Factory detection method and system for optical splitter

Abstract

The invention relates to a factory detection method and system for an optical splitter, and the method comprises the steps: deploying related modules, carrying out the collection of an appearance image of a to-be-detected optical splitter, carrying out the feature extraction, and judging whether the appearance of the to-be-detected optical splitter meets a preset requirement or not; the light source module emits a light signal, the light signal with the specified wavelength is divided into multiple paths of light signals after passing through the to-be-detected light splitter, and the intensity and power data of the multiple paths of light signals are recorded in sequence; calculating the splitting ratio, the insertion loss, the additional loss and the uniformity of the optical splitter to be detected according to the recorded intensity and power data of the multiple paths of optical signals; and the performance comprehensive judgment module judges whether the splitting ratio, the insertion loss, the additional loss and the uniformity of the to-be-detected optical splitter are in a preset splitting ratio range, a preset insertion loss range, a preset additional loss range and a preset uniformity range respectively and judges whether the to-be-detected optical splitter is qualified or not. Rapid and efficient detection is carried out before the optical splitter leaves a factory, and meanwhile, the labor consumption is reduced, so that the cost is reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of optical splitters, and particularly relates to a factory inspection method and system for optical splitters. Background Art

[0002] An optical splitter is one of the important passive components in an optical fiber link, mainly used to distribute the optical signal in one optical fiber to multiple optical fibers, or to combine the optical signals of multiple optical fibers into one optical fiber. Optical splitters are usually used in passive optical network (PON) systems, especially in FTTH (fiber to the home) applications, where they connect the optical line terminal (OLT) and the optical network unit (ONU) to enable multiple end-users to share one PON interface.

[0003] After the optical splitter manufacturer completes the production of the optical splitter, it is necessary to detect the various performances of the produced optical splitter before the optical splitter leaves the factory. Only after ensuring that each performance of the optical splitter meets the corresponding preset requirements can the optical splitter be shipped out of the factory.

[0004] When the existing optical splitter manufacturers conduct factory inspections, they set up special inspection stations for various performances and appearances. However, almost each station requires corresponding inspection personnel to perform corresponding operations for corresponding inspections. This method is very inefficient, and setting up multiple inspection stations results in low space utilization.

[0005] Therefore, how to perform rapid and efficient inspections before the optical splitter leaves the factory while reducing the amount of manual labor and thus reducing costs is a technical problem that urgently needs to be solved at present. Summary of the Invention

[0006] The purpose of the present invention is to provide a factory inspection method and system for optical splitters, which can perform rapid and efficient inspections before the optical splitters leave the factory while reducing the amount of manual labor and thus reducing costs.

[0007] To solve the above technical problems, the technical solutions adopted by the present invention are as follows:

[0008] In a first aspect, a factory inspection method for an optical splitter is provided, including the following steps:

[0009] S1: Deploy a light source module, an image acquisition module, a detection data acquisition module, a data calculation and analysis module, and a performance comprehensive judgment module at a specified position near the end process on the optical splitter production line. The image acquisition module is arranged at a specified position around the optical splitter to be detected;

[0010] S2: The image acquisition module acquires the appearance image of the optical splitter to be detected and transmits it to the data calculation and analysis module. The data calculation and analysis module extracts features from the acquired appearance image of the optical splitter to be detected, and based on the extracted features, analyzes and determines whether the appearance of the optical splitter to be detected meets the preset requirements. If so, execute step S3; if not, repair the optical splitter to be detected.

[0011] S3: The detection data acquisition module includes a fiber optic jumper test line and an optical fiber power meter. One end of the fiber optic jumper test line is connected to the output port of the light source module, and the other end is connected to the input port of the optical fiber power meter. The input port of the optical fiber power meter is connected to the designated output port of the optical splitter to be detected.

[0012] S4: The light source module emits an optical signal with a specified wavelength. The optical signal with the specified wavelength is split into multiple optical signals by the optical splitter to be detected, and the intensity and power data of the multiple optical signals are recorded in sequence.

[0013] S5: Calculate the splitting ratio, insertion loss, additional loss, and uniformity of the optical splitter to be measured according to the recorded intensity and power data of the multiple optical signals, and compare the calculated splitting ratio, insertion loss, additional loss, and uniformity of the optical splitter to be measured with the preset splitting ratio range, insertion loss range, additional loss range, and uniformity range respectively.

