rear of the optical fiber end face is taken as the initial angle of 0°. During the debugging process, the angle is increased in a clockwise or counterclockwise direction on the horizontal plane where the optical fiber is located until it increases to 180°; the bending degree of the optical fiber refers to the angle of the unfixed part of the optical fiber bent at the part directly irradiated by the supplementary light source under the premise that the optical fiber end face is fixed. The bending angle ranges from 0 to 90°. If the angle is too large, it will cause serious bending loss of the optical fiber, affecting the experimental results and even destroying the structure of the optical fiber;

[0010] It should be noted that, in general, the method for binarizing digital images is to manually select a threshold to divide the grayscale value of the image. The grayscale value greater than the threshold will be reassigned to 1, appearing white; the grayscale value less than the threshold will be reassigned to 0, appearing black. Since the artificial selection of the threshold will bring certain errors, especially the loss of image accuracy near the edge of the object, this method is applicable in simple scenarios, but for the optical fiber of the present invention, it is necessary to retain the edge characteristics of the object to the greatest extent. Therefore, the present invention uses the maximum inter-class variance method to process the high-definition grayscale image to achieve adaptive binarization.

[0011] Preferably, the output resolution of the initial image of the optical fiber end face is not less than 1280*720;

[0012] Furthermore, the color gamut conversion processing process specifically adopts a weighted average method to pre-process the initial image of the optical fiber end face. The weighted average method refers to performing weighted averaging on the R, G, and B values ​​of the initial image of the optical fiber end face according to a certain weight, and then converting the RGB primary color space into the YCbCr brightness color difference space, and then extracting the Y component to output the grayscale image of the optical fiber end face.

[0013] Further preferably, the steps of using median filtering to perform image salt and pepper noise filtering on the optimal optical fiber end face grayscale image are as follows: first, two rows of pixels are cached using a field programmable gate array FPGA and a static random access memory SRAM architecture, and processing begins when the third row of pixels arrives, forming a filtering template in the form of a 3*3 pixel matrix; secondly, a quick sorting algorithm is used to extract the median in the 3*3 pixel value matrix; finally, the original pixel value is replaced by the pixel median.

[0014] Preferably, the coordinates of the center position of the target area obtained by the optical fiber end face positioning method can be recorded and stored in real time using FPGA register resources. By using the hardware acceleration characteristics brought by the FPGA architecture, high-speed algorithm processing and recognition of the end face image can be performed with almost zero delay. At the same time, the result output after image processing is used to accurately control the stepper motor to drive the optical fiber to rotate.

[0015] In the second aspect, the present invention provides an application of the above-mentioned optical fiber end face positioning method in polarization-maintaining optical fiber fusion splicing. After obtaining high-precision core center position coordinates and stress zone center position coordinates through the optical fiber end face positioning method of the present invention, the corresponding optical fiber reference coordinates including the core center position reference coordinates and the stress zone center position reference coordinates can be selected according to the actual optical fiber production purpose and needs; the optical fiber end face positioning method of the present invention has a high advantage in the production of polarization-maintaining optical fiber core or quarter wave plate;

[0016] When the optical fiber end face positioning method of the present invention is used for polarization-maintaining optical fiber butt welding, based on the selected optical fiber reference coordinates as a reference, the currently obtained fiber core center position coordinates are first compared with the fiber core center position reference coordinates, and the stepper motor is driven to move in the plane where the optical fiber end face is located to adjust the fiber core center position, and then the currently obtained coordinates of the stress area center are compared with the stress area center position reference coordinates, and the stepper motor is driven to rotate to make consistency adjustment; it should be noted that when the stepper motor is driven to rotate, since the core position is fixed at this time, the horizontal and vertical coordinate values ​​of the core will not change too much, while the horizontal and vertical coordinates of the stress area centers on both sides of the optical fiber will change greatly.

