A method and system for testing optical fiber geometric parameters based on geometric templates

Abstract

The present invention belongs to the technical field of optical fiber testing, and discloses a method and system for testing optical fiber geometric parameters based on a geometric template. For any type of optical fiber, before the first geometric parameter test, a cross-sectional view of the optical fiber is obtained, and images of various claddings and various core layers in the cross-sectional view are extracted and used as a predefined image template for this type of optical fiber. When performing geometric parameter testing on this type of optical fiber, a cross-sectional view of the test optical fiber is obtained. Based on the predefined image template of this type of optical fiber, template matching is performed on all core layers and all claddings in the cross-sectional view of the test optical fiber to obtain the position information of each core layer and each cladding. Combining the position information, each core layer and each cladding are respectively matched with a number of standard geometric templates, and the geometric algorithm corresponding to the matched standard geometric template is called for fitting calculation to obtain the geometric parameter information of the optical fiber. The present invention has high universality and can quickly and intelligently realize the geometric parameter testing of various types of optical fibers.

Description

Technical Field

[0001] The present invention belongs to the technical field of optical fiber testing, and more specifically, relates to a method and system for testing geometric parameters of optical fibers based on a geometric template. Background Art

[0002] When testing the geometric parameters of current ordinary optical fibers, customized software is used. If it is necessary to test the geometric parameters of a new type of optical fiber, a complete set of test programs needs to be rewritten and the test parameters need to be adjusted. Non-professionals cannot respond quickly. Moreover, the development speed of new types of optical fibers is fast, and the current optical fiber testing instruments and methods cannot cope with it in a timely manner. In addition, the testing and measurement of existing special optical fibers are mainly completed manually by using a microscope, which has disadvantages such as insufficient measurement accuracy, high uncertainty, and long time consumption. Summary of the Invention

[0003] The present invention provides a method and system for testing geometric parameters of optical fibers based on a geometric template, so as to solve the problems in the prior art that the universality of the method and system for testing geometric parameters of optical fibers is poor, especially the problems that it is difficult to test and measure new types of special optical fibers, the timeliness is poor, or the measurement accuracy is low and the efficiency is low when using manual measurement.

[0004] The present invention provides a method for testing geometric parameters of optical fibers based on a geometric template. For any type of optical fiber, before the first geometric parameter test, a cross-sectional view of the optical fiber is obtained; images of various claddings and various core layers in the cross-sectional view are extracted, and the images of various claddings and various core layers are used as predefined image templates for the optical fiber and stored.

[0005] When testing the geometric parameters of the optical fiber, a cross-sectional view of the test optical fiber is obtained; based on the predefined image template of the optical fiber, template matching is performed on all core layers and all claddings in the cross-sectional view of the test optical fiber to obtain the position information of each core layer and each cladding of the test optical fiber; combining the position information, each core layer and each cladding are respectively matched with several standard geometric templates, and the geometric algorithm corresponding to the matched standard geometric template is called for fitting calculation to obtain the geometric parameter information of the optical fiber.

[0006] Preferably, after obtaining the cross-sectional view of the optical fiber or the cross-sectional view of the test optical fiber, the following is further included: preprocessing the cross-sectional view, and the preprocessing includes sharpening the edge of the optical fiber end face and adjusting the brightness and contrast of the image.

[0007] Preferably, one type of cladding corresponds to one cladding shape and one cladding size, and one type of core layer corresponds to one core layer shape and one core layer size.

[0008] Preferably, the standard geometric templates and their corresponding geometric algorithms can be imported into the optical fiber geometric parameter testing system by means of pre-storage and new addition.

[0009] Preferably, after the standard geometric template is imported into the optical fiber geometric parameter test system, it further includes: using the imported standard geometric template as a basic template, performing scaling and / or rotation processing on the basic template to generate a plurality of extended templates, and using both the basic template and the extended templates as standard geometric templates for subsequent matching.

[0010] Preferably, when performing template matching on all core layers and all cladding layers in the cross-sectional view of the test optical fiber respectively, calculate the similarity values between each core layer and each cladding layer in the cross-sectional view of the test optical fiber and a plurality of pre-defined image templates stored respectively, and perform template matching based on the maximum similarity value.