[0014] S6: The performance comprehensive judgment module judges whether the splitting ratio, insertion loss, additional loss, and uniformity of the optical splitter to be measured are respectively within the preset splitting ratio range, insertion loss range, additional loss range, and uniformity range. If so, determine that the optical splitter to be measured is qualified; otherwise, determine that the optical splitter to be measured is unqualified, and mark the performance values that do not meet the corresponding requirements on the optical splitter to be measured.

[0015] Preferably, the specific process of the data calculation and analysis module extracting features from the acquired appearance image of the optical splitter to be detected in step S2 is as follows:

[0016] S21: Convert the appearance image of the optical splitter to be detected into a corresponding grayscale image.

[0017] S22: Obtain a window with a specified size around the specified pixel point in the converted grayscale image. Using the pixel value of the specified pixel point as the threshold, compare the pixel value of each pixel point in the window with the threshold, mark the pixel points greater than the threshold as 1, and mark the pixel points less than or equal to the threshold as 0.

[0018] S23: Mark the pixels within the window of other pixel points in the converted grayscale image in sequence according to step S22, calculate the histogram of each window, and then construct the feature vector of the grayscale image, where the feature vector includes multiple feature values.

[0019] Preferably, the specific process of analyzing and judging whether the appearance of the optical splitter to be detected meets the preset requirements based on the extracted features in step S2 is as follows:

[0020] S24: Classify the appearance defects of the optical splitter into specified types, including scratches, indentations, and holes, and create corresponding feature values for each appearance defect type;

[0021] S25: Match the feature values in the feature vector with the corresponding feature values created for each appearance defect type, and judge whether the optical splitter to be detected has appearance defects according to the matching results.

[0022] Preferably, the specific process of calculating the splitting ratio, insertion loss, additional loss, and uniformity of the optical splitter to be detected according to the recorded intensity and power data of multiple optical signals in step S5 is as follows:

[0023] S51: Calculate the splitting ratio of the optical splitter to be detected through Ki = Pi / Ps*100%, where Ki is the splitting ratio of the optical branch, Pi is the driving power of the i-th optical branch, and Ps is the sum of the driving powers of all optical branches;

[0024] S52: Calculate the insertion loss of the optical splitter to be detected through IL = -10lg(po / pi)+0.2~0.4, where Pi is the driving power at the input end, po is the driving power at the output end, and 0.2~0.4 is the additional loss of the optical splitter;

[0025] S53: Calculate the uniformity of the optical splitter through uniformity = (maximum light intensity - minimum light intensity) / maximum light intensity, where the maximum light intensity refers to the light intensity value of the optical branch with the maximum light intensity among all optical branches, and the minimum light intensity is the light intensity value of the optical branch with the minimum light intensity among all optical branches.

[0026] Preferably, it also includes detecting the return loss of the optical splitter to be detected, and the specific process is as follows:

[0027] Detect the reflected power and incident power of the optical splitter, and then calculate the return loss according to return loss = 10*log(incident power / reflected power).

[0028] In a second aspect, a factory inspection system for an optical splitter is provided, which is used to implement the factory inspection method for an optical splitter described above. The system includes a light source module, an image acquisition module, a detection data acquisition module, a data calculation and analysis module, and a performance comprehensive judgment module. The light source module is connected to the optical splitter to be detected. The image acquisition module is arranged at a specified position around the optical splitter to be detected. The detection data acquisition module is connected to the optical splitter to be detected. The detection data acquisition module is connected to the data calculation and analysis module, and the data calculation and analysis module is connected to the performance comprehensive judgment module;

[0029] The image acquisition module is used to acquire the appearance image of the optical splitter to be detected and transmit it to the data calculation and analysis module;

[0030] The data calculation and analysis module extracts features from the acquired appearance image of the optical splitter to be detected, and analyzes and judges whether the appearance of the optical splitter to be detected meets the preset requirements based on the extracted features;

[0031] The detection data acquisition module includes a light jumper test line and an optical fiber power meter. The light source module emits an optical signal with a specified wavelength. The optical signal with the specified wavelength is divided into multiple optical signals after passing through the optical splitter to be detected. The light jumper test line and the optical fiber power meter acquire the intensity and power data of the multiple optical signals;

[0032] The data calculation and analysis module is used to calculate the splitting ratio, insertion loss, additional loss, and uniformity of the optical splitter to be detected according to the recorded intensity and power data of the multiple optical signals, and compare the calculated splitting ratio, insertion loss, additional loss, and uniformity of the optical splitter to be detected with the preset splitting ratio range, insertion loss range, additional loss range, and uniformity range respectively;

[0033] The performance comprehensive judgment module is used to judge whether the splitting ratio, insertion loss, additional loss, and uniformity of the optical splitter to be detected are respectively within the preset splitting ratio range, insertion loss range, additional loss range, and uniformity range.