[0017] In a third aspect, the present invention provides an application of the above-mentioned optical fiber end face positioning method in the production of a quarter wave plate. When the optical fiber end face positioning method of the present invention is used in the production of an all-fiber quarter wave plate, the optical fiber on one side selects the horizontal coordinate of the core as a reference, and the optical fiber on the other side should select the vertical coordinate of the core on this side as a reference. It is only necessary to rotate the horizontal coordinate of the center of the stress zone of the optical fiber end face at one end to align with the horizontal coordinate of the core, and rotate the vertical coordinate of the center of the stress zone at the other end to align with the vertical coordinate of the core. At this time, the optical fiber end faces at both ends should be 90°, and the fast axis and the slow axis of the optical fiber end face should remain perpendicular to each other, wherein the fast axis is an optical axis with a small refractive index and a faster light transmission speed, which passes vertically through the midpoint of the line connecting the centers of the two stress zones, and the slow axis is an optical axis passing through the end points of the two stress zones, which is a direction with a large refractive index and a slower transmission speed. The schematic diagram is as follows Figure 2 shown.

[0018] Compared with the prior art, the present invention has the following beneficial effects:

[0019] The optical fiber end face positioning method provided by the present invention not only cleverly reduces the difficulty of image recognition by acquiring the optical fiber end face image and coordinating the transformation and selection of the lighting mode, but also can realize the efficient and accurate docking of the polarization-maintaining optical fiber end face, thereby reducing the optical fiber fusion loss and maintaining a stable extinction ratio;

[0020] The optical fiber end face positioning method provided by the present invention is suitable for the end face alignment and fusion splicing of different polarization-maintaining optical fibers. It can be used for the core alignment of panda-type polarization-maintaining optical fibers, and can also be applied to other polarization-maintaining optical fibers, such as bowtie-type and elliptical polarization-maintaining optical fibers. It can also support 90° alignment of end faces in the preparation of optical fiber quarter-wave plates, and has a wide range of applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following is a brief introduction to the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. As shown in the accompanying drawings, the above and other purposes, features and advantages of the present invention will be clearer. The same reference numerals in all the drawings indicate the same parts. The drawings are not deliberately scaled to the actual size, and the focus is on illustrating the main purpose of the present invention.

[0022] Figure 1 A flow chart of a method for locating an optical fiber end face provided by an embodiment of the present invention;

[0023] Figure 2 A schematic diagram of the fast axis and slow axis of the optical fiber according to the present invention;

[0024] Figure 3 It is a grayscale image of the optical fiber end face obtained after the color gamut conversion process described in the specific implementation manner;

[0025] Figure 4 This is a schematic diagram of a specific lighting experiment performed on an optical fiber as described in a specific implementation manner;

[0026] Figure 5 It is a schematic diagram of the change of the gray value difference curve between the target area and the background area in the gray image of the optical fiber end face in the specific implementation manner;

[0027] Figure 6 This is a schematic diagram of the optical fiber end face after median filtering processing in a specific implementation manner;

[0028] Figure 7 This is a schematic diagram of the optical fiber end face after the adaptive binarization processing in the specific implementation manner;

[0029] Figure 8 It is a schematic diagram of the optical fiber end face alignment effect in a specific implementation manner. DETAILED DESCRIPTION

[0030] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0031] The following embodiments specifically use the alignment calibration when two Panda-type polarization-maintaining optical fibers are fused as an example to specifically explain the operation process and application process of the optical fiber end face positioning method described in the present invention. When the optical fibers are fused, the fusion end faces of the two optical fibers need to be moved to a predetermined position so that the distance between the two optical fibers is very small but there is a slight error, and they are not aligned. At this time, the positional relationship between the two optical fiber end faces needs to be adjusted to align the two end faces.

[0032] In order to overcome the error defects of existing optical fiber alignment, the embodiment of the present invention provides an optical fiber end face positioning method and a core alignment calibration method for polarization-maintaining optical fiber fusion splicing using the above method. In this embodiment, two panda-type polarization-maintaining optical fibers with a core diameter of 9 microns are selected, such as Figure 1 The optical fiber end face positioning method shown includes the following steps:

[0033] S100 places the end faces of two optical fibers with a cross-sectional cutting angle of 3° on the fiber fusion splicer platform for preliminary alignment. The two optical fibers are preliminarily aligned through the fiber displacement platform according to the side view image. The preliminary alignment here is only a preliminary alignment visible to the naked eye, which can greatly save the displacement amplitude of subsequent high-precision adjustment.