[0011] Preferably, when calculating the similarity values between each core layer and each cladding layer in the cross-sectional view of the test optical fiber and a certain pre-defined image template stored respectively, denote the pre-defined image template as the first template, move the first template on the cross-sectional view of the test optical fiber, calculate the similarity value between the first template and the covered window on the cross-sectional view, and obtain the estimated best position of the first template in the cross-sectional view based on the calculation result of the similarity value.

[0012] Preferably, when matching each core layer and each cladding layer with a plurality of standard geometric templates respectively, adopt a template matching method based on edge point gradient information.

[0013] Preferably, use the sobel operator to calculate the amplitude gradient and gradient direction of pixel points in the image; during the template matching process, use the pyramid hierarchical search strategy to search the ROI area.

[0014] On the other hand, the present invention provides an optical fiber geometric parameter test system based on a geometric template, including:

[0015] A pre-defined image template acquisition unit for acquiring and storing pre-defined image templates of various types of optical fibers, and the pre-defined image template of each optical fiber includes images of various cladding layers and various core layers of the optical fiber;

[0016] A standard geometric template acquisition unit for acquiring and storing a plurality of standard geometric templates and their corresponding geometric algorithms;

[0017] A first matching unit for performing template matching on all core layers and all cladding layers in the cross-sectional view of the test optical fiber respectively based on the pre-defined image template of the optical fiber, and obtaining the position information of each core layer and each cladding layer of the test optical fiber;

[0018] A second matching unit for combining the position information, matching each core layer and each cladding layer with a plurality of standard geometric templates respectively, and calling the geometric algorithm corresponding to the matched standard geometric template for fitting calculation to obtain the geometric parameter information of the optical fiber;

[0019] The optical fiber geometric parameter testing system based on a geometric template is used to implement the functions in the above-mentioned optical fiber geometric parameter testing method based on a geometric template.

[0020] One or more technical solutions provided in the present invention have at least the following technical effects or advantages:

[0021] In the present invention, a number of standard geometric templates and their corresponding geometric algorithms are pre-stored in the optical fiber geometric parameter testing system; for any type of optical fiber, before the first geometric parameter test, a cross-sectional view of the optical fiber is obtained, and images of various claddings and various cores in the cross-sectional view are extracted. The images of various claddings and various cores are used as predefined image templates for this type of optical fiber and stored; thus, the test preparation work is completed. When performing geometric parameter testing on this type of optical fiber, a cross-sectional view of the test optical fiber is obtained; based on the predefined image template of this type of optical fiber, template matching is performed on all cores and all claddings in the cross-sectional view of the test optical fiber to obtain the position information of each core and each cladding of the test optical fiber, that is, one matching is performed. Combining the position information, each core and each cladding are respectively matched with a number of standard geometric templates, that is, a second matching is performed. The geometric algorithm corresponding to the matched standard geometric template is called for fitting calculation to obtain the geometric parameter information of the optical fiber. Compared with the existing optical fiber measurement technologies, the present invention focuses on various types of optical fibers (including new special optical fibers). The present invention can quickly, conveniently and intelligently perform geometric measurements on almost any type of optical fiber by using two matches, with high universality, good timeliness, accurate results, high efficiency, greatly reducing manual participation, and enabling operators to quickly respond to the testing of new optical fibers. Description of the Drawings

[0022] Figure 1 It is a schematic flow chart of an optical fiber geometric parameter testing method based on a geometric template provided in Embodiment 1 of the present invention;

[0023] Figure 2 It is a schematic diagram of an elliptical geometric template;

[0024] Figure 3 It is a schematic diagram of a triangular geometric template;

[0025] Figure 4 It is a schematic diagram of a hexagonal geometric template. Detailed Embodiments

[0026] In order to better understand the above technical solutions, the above technical solutions will be described in detail below in conjunction with the accompanying drawings of the specification and specific embodiments.

[0027] Embodiment 1:

[0028] Embodiment 1 provides a method for testing geometric parameters of an optical fiber based on a geometric template, which mainly includes the following steps:

[0029] For any type of optical fiber, before the first geometric parameter test, obtain a cross-sectional view of the optical fiber; extract the images of various claddings and various cores in the cross-sectional view, and use the images of various claddings and various cores as the predefined image templates of the optical fiber and store them.