[0034] The beneficial effects of the present invention include:

[0035] The factory inspection method and system for an optical splitter provided by the present invention deploy relevant modules to collect the appearance image of the optical splitter to be inspected and extract features, and determine whether the appearance of the optical splitter to be inspected meets the preset requirements; the light source module emits an optical signal, and the optical signal with the specified wavelength is divided into multiple optical signals after passing through the optical splitter to be inspected, and the intensity and power data of the multiple optical signals are recorded in sequence; the splitting ratio, insertion loss, additional loss, and uniformity of the optical splitter to be measured are calculated according to the recorded intensity and power data of the multiple optical signals; the performance comprehensive judgment module determines whether the splitting ratio, insertion loss, additional loss, and uniformity of the optical splitter to be measured are respectively within the preset splitting ratio range, insertion loss range, additional loss range, and uniformity range to determine whether the optical splitter to be measured is qualified. It realizes rapid and efficient detection before the optical splitter leaves the factory, reduces the use of labor, and thus reduces costs.

[0036] First, by converting the appearance image of the optical splitter to be inspected into a corresponding grayscale image, the dimensionality reduction of the image is achieved, the complexity of subsequent image processing is reduced, and the image processing efficiency is improved. A window with a specified size around the specified pixel point is obtained, and taking the pixel value of the specified pixel point as the threshold, the pixel value of each pixel point in the window is compared with the threshold, and the pixel points greater than the threshold are marked as 1, and the pixel points less than or equal to the threshold are marked as 0; the histogram of each window is calculated, and then the feature vector of the grayscale image is constructed, realizing the accurate and rapid extraction of the key features of the appearance image of the optical splitter, and preparing for subsequent image analysis.

[0037] Second, by classifying the appearance defects of the optical splitter into specified types, the feature values in the feature vector are matched with the corresponding feature values created under each appearance defect type, and whether there are appearance defects in the optical splitter to be inspected is judged according to the matching result, realizing the convenient and efficient defect detection of the appearance of the optical splitter. The whole process does not require any manual participation at all, greatly reducing the labor, and at the same time the efficiency is effectively improved.

[0038] Finally, the splitting ratio of the optical splitter to be inspected is calculated by a formula, the insertion loss of the optical splitter to be inspected is calculated, and the key performance values such as the uniformity of the optical splitter are calculated, so as to perform precise detection on the optical splitter before leaving the factory, effectively ensuring that the optical splitter after leaving the factory can meet the corresponding requirements, and facilitating the timely repair of the optical splitters that do not meet the corresponding requirements, and improving the production qualification rate of the products. Description of the Drawings

[0039] Figure 1 It is a schematic flow chart of the factory inspection method for the optical splitter of the present invention.

[0040] Figure 2 It is a schematic flow chart of the feature extraction of the present invention.

[0041] Figure 3 Schematic diagram of the architecture of the factory inspection system for the optical splitter of the present invention. Detailed implementation manners

[0042] The following is a further detailed description of the present invention in conjunction with the attached Figures 1 to 3 drawings:

[0043] Embodiment 1

[0044] Referring to the attached Figure 1 drawings, a factory inspection method for an optical splitter includes the following steps:

[0045] S1: Deploy a light source module, an image acquisition module, a detection data acquisition module, a data calculation and analysis module, and a performance comprehensive judgment module at a specified position near the end process on the optical splitter production line. The image acquisition module is arranged at a specified position around the optical splitter to be detected. Step S1 is to deploy equipment for the factory inspection of the optical splitter, so as to perform subsequent relevant data acquisition and data analysis to know whether the optical splitter to be detected meets the corresponding performance and appearance requirements.

