Optical fiber end face angle measurement method and system, and storage medium

Abstract

The invention provides an optical fiber end face angle measurement method and system, and a storage medium, and the method comprises the steps: collecting an optical fiber end face image frame with a bevel edge feature, constructing a teaching data set, training a visual recognition network, predicting coordinate points at two ends of the bevel edge feature, and obtaining a recognition model; establishing a control strategy of an identification result and optical fiber radial rotation; placing the detection end of the to-be-detected optical fiber in a camera shooting area, and radially rotating the optical fiber according to a control strategy until an end face image frame of the to-be-detected optical fiber containing a bevel edge feature is identified and acquired; and constructing a virtual protractor, and aligning the original point of the virtual protractor with the vertex of the bevel edge feature in the to-be-measured optical fiber end face image frame to obtain an angle value on the virtual protractor corresponding to the bevel edge feature. Therefore, by means of optical fiber dynamic adjustment and image detection, the angle of the end face of the optical fiber is accurately measured by using a single camera.

Description

Technical Field

[0001] The invention relates to image detection technology, and in particular to a measurement method and system and a storage medium for detecting the end face angle of an optical fiber using the image detection technology. Background Art

[0002] Optical fiber is widely used in various fields, such as optical fiber communication, laser processing, optical fiber sensors, etc. With the advancement of optical fiber technology, the alignment accuracy of optical fiber end faces is increasingly required. The error of optical fiber end face angle will lead to signal loss and increased reflection. Therefore, the measurement of optical fiber end face angle is a key link to ensure the optical performance of optical fiber and the overall performance of the system.

[0003] The existing equipment for detecting the angle of the fiber end face is a fiber fusion splicer, which usually uses two cameras to sample different sides of the fiber, and calculates and determines the angle of the fiber end face based on the angle between the fiber end face in the fiber side image and the fiber itself. Since it only samples the two sides of the fiber, it is impossible to accurately obtain the maximum angle value of the fiber end face, which will cause errors in practical applications. Summary of the invention

[0004] To this end, the main purpose of the present invention is to provide a method and system for measuring the end face angle of an optical fiber, and a storage medium, so as to accurately measure the end face angle of an optical fiber using a single camera through dynamic adjustment of the optical fiber and image detection means.

[0005] In order to achieve the above object, according to one aspect of the present invention, a method for measuring an optical fiber end face angle is provided, the steps comprising:

[0006] Collect the image frames of the fiber end face with bevel features, build a teaching data set to train the visual recognition network, predict the coordinate points at both ends of the bevel features, obtain the recognition model, and establish the control strategy of the recognition results and the radial rotation of the fiber;

[0007] The detection end of the optical fiber to be tested is placed in the imaging area, and the optical fiber is radially rotated according to the control strategy until an image frame of the end face of the optical fiber to be tested containing the bevel edge feature is identified and obtained;

[0008] A virtual protractor is constructed, and the origin of the virtual protractor is aligned with the vertex of the bevel feature in the image frame of the optical fiber end face to be measured to obtain the angle value of the bevel feature on the virtual protractor corresponding to the bevel feature.

[0009] In a possible preferred embodiment, the step of establishing the identification result and the control strategy of optical fiber radial rotation includes:

[0010] Set the axial rotation speed of the optical fiber to be tested to match the recognition frame rate of the visual recognition network;

[0011] When the visual recognition network recognizes the bevel feature of the end face of the optical fiber to be tested, the rotation of the optical fiber to be tested is stopped.

[0012] In a possible preferred embodiment, the step of establishing the identification result and the control strategy of optical fiber radial rotation includes:

[0013] Set the axial rotation speed of the optical fiber to be tested to match the frame rate of the visual recognition network;

[0014] Set the radial rotation pre-start time of the optical fiber to be tested;

[0015] The rotation is started and after the pre-start time, when the visual recognition network recognizes the bevel edge feature of the end face of the optical fiber to be tested, the rotation of the optical fiber to be tested is stopped.