[0030] When performing a geometric parameter test on the optical fiber, obtain a cross-sectional view of the test optical fiber; based on the predefined image template of the optical fiber, perform template matching on all cores and all claddings in the cross-sectional view of the test optical fiber respectively to obtain the position information of each core and each cladding of the test optical fiber; combine the position information, match each core and each cladding with several standard geometric templates respectively, and call the geometric algorithms corresponding to the matched standard geometric templates for fitting calculation to obtain the geometric parameter information of the optical fiber.

[0031] That is, the present invention includes two stages: a preparation stage and a test stage.

[0032] Preparation stage: Obtain and store predefined image templates of multiple types of optical fibers, and the predefined image template of each optical fiber includes the images of various claddings and various cores of the optical fiber; obtain and store multiple standard geometric templates and their corresponding geometric algorithms.

[0033] Test stage: Obtain a cross-sectional view of the test optical fiber; perform one-time matching to identify all cores and all claddings in the cross-sectional view of the test optical fiber; perform secondary matching and use the geometric algorithms corresponding to the standard geometric templates for fitting calculation to obtain geometric parameter information.

[0034] Among them, the standard geometric templates and their corresponding geometric algorithms can be imported into the optical fiber geometric parameter test system by means of pre-storage and new addition. That is, the present invention can add new geometric templates and corresponding geometric algorithms at any time according to needs.

[0035] In addition, after the standard geometric template is imported into the optical fiber geometric parameter test system, the following steps may further be included: Use the imported standard geometric template as a basic template, perform scaling and / or rotation processing on the basic template to generate several extended templates, and use the basic template and the extended templates as standard geometric templates for subsequent matching. By expanding the basic template, the present invention can generate a large number of templates after a user provides a standard geometric template, which are used to match the corresponding images of the template with different angles and different sizes.

[0036] After obtaining the cross-sectional view of the optical fiber or obtaining the cross-sectional view of the test optical fiber, the following steps may further be included: Perform preprocessing on the cross-sectional view, and the preprocessing includes sharpening the edge of the optical fiber end face and adjusting the brightness and contrast of the image.

[0037] The present invention will be specifically described below.

[0038] Refer to Figure 1 , a method for testing geometric parameters of an optical fiber based on a geometric template provided by the present invention includes the following steps:

[0039] Step 1, generate a predefined image template.

[0040] For any type of optical fiber (including new special optical fibers, ordinary optical fibers, etc.), first obtain a cross-sectional view of the optical fiber, then extract the images of various claddings and various core layers in the cross-sectional view, and finally import the images of various claddings and various core layers as the predefined image template of the optical fiber into the system for storage. This step belongs to the preparation stage.

[0041] For example, adjust the optical path and the monochromator, and use a camera to obtain the cross-sectional view of the illuminated cladding and core layer in real time. If it is a multi-core optical fiber, obtain the cross-sectional view of the optical fiber when all core layers are fully illuminated.

[0042] One type of cladding corresponds to one cladding shape and one cladding size, and one type of core layer corresponds to one core layer shape and one core layer size. For example, if an optical fiber includes three circular cores of different sizes and one circular cladding of one size, the predefined image template of the optical fiber includes three core layer images and one cladding image. When testing this type of optical fiber later, use the above three core layer images and one cladding image to match and locate all core layers and one cladding in the cross-sectional view of the test optical fiber.

[0043] For more accurate and rapid subsequent template matching, the cross-sectional view of the obtained optical fiber can also be preprocessed. The preprocessing includes sharpening the edge of the optical fiber end face and adjusting the brightness and contrast of the image. For example, use the square transformation y = x 2 Reduce the contrast in the dark area to sharpen the edge of the optical fiber end face, and adjust the overall brightness and contrast of the image to improve the clarity of the obtained image.

[0044] Step 2, core-cladding template matching.

[0045] Perform one matching to identify and locate all core layers and claddings in the test optical fiber.

[0046] The step 2 includes the following sub-steps:

[0047] Step 201, when testing the geometric parameters of the optical fiber, obtain the cross-sectional view of the test optical fiber.

[0048] Step 202, based on the predefined image template of the optical fiber, perform template matching on all core layers and all claddings in the cross-sectional view of the test optical fiber respectively to obtain the position information of each core layer and each cladding of the test optical fiber.

[0049] When performing template matching on all core layers and all cladding layers in the cross-sectional diagram of the test optical fiber respectively, similarity values between each core layer and each cladding layer in the cross-sectional diagram of the test optical fiber and a number of stored predefined image templates are calculated, and template matching is performed based on the maximum similarity value.