[0046] S2: The image acquisition module acquires the appearance image of the optical splitter to be detected and transmits it to the data calculation and analysis module. The data calculation and analysis module extracts the features of the acquired appearance image of the optical splitter to be detected and analyzes and judges whether the appearance of the optical splitter to be detected meets the preset requirements based on the extracted features. If so, execute step S3; if not, repair the optical splitter to be detected. This process realizes automatically acquiring the appearance image of the detected optical splitter, then automatically extracting the image features of the acquired appearance image, and then analyzing whether there are appearance defects in the optical splitter to be detected through the analysis of the extracted image features.

[0047] S3: The detection data acquisition module includes a light jumper test line and an optical fiber power meter. One end of the light jumper test line is connected to the output port of the light source module, and the other end is connected to the input port of the optical fiber power meter. The input port of the optical fiber power meter is connected to the specified output port of the optical splitter to be detected. The detection data acquisition module is mainly used to obtain the intensity and optical power of the optical signals of each optical branch.

[0048] S4: The light source module emits an optical signal with a specified wavelength. The optical signal with the specified wavelength is divided into multiple optical signals after passing through the optical splitter to be detected, and the intensity and power data of the multiple optical signals are recorded in sequence, which is convenient for setting and analyzing relevant parameters at a specific wavelength.

[0049] S5: Calculate the splitting ratio, insertion loss, additional loss, and uniformity of the optical splitter under test based on the intensity and power data of the recorded multi-channel optical signals, and compare the calculated splitting ratio, insertion loss, additional loss, and uniformity of the optical splitter under test with the preset splitting ratio range, insertion loss range, additional loss range, and uniformity range respectively;

[0050] S6: Use the performance comprehensive judgment module to judge whether the splitting ratio, insertion loss, additional loss, and uniformity of the optical splitter under test are respectively within the preset splitting ratio range, insertion loss range, additional loss range, and uniformity range. If so, determine that the optical splitter under test is qualified; otherwise, determine that the optical splitter under test is unqualified, and mark the performance values that do not meet the corresponding requirements on the optical splitter under test.

[0051] In this embodiment, relevant modules are deployed to collect the appearance image of the optical splitter to be detected and perform feature extraction, and judge whether the appearance of the optical splitter to be detected meets the preset requirements; the light source module emits an optical signal, and the optical signal with the specified wavelength is divided into multi-channel optical signals after passing through the optical splitter to be detected, and the intensity and power data of the multi-channel optical signals are recorded in sequence; calculate the splitting ratio, insertion loss, additional loss, and uniformity of the optical splitter under test according to the recorded intensity and power data of the multi-channel optical signals; use the performance comprehensive judgment module to judge whether the splitting ratio, insertion loss, additional loss, and uniformity of the optical splitter under test are respectively within the preset splitting ratio range, insertion loss range, additional loss range, and uniformity range to determine whether the optical splitter under test is qualified. It realizes rapid and efficient detection before the optical splitter leaves the factory, reduces the use of manpower, and thus reduces costs.

[0052] Embodiment 2

[0053] Based on Embodiment 1, refer to Figure 2 , the specific process of the data calculation and analysis module in step S2 for feature extraction of the collected appearance image of the optical splitter to be detected is as follows:

[0054] S21: Convert the appearance image of the optical splitter to be detected into a corresponding grayscale image. By converting to a grayscale image, since a grayscale image has only a single luminance channel, compared with the three color channels of a color image, the grayscale image reduces the complexity of data processing. This simplification can not only reduce the computational amount of image processing, improve the processing speed, but also reduce the storage space requirement to a certain extent. Therefore, by removing the color information in the image and only retaining the luminance information, the subsequent image processing operations are simplified.

[0055] S22: Obtain a window of a specified size around a specified pixel point in the converted grayscale image. Using the pixel value of the specified pixel point as a threshold, compare the pixel values of each pixel point within the window with the threshold. Mark the pixel points greater than the threshold as 1 and the pixel points less than or equal to the threshold as 0, achieving the classification of pixel points in the grayscale avatar and providing convenience for subsequent feature extraction.

[0056] S23: Sequentially mark the pixels within the windows of other pixel points in the converted grayscale image according to step S22, calculate the histogram of each window, and then construct the feature vector of this grayscale image. The feature vector includes multiple feature values.

[0057] In this embodiment, the specific process of analyzing and determining whether the appearance of the optical splitter to be detected meets the preset requirements based on the extracted features in step S2 is as follows:

[0058] S24: Classify the appearance defects of the optical splitter into specified types, including scratches, indentations, and holes, and create corresponding feature values for each type of appearance defect;

[0059] S25: Match the feature values in the feature vector with the corresponding feature values created for each type of appearance defect, and determine whether there are appearance defects in the optical splitter to be detected according to the matching results.