[0034] S200 then changes the lighting conditions of the two optical fibers respectively to obtain the best optical fiber end face grayscale image. Specifically, a plurality of initial images of the optical fiber end faces, i.e., RGB images of the optical fiber end faces, are obtained respectively under different lighting conditions. In this embodiment, the RGB images of the two optical fiber end faces are obtained by using a prism system, and the image resolution is 1280*720, and each frame of the acquired image is transmitted to the FPGA device in real time. The color gamut conversion processing is performed on the initial images of the optical fiber end faces to obtain the corresponding optical fiber end face grayscale image. The grayscale value difference between the target area and the background area in each optical fiber end face grayscale image is obtained and compared. In this embodiment, the target area includes the fiber core and the center of the stress zone. The optical fiber end face grayscale image with the largest grayscale value difference is the best optical fiber end face grayscale image.

[0035] The color gamut conversion process specifically adopts a linear weighted method. Since Verilog HDL cannot perform floating-point operations in FPGA, this embodiment is specifically based on the conversion formula from the RGB primary color space to the YCbCr brightness color difference space, namely Y=0.299*R+0.587*G+0.114*B. By converting the floating-point number in the formula to a fixed point, the formula is converted by first expanding it by 256 times and then shifting it right by 8Bit. The final conversion formula is: Y=((77*R+150*G+29*B)>>8). At this time, combined with the design concept of the pipeline, the FPGA's rich adder, multiplier, shift register and other resources are used to transplant the formula to the FPGA platform to achieve hardware acceleration, extract the Y component, and finally extract the Y component to obtain a grayscale image of the optical fiber end face, such as Figure 3 The figure shows the best fiber end face grayscale image obtained in this embodiment.

[0036] In order to obtain the best fiber end face grayscale image, the fiber optic lighting conditions are changed and adjusted by adjusting the light source color temperature, power, the lighting angle of the fiber optic and the bending degree of the fiber optic itself. Specifically, in this embodiment, three types of light sources are selected, namely white light, warm white light and red light, and four power levels are selected, namely 0.5W, 1W, 1.2W and 1.5W. The lighting angle of the fiber optic is based on the horizontal plane where the fiber optic is located, and the rear of the fiber optic end face is taken as the initial angle, namely 0°. During the debugging process, the angle is increased in a clockwise or counterclockwise direction on the horizontal plane where the fiber optic is located until it increases to 180°. Each increase of 10° is regarded as a measured lighting angle until the fiber optic is illuminated from 0° to 180°, and the clockwise or counterclockwise direction is performed once each.

[0037] When adjusting the degree of bending of the optical fiber itself, the end face of the fixed optical fiber remains unchanged, and the middle position of the bent optical fiber is recorded as 0° when the optical fiber is straight. After that, each incremental bending of 15° is regarded as a measurement point of the optical fiber bending, until the measurement of the 0-90° bending of the optical fiber is completed. In this embodiment, white light is specifically selected as the initial light source with a power of 0.5W and a fixed lighting angle of 0°, while the optical fiber is kept straight. First, the degree of bending of the optical fiber is regarded as the only variable in the experiment, and the type of light source, power and lighting angle are kept unchanged. After the test and data recording of the variable of the degree of bending of the optical fiber are completed, the experimental variables are replaced one by one in the order of lighting angle, power and light source type until all experimental measurements and data recording are completed. The schematic diagram of the specific lighting experiment on the optical fiber is shown as follows Figure 4 At the same time, during the lighting change process, the grayscale values ​​of the fiber core and the center of the stress area are displayed and recorded in real time by the host computer through the serial port, and a grayscale value difference curve between the target area and the background area is drawn, as shown in Figure 5As shown. Finally, according to the curve, find the position where the grayscale difference between the fiber core and the center of the stress area is the largest, and fix the variables at this time; this embodiment is specifically in a laboratory space in sunny weather, and the lighting conditions corresponding to the best fiber end face grayscale images obtained by the two optical fibers are consistent, specifically a 1W white light source, 50° irradiation of the optical fiber on the left side, and the degree of bending of the optical fiber is 45°. It should be noted that during specific operations, the user can select the best light source power and lighting angle according to the environment, changes in optical fiber types and actual lighting conditions. The basis for selection is that the grayscale value difference between the target area of ​​the optical fiber and the background area is maximized;

[0038] S300 transplants the median filtering algorithm to the FPGA platform to process the optimal fiber end face grayscale image, filter the salt and pepper noise, smooth the image, and obtain a higher quality fiber end face image referred to as a high-definition grayscale image. The schematic diagram is as follows Figure 6 The steps of algorithm transplantation include: first, using the BRAM resources in the FPGA to cache two rows of pixels, waiting for the third row of pixels to arrive and start processing, forming a filter template in the form of a 3*3 pixel matrix; second, using the quick sorting algorithm to extract the median value in the 3*3 pixel value matrix, wherein the quick sorting algorithm schematic diagram is shown in Figure 7 Finally, the median value of the nine pixels is used to replace the original pixel value. It should be noted that the FPGA will consume three clock cycles when processing the median filter algorithm. In order to keep synchronization with the pixel data, the line and field synchronization signals also need to be offset by three clocks accordingly.