[0050] When calculating the similarity values between each core layer and each cladding layer in the cross-sectional diagram of the test optical fiber and a certain predefined image template stored respectively, the predefined image template is denoted as the first template. The first template is moved on the cross-sectional diagram of the test optical fiber, the similarity value between the first template and the covered window on the cross-sectional diagram is calculated, and the estimated best position of the first template in the cross-sectional diagram is obtained based on the calculation result of the similarity value.

[0051] To better understand a single matching process, the single template matching algorithm is further described below.

[0052] Template matching can identify the part on the image that matches the predefined template. It is a process of moving the template on the entire image and calculating the similarity between the template and the covered window on the image. Template matching is achieved through two-dimensional convolution. In convolution, the value of the output pixel is obtained by multiplying the elements of two matrices and summing the results. One matrix represents the image itself, and the other matrix is the template, which is the convolution kernel. Let the original image and the template be f(x, y) and t(x, y) respectively, and the enhanced image after convolution is g(x, y). The convolution formula is: g(x, y) = f(x, y) * t(x, y), where * represents the convolution of the two functions. The discrete convolution formula is as follows: M and N are the sizes of the template.

[0053] By providing a basic template for comparison, the template is positioned at each possible position on the image to find a similar template in the source image. The matching process is to compare the source images in the dataset with the templates in the library in sequence through indexing. During the matching process, the comparison calculation results between the template and the image are stored in a matrix, and the estimated best position of the template in the dataset is given through a scoring algorithm. The key of the scoring algorithm lies in calculating the similarity of two images with the same dimension, which is essentially the sum of the multiplications of the corresponding pixel values of the images. The basic method for calculating the image correlation is to calculate the cross-correlation between them. This method will produce deviations due to the global brightness change of the pixels, but the brightness of the obtained cross-sectional diagram of the optical fiber is basically consistent under normal circumstances, so the deviation of this method can be ignored. Among them, Cross-Correlation(Image1Image2) = ∑ x,y Image1(x, y) × Image2(x, y), and the larger the value, the better the matching degree.

[0054] Step 3, geometric template matching.

[0055] Perform secondary matching to match the corresponding standard geometric template, and call the geometric algorithm corresponding to the matched standard geometric template to perform geometric data operations to obtain test data. That is, automatically match the core cladding according to the imported standard geometric template and call the fitting and testing algorithms of the corresponding geometric template to obtain the key geometric information of the image.

[0056] The primary matching in the present invention is template matching based on gray values to find the position of the core layer cladding in the image in a certain order (for example, from left to right and from top to bottom). The secondary matching in the present invention mainly uses a large number of templates generated by rotation and size change to match the correct standard geometric template at several fixed positions. The present invention can use the primary matching to locate the position of the core cladding, which facilitates the geometric algorithm in the secondary matching (i.e., geometric template matching) to find the center point coordinates of the core cladding faster and more accurately and fit and calculate other geometric parameters, and can also reduce the computational amount of searching for the position of the core cladding during geometric template matching and improve the matching accuracy.

[0057] The said Step 3 includes the following sub-steps:

[0058] Step 301, combine the position information and match each core layer and each cladding with a number of standard geometric templates respectively.

[0059] Among them, a number of standard geometric templates and their corresponding geometric algorithms are pre-stored in the system during the preparation stage. For example, it includes standard geometric templates and geometric algorithms such as circles, ellipses, triangles, polygons, etc. The geometric algorithms for standard geometric shapes can use existing algorithms.

[0060] When matching each core layer and each cladding with a number of standard geometric templates respectively, use the template matching method based on edge point gradient information. Use the sobel operator to calculate the amplitude gradient and gradient direction of the pixel points in the image; during the template matching process, use the pyramid hierarchical search strategy to search the ROI area.

[0061] To better understand the secondary matching process, the secondary template matching algorithm is further described below.

[0062] Since the standard geometric template has obvious edge features, a template matching method based on edge point gradient information is adopted. First, create a data set or template model from the edge of the template image, and this data set or template model will be used to search for images similar to or the same as the template image in the target image. Use the sobel operator to find the gradient and direction of the image. The sobel operator can perform planar convolution with the image in two directions, horizontal and vertical. For a certain point (x, y) in the template image, two gradient vectors G x , G y; The amplitude gradient of the pixel point: Calculate the gradient direction:

[0063] When calculating the gradient information of the edge points of the searched image (target image), after obtaining the ROI area of the bottom layer of the searched image by the image pyramid algorithm, the template image traverses and searches in the ROI area with the center point as the origin. The search process is to calculate the average value of the inner product of the gradient vectors of the corresponding edge points of the template and the sub-image.