[0016] In a possible preferred embodiment, the step of constructing a virtual protractor and aligning the origin of the virtual protractor with the vertex of the bevel feature in the image frame of the optical fiber end face to be measured comprises:

[0017] Compare the coordinate values ​​of the two end points of the bevel feature, determine the coordinates of the vertex of the bevel feature, and set them as the origin of the virtual protractor;

[0018] According to the preset virtual protractor radius, the 0° and 90° points of the virtual protractor are determined at the origin respectively;

[0019] According to the preset virtual protractor radius, draw a semicircle, and according to the obtained point position, position the scale to draw scale lines in equal parts, and add scale labels to the scale lines.

[0020] In a possible preferred embodiment, the optical fiber end face angle measurement method further comprises the following steps:

[0021] According to the image recognition result of the optical fiber to be tested, the distance between the coordinate points at both ends of the bevel feature is calculated to obtain the bevel feature length;

[0022] According to the characteristic length of the hypotenuse, the radius of the virtual protractor is dynamically adjusted to scale the virtual protractor proportionally.

[0023] In a possible preferred embodiment, the step of obtaining the angle value on the virtual protractor corresponding to the bevel feature includes:

[0024] Determine the coincident scale line of the virtual protractor at which the bottom point of the hypotenuse feature is located, and read the value of the corresponding scale label as the angle value.

[0025] In a possible preferred embodiment, the step of obtaining the angle value on the virtual protractor corresponding to the bevel feature includes:

[0026] Identify the coincident scale lines of the virtual protractor where the hypotenuse feature is located, and read the value of the corresponding scale label as the angle value.

[0027] In order to achieve the above-mentioned purpose, according to another aspect of the present invention, there is also provided an optical fiber end face angle measuring device, which includes: a visual recognition unit, an optical fiber clamp, and a rotating device, wherein the optical fiber clamp is provided with an optical fiber groove to axially position the optical fiber to be measured and allow the end face of the optical fiber to be measured to extend from one end of the optical fiber clamp to the camera area of ​​the visual recognition unit, the rotating device is matched with the optical fiber to be measured, and the visual recognition unit is controlled and connected to the rotating device, so as to control the rotating device to rotate the optical fiber to be measured according to the steps of any of the above-mentioned optical fiber end face angle measuring methods until the visual recognition unit obtains the complete bevel edge feature of the optical fiber to be measured, and then the end face angle of the optical fiber to be measured is measured with a virtual protractor.

[0028] In a possible preferred embodiment, the optical fiber end face angle measuring device further includes: a light source unit, a display screen, and a ruler, wherein the ruler is arranged at one end of the optical fiber clamp to guide the optical fiber to be measured to extend outward to a positioning position within the camera area of ​​the visual recognition unit, the light emitting side of the light source unit faces the optical fiber to be measured within the camera area, and the display screen is connected to the visual recognition unit to show the relative measurement posture of the end face of the optical fiber to be measured and the virtual protractor.

[0029] In order to achieve the above object, according to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, wherein when the computer program is executed, the steps of any of the above-mentioned methods for measuring the optical fiber end face angle are implemented.

[0030] The optical fiber end face angle measurement method, system and storage medium provided by the present invention cleverly link the image recognition technology with the automation technology control, so as to achieve accurate capture and identification of the lateral end face profile of the measured optical fiber through a single camera, so as to provide virtual protractor technology, making the entire measurement process more visualized and intuitive. Compared with traditional edge detection and slope calculation algorithms that only output results but cannot intuitively evaluate or verify the accuracy of the results, the virtual protractor technology enables testers to more intuitively observe the actual bevel of the measured optical fiber end face and the position of the scale on the virtual protractor, thereby facilitating the testers to verify the authenticity and accuracy of the measurement results, providing observable verification conditions for the measurement results, thereby effectively improving the reliability of the optical fiber end face angle measurement results. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] The drawings constituting a part of this application are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0032] Figure 1 A schematic diagram of the steps of the optical fiber end face angle measurement method of the present invention;

[0033] Figure 2 Schematic diagrams of several side views of the fiber end face that may appear in the imaging area when the fiber is rotated;

[0034] Figure 3 It is a schematic diagram of a virtual protractor in the optical fiber end face angle measurement method of the present invention;

[0035] Figure 4 It is a schematic structural diagram of the optical fiber end face angle measurement system of the present invention.