[0060] By classifying the appearance defects of the optical splitter into specified types, matching the feature values in the feature vector with the corresponding feature values created for each type of appearance defect, and determining whether there are appearance defects in the optical splitter to be detected according to the matching results, convenient and efficient defect detection of the appearance of the optical splitter is achieved. The entire process requires no human participation at all, greatly reducing labor while effectively improving efficiency.

[0061] Embodiment 3

[0062] Based on Embodiment 1 or Embodiment 2, the specific process of calculating the splitting ratio, insertion loss, additional loss, and uniformity of the optical splitter to be measured according to the recorded intensity and power data of multiple optical signals in step S5 is as follows:

[0063] S51: Calculate the splitting ratio of the optical splitter to be detected through Ki = Pi / Ps*100%, where Ki is the splitting ratio of the optical branch, Pi is the driving power of the i-th optical branch, and Ps is the sum of the driving powers of all optical branches;

[0064] S52: Calculate the insertion loss of the optical splitter to be detected through IL = -10lg(po / pi)+0.2~0.4, where Pi is the driving power at the input end, po is the driving power at the output end, and 0.2~0.4 is the additional loss of the optical splitter;

[0065] S53: Calculate the uniformity of the optical splitter through the formula: uniformity = (maximum optical intensity - minimum optical intensity) / maximum optical intensity, where the maximum optical intensity refers to the optical intensity value of the optical branch with the maximum optical intensity among all optical branches, and the minimum optical intensity is the optical intensity value of the optical branch with the minimum optical intensity among all optical branches.

[0066] Calculate key performance values such as the splitting ratio of the optical splitter to be detected, the insertion loss of the optical splitter to be detected, and the uniformity of the optical splitter. Conduct precise detection on the optical splitter before leaving the factory, effectively ensuring that the optical splitter after leaving the factory can meet the corresponding requirements, and facilitating timely repair of the optical splitters that do not meet the corresponding requirements, thereby improving the production qualification rate of the product.

[0067] In this embodiment, it also includes detecting the return loss of the optical splitter to be detected. The specific process is as follows:

[0068] Detect the reflected power and incident power of the optical splitter, and then calculate the return loss according to the formula: return loss = 10 * log (incident power / reflected power).

[0069] An ex-factory detection system for an optical splitter, used to implement the ex-factory detection method of the optical splitter described above. Refer to Figure 3 , including a light source module, an image acquisition module, a detection data acquisition module, a data calculation and analysis module, and a performance comprehensive judgment module. The light source module is connected to the optical splitter to be detected. The image acquisition module is arranged at a specified position around the optical splitter to be detected. The detection data acquisition module is connected to the optical splitter to be detected. The detection data acquisition module is connected to the data calculation and analysis module, and the data calculation and analysis module is connected to the performance comprehensive judgment module.

[0070] An image acquisition module is used to acquire the appearance image of the optical splitter to be detected and transmit it to the data calculation and analysis module; the data calculation and analysis module extracts features from the acquired appearance image of the optical splitter to be detected and analyzes and judges whether the appearance of the optical splitter to be detected meets the preset requirements based on the extracted features. The detection data acquisition module includes a fiber optic jumper test line and an optical fiber power meter. The light source module emits an optical signal with a specified wavelength. The optical signal with the specified wavelength is divided into multiple optical signals after passing through the optical splitter to be detected. The fiber optic jumper test line and the optical fiber power meter acquire the intensity and power data of the multiple optical signals. The data calculation and analysis module is used to calculate the splitting ratio, insertion loss, additional loss, and uniformity of the optical splitter to be measured according to the recorded intensity and power data of the multiple optical signals, and compare the calculated splitting ratio, insertion loss, additional loss, and uniformity of the optical splitter to be measured with the preset splitting ratio range, insertion loss range, additional loss range, and uniformity range respectively. The performance comprehensive judgment module is used to judge whether the splitting ratio, insertion loss, additional loss, and uniformity of the optical splitter to be measured are respectively within the preset splitting ratio range, insertion loss range, additional loss range, and uniformity range.