[0064] The pyramid search uses a coarse-to-fine search method to effectively improve the matching speed and performs hierarchical processing on the template image and the image to be matched. 2×2 = 4 pixels can be averaged into one pixel, thus forming a secondary image, and then the same method is used to obtain a tertiary image with a lower resolution. After n times of hierarchical processing, n processed images are obtained. First, start from the layer with the lowest resolution, find the rough matching position first, and then perform matching on the image with a higher resolution (i.e., the image processed for the (n - 1)th time), and only perform matching at the rough position calculated last time, and so on, until the exact matching position is found in the image to be matched with the highest resolution.

[0065] Step 302: Call the geometric algorithm corresponding to the matched standard geometric template for fitting calculation to obtain the geometric parameter information of the optical fiber.

[0066] If an elliptical geometric template as Figure 2 shown is matched, then run the elliptical geometric algorithm to obtain the center coordinates of the ellipse and the lengths of the major and minor semi-axes; if a triangle as Figure 3 shown or a polygon image such as a hexagon as Figure 4 shown is matched, then run the corresponding polygon algorithm to obtain the center, vertex coordinates, and the distance from the vertex to the center.

[0067] The measurement results include: the position of the center of the cladding circle, the lengths of the major and minor semi-axes of the ellipse, the position of the center of the core circle, the lengths of the major and minor semi-axes of the ellipse; use formulas to calculate the diameter and out-of-roundness of the cladding circle, the diameter and out-of-roundness of the core circle; calculate the core pitch based on the obtained multiple center coordinates of the core layer, etc.

[0068] Example 2:

[0069] Example 2 provides an optical fiber geometric parameter testing system based on a geometric template, including:

[0070] A predefined image template acquisition unit for acquiring and storing predefined image templates of various types of optical fibers, and the predefined image template of each optical fiber includes images of various claddings and various core layers of the optical fiber;

[0071] A standard geometry template acquisition unit for acquiring and storing multiple standard geometry templates and their corresponding geometric algorithms;

[0072] A first matching unit for performing template matching on all core layers and all cladding layers in the cross-sectional view of the test optical fiber respectively based on the predefined image template of this type of optical fiber, to obtain the position information of each core layer and each cladding layer of the test optical fiber;

[0073] A second matching unit for combining the position information to match each core layer and each cladding layer with several standard geometry templates respectively, and calling the geometric algorithm corresponding to the matched standard geometry template to perform fitting calculation to obtain the geometric parameter information of the optical fiber;

[0074] The optical fiber geometric parameter test system based on geometric templates is used to implement the functions in the optical fiber geometric parameter test method based on geometric templates described in Embodiment 1.

[0075] Since Embodiment 2 is a system corresponding to the method provided in Embodiment 1, it will not be elaborated here. For the system functions, refer to the description of Embodiment 1.

[0076] The optical fiber geometric parameter test method and system based on geometric templates provided by the embodiments of the present invention at least include the following technical effects:

[0077] The present invention applies image processing and machine vision technologies. Based on the predefined image template, it can efficiently locate the positions of the core and cladding layers. Based on the standard geometry template, it can quickly construct templates corresponding to various types of optical fibers. And in the application, if a new type of optical fiber without a pre-stored geometric image is encountered, the corresponding template and algorithm can be added. This is much simpler and faster than rewriting a set of programs. The present invention has high universality and can be used for geometric tests of currently existing optical fibers or various optical fibers that will appear in the future, that is, it can be extended to geometric tests of almost any type of optical fiber such as multi-cladding, multi-core, polygon, and special-shaped optical fibers. In addition, the present invention can also reduce costs, simplify operations, and improve test efficiency; it can quickly, conveniently and intelligently perform geometric measurements on various types of optical fibers including special optical fibers, with accurate results, greatly reducing manual participation, enabling operators to quickly respond to the tests of any new type of optical fiber.

[0078] Finally, it should be noted that the above specific embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the examples, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered within the scope of the claims of the present invention.