[0036] Description of Reference Numerals

[0037] Visual recognition unit 1, optical fiber clamp 2, rotating device 3, optical fiber slot 4, optical fiber to be tested 5, light source unit 6, display screen 7, scale 8, camera 11, visual recognition module 12. DETAILED DESCRIPTION

[0038] In order to enable those skilled in the art to better understand the technical solution of the present invention, the specific technical solution of the present invention will be clearly and completely described in conjunction with the embodiments below to help those skilled in the art further understand the present invention. Obviously, the embodiments described in this case are only embodiments of a part of the present invention, not all of the embodiments. It should be pointed out that for those of ordinary skill in the art, the embodiments in this application and the features in the embodiments can be combined with each other without departing from the concept of the present invention and without conflicting with each other. Based on the embodiments in the present invention, all other embodiments obtained without creative work by those of ordinary skill in the art should belong to the disclosure and protection scope of the present invention.

[0039] In addition, the terms "first", "second", "S100", "S200", etc. in the specification, claims and drawings of the present invention are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the features used in this way can be interchanged where appropriate, so that the embodiments of the present invention described here can be implemented in an order other than those described here. At the same time, the stages recorded in each step are not mandatory to be implemented in the same step. It should be understood that the implementation order of the contents in each step stage can be adjusted and interchanged without violating the inventive concept, so that the step embodiments of the present invention described here can be implemented in an order other than those described here. In addition, the terms "including" and "having" and any of their variations in the present invention are intended to cover non-exclusive inclusions. Unless otherwise clearly specified and limited, the terms "set", "layout", "install", "connect", and "connect" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected, or indirectly connected through an intermediate medium, or it can be the internal connection of two components. For ordinary technicians in this field, the specific meanings of the above terms in this case can be understood according to specific circumstances and in combination with the prior art.

[0040] In order to use a single camera to accurately measure the fiber end face angle through dynamic fiber adjustment and image detection, such as Figures 1 to 3 As shown, the present invention provides a method for measuring the end face angle of an optical fiber, and its exemplary steps include:

[0041] Step S100 collects an image frame of the optical fiber end face with a bevel feature, constructs a teaching data set to train a visual recognition network, predicts the coordinate points at both ends of the bevel feature, obtains a recognition model, and establishes a control strategy for the recognition result and radial rotation of the optical fiber.

[0042] Step S200 places the detection end of the optical fiber to be tested in the imaging area, and radially rotates the optical fiber according to the control strategy until an image frame of the end face of the optical fiber to be tested containing a bevel edge feature is identified and acquired.

[0043] Step S300 constructs a virtual protractor, aligns the origin of the virtual protractor with the vertex of the bevel feature in the image frame of the optical fiber end face to be measured, so as to obtain the angle value of the bevel feature on the virtual protractor corresponding to the bevel feature.

[0044] Specifically, in traditional fiber end face angle measurement schemes, the optical fiber is usually in a stationary state, so multiple cameras need to be arranged at different sides of the optical fiber in order to find a suitable measurement angle. However, this method cannot accurately obtain the maximum angle value of the optical fiber end face, which will cause measurement errors in practical applications.

[0045] To this end, the present invention takes a different approach. Considering that the end face of the optical fiber is captured by a single camera, if the optical fiber is continuously rotated axially, lateral frames of different forms of the optical fiber can be captured (such as Figure 2 As shown), and by controlling the step amount of the rotation angle, a frame close to the maximum angle value of the fiber end face can be found, thereby improving the measurement accuracy. Therefore, the present invention conceives a dynamic design of the light to identify the frame close to the maximum angle value of the fiber end face through image recognition technology for subsequent virtual protractor measurement.

[0046] In order to use the existing image recognition technology to train the visual recognition network and build a recognition model, the example preferably uses the oblique edge feature (such as Figure 2 A) As the main identification feature, the reason is that the bevel is a direct reflection of the cutting angle of the fiber end face. Compared with other irregular features, the bevel is always present and clear, with high stability and consistency. Secondly, the bevel has geometric regularity, which can facilitate the rapid positioning of existing visual recognition algorithms, and can further realize the automatic alignment of the virtual protractor.