[0071] In summary, the factory inspection method and system for the optical splitter provided by the present invention deploy relevant modules to acquire and extract features from the appearance image of the optical splitter to be detected, and judge whether the appearance of the optical splitter to be detected meets the preset requirements; the light source module emits an optical signal, and the optical signal with the specified wavelength is divided into multiple optical signals after passing through the optical splitter to be detected, and the intensity and power data of the multiple optical signals are recorded in sequence; the splitting ratio, insertion loss, additional loss, and uniformity of the optical splitter to be measured are calculated according to the recorded intensity and power data of the multiple optical signals; the performance comprehensive judgment module judges whether the splitting ratio, insertion loss, additional loss, and uniformity of the optical splitter to be measured are respectively within the preset splitting ratio range, insertion loss range, additional loss range, and uniformity range to determine whether the optical splitter to be measured is qualified. It realizes fast and efficient detection before the optical splitter leaves the factory, reduces the use of labor, and thus reduces costs.

[0072] By converting the appearance image of the optical splitter to be detected into a corresponding grayscale image, the dimensionality of the image is reduced, the complexity of subsequent image processing is decreased, and the image processing efficiency is improved. A window of a specified size around a specified pixel point is obtained. Using the pixel value of the specified pixel point as a threshold, the pixel values of each pixel point within the window are compared with the threshold. Pixel points with values greater than the threshold are marked as 1, and pixel points with values less than or equal to the threshold are marked as 0. The histogram of each window is calculated, and then the feature vector of this grayscale image is constructed, achieving the accurate and rapid extraction of the key features of the optical splitter appearance image, preparing for subsequent image analysis. By classifying the appearance defects of the optical splitter into specified types, the feature values in the feature vector are matched with the corresponding feature values created for each appearance defect type, and based on the matching results, it is determined whether the optical splitter to be detected has appearance defects, realizing the convenient and efficient defect detection of the optical splitter appearance. The entire process requires no human participation at all, greatly reducing labor and effectively improving efficiency at the same time. By calculating key performance values such as the splitting ratio of the optical splitter to be detected, the insertion loss of the optical splitter to be detected, and the uniformity of the optical splitter through formulas, precise detection is carried out before the optical splitter leaves the factory, effectively ensuring that the optical splitters after leaving the factory can meet the corresponding requirements, and facilitating the timely repair of the optical splitters that do not meet the corresponding requirements, improving the production qualification rate of the product.

Claims

1. A factory inspection method for an optical splitter, characterized in that: The following steps are involved: S1: deploying a light source module, an image acquisition module, a detection data acquisition module, a data calculation and analysis module, and a comprehensive performance judgment module at a designated position near the terminal process on the optical splitter production line, wherein the image acquisition module is arranged at a designated position around the optical splitter to be detected; S2: The image acquisition module acquires the appearance image of the optical splitter to be detected, and transmits it to the data calculation and analysis module, the data calculation and analysis module extracts features from the acquired appearance image of the optical splitter to be detected, and analyzes and determines whether the appearance of the optical splitter to be detected meets the preset requirements based on the extracted features. If so, execute step S3; if not, return the optical splitter to be detected for repair; S3: The detection data acquisition module includes a light jumper test line and a fiber optic power meter. One end of the light jumper test line is connected to the output port of the light source module, and the other end is connected to the input port of the fiber optic power meter. The input port of the fiber optic power meter is connected to the designated output port of the optical splitter to be detected. S4: The light source module emits an optical signal of a specified wavelength, the optical signal of the specified wavelength is divided into multiple optical signals after passing through the optical splitter to be detected, and the intensity and power data of the multiple optical signals are recorded in sequence; S5: Calculate the splitting ratio, insertion loss, additional loss, and uniformity of the optical splitter to be tested according to the recorded intensity and power data of the multiple optical signals, and compare the calculated splitting ratio, insertion loss, additional loss, and uniformity of the optical splitter to be tested with the preset splitting ratio range, insertion loss range, additional loss range, and uniformity range, respectively; S6: The comprehensive performance judgment module is used to judge whether the splitting ratio, insertion loss, additional loss and uniformity of the optical splitter to be tested are respectively within the preset splitting ratio range, insertion loss range, additional loss range and uniformity range. If so, the optical splitter to be tested is judged to be qualified; otherwise, the optical splitter to be tested is judged to be unqualified, and the performance values ​​that do not meet the corresponding requirements are marked on the optical splitter to be tested.