[0039] S400 processes the high-definition grayscale image using the maximum inter-class variance method to achieve adaptive binarization, and further locates the center of the fiber core and the stress zone. The schematic diagram is shown in FIG. Figure 7 As shown; first, calculate the histogram of the image, that is, all the pixels of the image are divided into 256 grayscale value intervals from 0 to 255, and count the number of pixels in each interval; second, normalize the histogram, that is, divide the number of pixels in each interval by the total number of pixels to limit it between 0 and 1; third, set a classification threshold, that is, a grayscale level, and start iterating from 0; fourth, through the normalized histogram, count the pixels of grayscale levels 0 to i and define the pixels with pixel values ​​in this range as foreground pixels, the proportion of foreground pixels in the entire image is w0, and count the average grayscale of foreground pixels u0; count the pixels of grayscale levels i to 255 and define the pixels with pixel values ​​in this range as background pixels, the proportion of background pixels in the entire image is w1, and count the average grayscale of background pixels u1; here, let the total average grayscale of the image be u2, and the inter-class variance be g. Where: u2 = w0*u0+w1*u1, g = w0(u0-u2) 2 +w1(u1-u2) 2 , substituting u2 into g, we can get: g = w0w1 (u0-u1)2 ; Fifth, the gray value of the threshold is increased by 1, and the process goes to step 4 until i reaches 256 and the iteration ends; finally, the value corresponding to the maximum g is used as the optimal threshold of the entire image to realize the adaptive binarization of the gray image of the fiber end face.

[0040] S500 uses 12 registers to record the horizontal and vertical coordinates of the cores of the two optical fiber end faces and the center of the stress area.

[0041] According to the requirements of realizing polarization-maintaining optical fiber core production, the optical fibers on both sides should simultaneously select the horizontal coordinate or vertical coordinate as the reference, and the stepper motor rotates the optical fiber until the horizontal and vertical coordinates of the center of the stress zone are consistent with the selected core reference horizontal coordinate or vertical coordinate, and then stops. At this time, the fast axis and slow axis of the optical fiber core end faces on both sides are consistent;

[0042] According to the function to be realized, the horizontal and vertical coordinates of the fiber core reference in the register are compared in real time with the horizontal and vertical coordinates of the center of the stress zone. When the values ​​are different, the stepper motor will be driven to rotate the optical fiber until the values ​​are the same to complete the corresponding function. It should be noted that since the position of the fiber core is fixed, the horizontal and vertical coordinates of the fiber core position will not change during the rotation of the optical fiber, while the horizontal and vertical coordinates of the center of the stress zone will continue to change with the rotation of the optical fiber until the expected function is achieved and the optical fiber stops rotating. The schematic diagram of the optical fiber end face alignment effect is shown in the figure. Figure 8 shown.

[0043] Test Example 1: Extinction Ratio Test

[0044] The above method is used to test the extinction ratio of the optical fiber after the core. This test uses the extinction ratio measurement method to test the effect. The test process is as follows:

[0045] Before pipeline splicing, one of the optical fibers in a group of optical fibers to be spliced ​​is inserted into the adapter and then connected to the extinction ratio tester. The extinction ratio value on the screen is read and the reading at this time is recorded as the initial extinction ratio. In the same way, the optical fiber after splicing is connected to the extinction ratio tester and the extinction ratio is read as the extinction ratio after splicing. The number of samples is five groups of optical fibers to be spliced. Each sample is measured 5 times and the average value of the 5 times is recorded in the table below.