[0047] Therefore, in this example, in step S100, an optical fiber end face image frame with a bevel feature is collected to construct a teaching data set to train a visual recognition network and predict the coordinate points at both ends of the bevel feature. Example steps for obtaining a recognition model include:

[0048] Step S110 collects a large number of fiber end surface images with different slopes (refer to Figure 3 The feature A in the figure is marked, and the position of the hypotenuse feature line is marked to establish a teaching data set;

[0049] Step S120 constructs a pre-trained visual recognition network, such as ResNet50V2, to predict the four coordinate values ​​of the two end points of the hypotenuse. At the same time, transfer learning can be used to freeze the weights of the pre-trained model and add a customized top layer.

[0050] In addition, in a preferred example, early stopping and learning rate decay callback functions can be used to optimize the training process. After the training is completed, a recognition model can be obtained, and the model can directly output the coordinates of the two points on the hypotenuse for subsequent control of the rotation device to stop and align the virtual protractor.

[0051] The design of the recognition model also allows for flexible training, saving, and loading of models to process various types of fiber end face images, including different fiber types, lighting conditions, and backgrounds. It also supports feature detection and visualization in real-time video streams.

[0052] In one example, the camera can be set above the end face of the optical fiber, and the image of the rotating optical fiber end face can be collected in real time and transmitted to the visual recognition network for analysis. The camera is responsible for ensuring that the image is clear and has sufficient resolution for subsequent algorithm processing. For example: import two commonly used image processing libraries, cv2 and numpy, for image capture and processing, numerical processing; use the capture_frame method to capture a frame of image, and the preprocess_frame method to preprocess the captured frame (such as denoising and contrast enhancement); use the close method to release camera resources, and call this method when the camera is not needed, etc.

[0053] During the visual recognition process, the fiber end face image frames can be pre-processed by denoising, contrast enhancement, etc. The trained recognition model can be combined with the feature detection logic of the real-time image analysis algorithm to identify the bevel features of the fiber end face that meet the preset conditions, and mark the bevel and the endpoint aligned with the virtual protractor origin.

[0054] At this point, by defining a feature detection class and a main control loop, an automated fiber end face detection process can be implemented. For example, the detect_feature method is used to detect features in an image. If a valid feature is detected (the coordinates are within the image range), the rotation of the rotating device will be stopped, and the draw_feature method is used to draw the detected features on the image.

[0055] Further, in this example, in step S100, the example steps of establishing the control strategy of the identification result and the radial rotation of the optical fiber include:

[0056] Step S130 sets the axial rotation speed of the optical fiber to be tested to match the recognition frame rate of the visual recognition network.

[0057] Among them, in order to more accurately find the frame close to the maximum angle value of the optical fiber end face, the step amount of the optical fiber axial rotation can be adjusted. For example, a FiberFixture instance is established to rotate at a speed of 10° / second. Of course, it can also be adjusted according to the actual camera frame rate and the recognition frame rate of the recognition network. This example does not limit it.

[0058] Step S140 sets the radial rotation pre-start time of the optical fiber to be tested.

[0059] During the startup test, when the rotating device is started, a short acceleration process is required to reach a stable 10° / second. Setting a pre-start time of, for example, 5 seconds can ensure that the motor speed is completely stable and eliminate jitter and instability during startup. At the same time, after the camera is started, it also takes a certain amount of time to adjust the exposure and focus parameters. Therefore, these 5 seconds can also allow the system to reach a stable state, thereby obtaining a clearer image and reducing the probability of misrecognition.

[0060] Step S150 starts the rotation and after the pre-start time, when the visual recognition network recognizes the bevel feature of the end face of the optical fiber to be tested, stops the rotation of the optical fiber to be tested.

[0061] Specifically, when the optical fiber starts to rotate, the camera will continue to provide image frames to the visual recognition network until the recognition network identifies an image frame containing the bevel feature of the current optical fiber end face to be measured, and then outputs a hardware control command to stop the rotating device, so that the current optical fiber end face is stationary sideways facing the camera area, ready for virtual protractor measurement.