2. The factory inspection method of an optical splitter according to claim 1, characterized in that: The specific process of the data calculation and analysis module in step S2 performing feature extraction on the collected appearance image of the optical splitter to be detected is as follows: S21: converting the appearance image of the optical splitter to be detected into a corresponding grayscale image; S22: Obtain a window of a specified size around a designated pixel in the converted grayscale image, take the pixel value of the designated pixel as a threshold, compare the pixel value of each pixel in the window with the threshold, mark the pixel value greater than the threshold as 1, and mark the pixel value less than or equal to the threshold as 0; S23: sequentially marking the pixels in the window of other pixel points in the converted grayscale image according to step S22, calculating the histogram of each window, and then constructing a feature vector of the grayscale image, wherein the feature vector includes multiple eigenvalues.

3. The factory inspection method of an optical splitter according to claim 2, characterized in that: The specific process of analyzing and judging whether the appearance of the optical splitter to be tested meets the preset requirements based on the extracted features in step S2 is as follows: S24: classify the appearance defects of the optical splitter into specified types, including scratches, indentations, and holes, and create corresponding feature values ​​under each appearance defect type; S25: Matching the eigenvalues ​​in the eigenvector with corresponding eigenvalues ​​created under each appearance defect type, and judging whether the optical splitter to be inspected has an appearance defect according to the matching result.

4. The factory inspection method of an optical splitter according to claim 1, characterized in that: The specific process of calculating the splitting ratio, insertion loss, additional loss and uniformity of the optical splitter to be tested according to the recorded intensity and power data of the multi-channel optical signals in step S5 is as follows: S51: Calculate the splitting ratio of the optical splitter to be tested by Ki=Pi / Ps*100%, where Ki is the splitting ratio of the optical branch, Pi is the driving power of the i-th optical branch, and Ps is the sum of the driving powers of all optical branches; S52: Calculate the insertion loss of the optical splitter to be tested by IL=-10lg(po / pi)+0.2~0.4, where Pi is the driving power of the input end, po is the driving power of the output end, and 0.2~0.4 is the additional loss of the optical splitter; S53: Calculate the uniformity of the optical splitter by uniformity = (maximum light intensity - minimum light intensity) / maximum light intensity, where the maximum light intensity refers to the light intensity value of the optical branch with the largest light intensity among all optical branches, and the minimum light intensity refers to the light intensity value of the optical branch with the smallest light intensity among all optical branches.

5. The factory inspection method of an optical splitter according to claim 4, characterized in that: It also includes testing the return loss of the optical splitter to be tested. The specific process is as follows: The reflected power and incident power of the optical splitter are detected, and the return loss is calculated according to return loss = 10*log (incident power / reflected power).

6. A factory inspection system for an optical splitter, used to implement a factory inspection method for an optical splitter as claimed in any one of claims 1 to 5, characterized in that: It includes a light source module, an image acquisition module, a detection data acquisition module, a data calculation and analysis module and a comprehensive performance judgment module, wherein the light source module is connected to the optical splitter to be detected, the image acquisition module is arranged at a designated position around the optical splitter to be detected, the detection data acquisition module is connected to the optical splitter to be detected, the detection data acquisition module is connected to the data calculation and analysis module, and the data calculation and analysis module is connected to the comprehensive performance judgment module; The image acquisition module is used to acquire the appearance image of the optical splitter to be detected and transmit it to the data calculation and analysis module; The data calculation and analysis module extracts features from the collected appearance image of the optical splitter to be detected, and analyzes and determines whether the appearance of the optical splitter to be detected meets preset requirements based on the extracted features; The detection data acquisition module includes a light jumper test line and an optical fiber power meter. The light source module emits an optical signal of a specified wavelength. The optical signal of the specified wavelength is divided into multiple optical signals after passing through an optical splitter to be detected. The light jumper test line and the optical fiber power meter obtain the intensity and power data of the multiple optical signals. The data calculation and analysis module is used to calculate the splitting ratio, insertion loss, additional loss and uniformity of the optical splitter to be tested according to the intensity and power data of the recorded multi-channel optical signals, and compare the calculated splitting ratio, insertion loss, additional loss and uniformity of the optical splitter to be tested with the preset splitting ratio range, insertion loss range, additional loss range and uniformity range respectively; The comprehensive performance judgment module is used to judge whether the splitting ratio, insertion loss, additional loss and uniformity of the optical splitter to be tested are respectively within the preset splitting ratio range, insertion loss range, additional loss range and uniformity range.

Images