Claims

1. A method for testing the geometric parameters of an optical fiber based on a geometric template, characterized in that, for any type of optical fiber, before the first geometric parameter test, a cross-sectional view of the optical fiber is obtained; images of various claddings and various core layers in the cross-sectional view are extracted, and the images of various claddings and various core layers are used as predefined image templates for the optical fiber and stored; when performing a geometric parameter test on the optical fiber, a cross-sectional view of the test optical fiber is obtained; based on the predefined image template of the optical fiber, template matching is performed on all core layers and all claddings in the cross-sectional view of the test optical fiber to obtain the position information of each core layer and each cladding of the test optical fiber; in combination with the position information, each core layer and each cladding are respectively matched with a number of standard geometric templates, and the geometric algorithm corresponding to the matched standard geometric template is called for fitting calculation to obtain the geometric parameter information of the optical fiber.

2. The method for testing the geometric parameters of an optical fiber based on a geometric template according to claim 1, characterized in that, after obtaining the cross-sectional view of the optical fiber or the cross-sectional view of the test optical fiber, it further includes: preprocessing the cross-sectional view, and the preprocessing includes sharpening the edge of the optical fiber end face and adjusting the brightness and contrast of the image.

3. The method for testing the geometric parameters of an optical fiber based on a geometric template according to claim 1, characterized in that, one type of cladding corresponds to one type of cladding shape and one type of cladding size, and one type of core layer corresponds to one type of core layer shape and one type of core layer size.

4. The method for testing the geometric parameters of an optical fiber based on a geometric template according to claim 1, characterized in that, the standard geometric template and its corresponding geometric algorithm can be imported into the optical fiber geometric parameter test system by means of pre-storage and new addition.

5. The method for testing the geometric parameters of an optical fiber based on a geometric template according to claim 4, characterized in that, after the standard geometric template is imported into the optical fiber geometric parameter test system, it further includes: taking the imported standard geometric template as a basic template, performing scaling and / or rotation processing on the basic template to generate a number of extended templates, and using the basic template and the extended templates as standard geometric templates for subsequent matching.

6. The method for testing the geometric parameters of an optical fiber based on a geometric template according to claim 1, characterized in that, when performing template matching on all core layers and all claddings in the cross-sectional view of the test optical fiber, the similarity values of each core layer and each cladding in the cross-sectional view of the test optical fiber with a number of stored predefined image templates are calculated, and template matching is performed based on the maximum similarity value.

7. The method for testing the geometric parameters of an optical fiber based on a geometric template according to claim 6, characterized in that, when calculating the similarity value of each core layer and each cladding in the cross-sectional view of the test optical fiber with a certain predefined image template stored, the predefined image template is denoted as the first template, the first template is moved on the cross-sectional view of the test optical fiber, the similarity value between the first template and the covered window on the cross-sectional view is calculated, and the estimated best position of the first template in the cross-sectional view is obtained based on the calculation result of the similarity value.

8. The method for testing the geometric parameters of an optical fiber based on a geometric template according to claim 1, It is characterized in that when matching each core layer and each cladding layer with a number of standard geometric templates respectively, a template matching method based on edge point gradient information is adopted 9. The geometric parameter testing method for optical fiber based on geometric template according to claim 8 It is characterized in that the amplitude gradient and gradient direction of pixel points in the image are calculated by using the sobel operator; during the template matching process, the pyramid hierarchical search strategy is used to search the ROI area 10. A geometric parameter testing system for optical fiber based on geometric template It is characterized in that comprising a predefined image template acquisition unit, configured to acquire and store predefined image templates of various types of optical fibers, and the predefined image template of each optical fiber includes images of various cladding layers and various core layers of this optical fiber a standard geometric template acquisition unit, configured to acquire and store a number of standard geometric templates and their corresponding geometric algorithms a first matching unit, configured to respectively perform template matching on all core layers and all cladding layers in the cross-sectional view of the test optical fiber based on the predefined image template of this optical fiber, and obtain the position information of each core layer and each cladding layer of the test optical fiber a second matching unit, configured to combine the position information, respectively match each core layer and each cladding layer with a number of standard geometric templates, and call the geometric algorithm corresponding to the matched standard geometric template to perform fitting calculation to obtain the geometric parameter information of the optical fiber the geometric parameter testing system for optical fiber based on geometric template is used to implement the functions in the geometric parameter testing method for optical fiber based on geometric template according to any one of claims 1-9

Images