[0062] During this process, the control strategy sends control instructions to the control unit of the rotating device according to the feature recognition results to control the start and stop. At the same time, by reading the data of the rotary encoder in the rotating device, it is ensured that the motor of the rotating device stops completely at the correct position. In addition, after the motor stops, the control unit can continuously monitor the position and state to ensure that there is no additional rotation and drift. For example, a FiberFixture class is defined, and the necessary libraries: time and threading are imported to set the current position, target position, rotation state, and monitoring thread. The rotation state is set to True with the Start_rotation method, and the start rotation instruction is sent. If an "OK" response is received, the monitoring thread is started and whether it is successful is returned. The rotation state is set to False with the Stop_rotation method, and the stop rotation instruction is sent. If an "OK" response is received, wait for the monitoring thread to end and return whether it is successful. The _monitor_position method is used to continuously run in the rotating state, query the current position every 0.1 seconds, and update the current_position attribute.

[0063] In addition, it should be noted that setting the pre-start time is not a hard condition. In an optional implementation, the setting can be canceled according to the improvement of hardware conditions. Therefore, there is no limitation in this example, and technical personnel in this field can choose to implement it according to actual conditions.

[0064] Further, in this example, in step S300, the example steps of constructing a virtual protractor and aligning the origin of the virtual protractor with the vertex of the bevel feature in the image frame of the optical fiber end face to be measured include:

[0065] Step S310 compares the coordinate values ​​of the two end points of the bevel feature, determines the coordinates of the vertex of the bevel feature, and sets it as the origin of the virtual protractor;

[0066] For example, in step S200, the process of placing the optical fiber detection end to be tested in the camera area, that is, the process of aligning the optical fiber detection end with a preset reference coordinate system in the camera area, if the optical fiber is aligned parallel to the Y axis of the reference coordinate system, after such setting, the two endpoint coordinates (X1, Y1) and (X2, Y2) of the hypotenuse feature subsequently identified can be determined by comparing the size of the vertical coordinate values ​​of the two endpoints to determine the vertex coordinates of the hypotenuse feature.

[0067] Step S320 determines the 0° and 90° points of the virtual protractor at the origin according to the preset virtual protractor radius.

[0068] For example, the virtual protractor's horizontal coordinate is added with the preset virtual protractor radius value r to find the point at the 0° position (X0, Y0), and then the point at the 90° position (X90, Y90) is found by keeping the horizontal coordinate of the protractor origin unchanged and subtracting r from the vertical coordinate, and this point will be on the side straight line of the optical fiber.

[0069] Step S330 draws a semicircle according to the preset virtual protractor radius, positions the scale according to the obtained points to draw scale lines in equal parts, and adds scale labels to the scale lines.

[0070] For example, use matplotlib (a commonly used Python drawing library that can be used to draw various graphics, including the required scale lines and semicircles) to draw the semicircular part of the protractor, and the scale lines are evenly spaced between 0° and 180°. For each angle, the linspace function of numpy can be used to generate points on the arc, and then the cos and sin functions can be used to calculate the coordinates of these points. And use np.deg2rad(deg) to convert degrees to radians to facilitate subsequent trigonometric function calculations, and ax.text can be used to mark the corresponding angle values ​​at the end points of the scale lines.

[0071] In addition, in order to realize dynamic adjustment of the virtual protractor according to the size of the bevel feature, in another example, the example steps of constructing the virtual protractor further include:

[0072] Step S340: calculating the distance between the coordinate points at both ends of the bevel feature according to the image recognition result of the optical fiber to be tested, and obtaining the bevel feature length;

[0073] Step S350 dynamically adjusts the radius of the semicircle according to the characteristic length of the hypotenuse to scale the virtual protractor proportionally.

[0074] Specifically, dynamically adjusting the radius of the virtual protractor mainly involves updating the length of the scale lines and the distribution of the scale points when redrawing. When the radius is changed, that is, the end angle changes, the end point of the scale (including the 90° point) will be recalculated based on the new radius. For example, the ax.clear() method is used to clear the current figure, and the draw_protractor and draw_ticks methods are called to redraw the protractor and scale lines.

[0075] Further, in this example, in step S300, the step of obtaining the angle value on the virtual protractor corresponding to the bevel feature includes:

[0076] Step S360 identifies the overlapping scale lines of the virtual protractor where the hypotenuse feature is located, and reads the value of the corresponding scale label as the angle value.

