Recognition And Judgment Of Stress Zone Of Polarization Maintaining Fiber

With the development of science and technology,polarization-maintaining fiber has been widely used in fields such as coherent optical communication,interference sensing,and integrated polarization devices due to its strong ability to maintain linearly polarized light.Linearly polarized light needs to be incident along the polarization axis of the stress zone of the polarization-maintaining fiber to maintain the polarization state.In the butt coupling of the polarization-maintaining fiber,the polarization axis needs to be aligned to reduce the polarization loss.Therefore,detecting the azimuth angle of the polarization axis in the stress zone of the polarization-maintaining fiber is crucial.However,these technologies cannot combine high detection accuracy,high efficiency,and good practicability into one,and there is an urgent need for a more efficient,more accurate,and more practical method for identifying the stress zone of a polarization-maintaining fiber.Aiming at this key issue,this paper proposes a unique method for obtaining the image of the end face of the optical fiber-bending optical fiber and lighting method,which realizes the collection of the best-effect image of the core of the polarization maintaining optical fiber.The control variable method was used to experimentally study the effects of polarization-maintaining fiber core images under different fiber bending radii,different lighting angles,different light powers and different light colors.The experimental results show that when the panda-type polarization-maintaining fiber with a cladding diameter of 125 μm has a bending radius of 10 mm and a 1.5 W power red LED lamp is incident from a lighting angle of about 45°,the image effect of the polarization-maintaining fiber core is best.That is,the brightness difference between the core and the cladding of the polarizationmaintaining fiber is the largest.After obtaining the best end-face image,based on the characteristic of the large brightness difference between the center of the stress zone and the cladding,the segmentation algorithm is used to extract the brightness center positions of the two stress zones.The connection is the slow axis of the polarization-maintaining fiber and the azimuth angle is calculated.This is extremely important for realizing high-efficiency coupling and high extinction ratio in the polarization-maintaining fiber fusion splicer.On the other hand,this paper proposes a new panda-type polarization-maintaining fiber stress zone positioning method—a three-dimensional image-based polarization-maintaining fiber stress zone identification method,which realizes the accurate identification and judgment of the stress zone of the panda-type polarization-maintaining fiber.Collecting the side-view light intensity image of the polarization-maintaining fiber rotated 3600 at 0.2370 intervals via the CMOS camera driven by FPGA,reading the side-view light intensity data from the memory and drawing the side-view light intensity distribution curve.The side-view light intensity distribution curve adopts mean value smoothing to reduce noise and reduce the distortion of the curve.The results show that the light intensity distribution curve has always maintained a three-peak shape,and the five eigenvalues of the three peaks HL,HC,HR and the peak widths DL and DR of the three peaks vary with the orientation of the polarization axis.The angle changes sensitively.In the process of recording one rotation of the fiber,when the characteristic value of the light intensity distribution curve collected by the upper CMOS satisfies HL=HR<HC and DL＝DR,the corresponding polarization axis azimuth angle is 90°.Using this as a benchmark,calculate the polarization axis azimuth θβ corresponding to the characteristic value of the light intensity distribution curve for each rotation of the fiber at 0.237°,and record the polarization axis azimuth θα based on the end image based on the detection of the central brightness area of the stress zone.Perform the same operation on the side CMOS to obtain a polarization axis azimuth angle θγ,which is transformed into the same coordinate system as the upper CMOS polarization axis azimuth angle θγ=(θβ+2700)mod 360°.The database is established according to the mapping relationship between the characteristic value of the light intensity distribution curve collected by the above CMOS and the mean value θ=(θα+θβ+θγ)/3 of the azimuth of the polarization axis.In actual detection,the corresponding polarization axis azimuth angle is found in the database according to the characteristic value of the side-view light intensity distribution curve,the pulse number calculated by the corresponding angle is sent to the stepping moto,and the polarization maintaining fiber is rotated to the vertical position of the slow axis.The experimental results show that the positioning accuracy of the polarization-maintaining fiber under different initial polarization axis azimuth angles are all within 0.5°.At the same time,this paper proposes a recognition algorithm for fiber end face cutting angle based on straight edge detection.In the process of rotating the fiber for one circle,the cutting angle of the fiber end face is obtained in real time according to the recognition algorithm to ensure that the fiber end face is flat and the cutting angle is within 1°.