[0046] Extinction ratio of optical fiber before and after splicing:

[0047] Experimental samples Group 1 Group 2 Group 3 Group 4 Group 5 Initial extinction ratio 43.3dB 47.9dB 46.1dB 46.7dB 47.4dB Extinction ratio after splicing 35.8dB 36.6dbB 35.5dB 37.6dB 38.3dB

[0048] The results of Test Example 1 show that the Panda-type polarization-maintaining optical fiber aligned by the above-mentioned core alignment method has a good extinction ratio after fusion splicing.

[0049] Test Example 2: Splice loss test

[0050] The test system includes a polarization-maintaining light source and an optical power meter. First, adjust the output power of the polarization-maintaining light source so that the output power is within the range of the optical power meter. After the system is preheated, read the stable optical power value P0 as a reference, switch the optical power meter to relative mode, and perform a zeroing operation; connect the fused optical fiber to the system, and note that the distance from the fusion point to the optical power meter is kept constant and greater than 2m each time. Measure the optical power value at this time, which is the fusion loss of the optical fiber after fusion. After multiple experiments, take the average value. The measurement results are as follows:

[0051] Number of experiments first Second time The third time Fourth Fifth Splice loss 0.033dB 0.035dB 0.042dB 0.037dB 0.039dB

[0052] The results of test example 2 show that the splicing loss of the Panda-type polarization-maintaining optical fiber aligned by the above-mentioned core alignment method is very low after splicing.

[0053] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. A method for locating an optical fiber end face, characterized in that: The following steps are involved: Set different lighting conditions and obtain initial images of the optical fiber end face under different lighting conditions; Performing color gamut conversion processing on the initial image of the optical fiber end face to obtain a corresponding optical fiber end face grayscale image; obtaining the grayscale value difference between the target area and the background area in each optical fiber end face grayscale image and comparing them, and the optical fiber end face grayscale image with the largest grayscale value difference is the optimal optical fiber end face grayscale image; The best optical fiber end face grayscale image is subjected to image salt and pepper noise filtering by using median filtering to obtain a higher quality optical fiber end face grayscale image marked as a high-definition grayscale image; The maximum inter-class variance algorithm is used to iterate the high-definition grayscale image to realize adaptive binarization, so as to obtain the center position coordinates of the target area in the image, which are the core center position coordinates and the stress area center position coordinates in the polarization-maintaining optical fiber; The color gamut conversion processing process specifically adopts the weighted average method to pre-process the initial image of the optical fiber end face. The weighted average method refers to converting the RGB primary color space into the YCbCr brightness color difference space after weighted averaging the R, G, and B values ​​of the initial image of the optical fiber end face according to a certain weight, and then extracting the Y component to output the grayscale image of the optical fiber end face.

2. The optical fiber end face positioning method according to claim 1, characterized in that: The different lighting conditions include: the color temperature of the supplementary light source, the power of the supplementary light source, the lighting angle of the supplementary light source and the bending degree of the optical fiber.

3. A method for locating an optical fiber end face according to claim 2, characterized in that: The color temperature of the supplementary light source can be changed within a range including white light, warm white light and red light.

4. The optical fiber end face positioning method according to claim 2, characterized in that: The power selection range of the supplementary light source is 0.5-1.5W.

5. The optical fiber end face positioning method according to claim 1, characterized in that: The output resolution of the initial image of the optical fiber end face is not less than 1280*720.

6. The optical fiber end face positioning method according to claim 1, characterized in that: The steps of using median filtering to perform salt and pepper noise filtering on the optimal optical fiber end face grayscale image are as follows: first, two rows of pixels are cached using an FPGA+SRAM architecture, and processing begins when the third row of pixels is obtained to form a filtering template in the form of a 3*3 pixel matrix; second, a quick sorting algorithm is used to extract the median in the 3*3 pixel value matrix; finally, the original pixel value is replaced by the pixel median.

7. The optical fiber end face positioning method according to claim 1, characterized in that: The coordinates of the center position of the target area are recorded and stored in real time using FPGA register resources, and the hardware acceleration characteristics brought by the FPGA architecture are used to perform high-speed algorithm processing and recognition on the end face image. At the same time, the results output after image processing are used to accurately control the stepper motor to drive the optical fiber to rotate.

8. Application of the optical fiber end face positioning method as described in any one of claims 1 to 7 in polarization-maintaining optical fiber fusion splicing.

9. Application of the optical fiber end face positioning method as described in any one of claims 1 to 7 in the production of a quarter wave plate.

Images