[0077] In addition, to enhance the visual effect, the scale lines that coincide with the end face bevel and the 0° scale line can be marked in yellow in the virtual protractor, and the other scale lines can be kept in the default color (such as black), and the corresponding angle value can be marked at the angle between the two. Use ax.plot to color the scale lines. Use ax.text to mark the corresponding angle value at the end of the scale line. This facilitates visual verification by the inspector.

[0078] Further, corresponding to the example of dynamically adjusting the radius of the virtual protractor, in step S300, the example steps of obtaining the angle value on the virtual protractor corresponding to the bevel feature include:

[0079] Step S360' determines the coincident scale line of the virtual protractor where the bottom point of the hypotenuse feature is located, and reads the value of the corresponding scale label as the angle value.

[0080] On the other hand, corresponding to the above-mentioned example of the optical fiber end face angle measurement method, as Figure 4 As shown, the present invention also provides an optical fiber end face angle measuring device, which includes: a visual recognition unit, an optical fiber clamp, and a rotating device, wherein the optical fiber clamp is provided with optical fiber grooves corresponding to various specifications and sizes for axial positioning of the optical fiber to be measured, and allows the end face of the optical fiber to be measured to extend vertically from one end of the optical fiber clamp to the camera area of ​​the visual recognition unit to form a positioning.

[0081] The visual recognition unit includes a camera and a visual recognition module, such as Figure 4 As shown, in order to obtain a lateral image of the end face of the optical fiber to be tested, in this example, the camera example is set directly above the optical fiber, wherein the camera is responsible for collecting the lateral magnified image of the end face of the optical fiber to be tested and transmitting the image data to the visual recognition module. Of course, in other optional examples, the camera can also be set on the side of the optical fiber, and this embodiment does not limit it.

[0082] The rotating device is matched with the optical fiber to be tested, and the visual recognition unit is control-connected to the rotating device, so as to manipulate the rotating device to rotate the optical fiber to be tested according to the program steps of the optical fiber end face angle measurement method, until the visual recognition unit obtains the complete bevel edge feature of the optical fiber to be tested, and then the end face angle of the optical fiber to be tested is measured by a virtual protractor.

[0083] Specifically, in this example, the rotating device can be implemented by using known prior art to axially rotate the optical fiber to be tested under the control of the visual recognition module, so as to change the lateral position of the end face of the optical fiber to be tested in the camera imaging area, so that the camera can capture the frame of the side change of the optical fiber to be tested, and identify the frame corresponding to the complete bevel feature of the optical fiber to be tested in these frame images through the visual recognition module (such as Figure 2 As shown in Figure A), stop the rotation of the rotating device, keep the end face of the optical fiber to be measured still, and use the virtual protractor program to complete the measurement of the angle of the end face of the optical fiber to be measured.

[0084] Furthermore, in order to improve the recognition rate of the visual recognition unit and avoid the reduction of the recognition rate due to changes in the ambient light source, in an optional embodiment, the optical fiber end face angle measuring device also includes: a light source unit, wherein the light source unit is preferably a ring light, and its light emitting side faces the optical fiber to be tested in the camera area, so as to provide a stable light source environment for the camera area and create a shadow-free condition for the optical fiber to be tested as much as possible, thereby reducing the probability of misjudgment of the visual recognition unit and thereby improving the visual recognition rate.

[0085] Furthermore, in order to facilitate the test personnel to verify the authenticity and accuracy of the measurement results and provide observable verification conditions for the measurement results, in an optional embodiment, the optical fiber end face angle measurement device also includes: a display screen, wherein the display screen is connected to the visual recognition unit, showing the relative measurement posture of the optical fiber end face to be tested and the virtual protractor, so that the test personnel can observe the complete bevel characteristics of the optical fiber to be tested and the corresponding scale on the virtual protractor with the naked eye.

[0086] Furthermore, in order to locate the position of the optical fiber end face to be measured within the imaging area to facilitate subsequent virtual protractor measurement, in an optional embodiment, the optical fiber end face angle measuring device also includes: a ruler, wherein the ruler is arranged at one end of the optical fiber clamp to guide and control the protruding length of the end face of the optical fiber to be measured until it is positioned within the imaging area of ​​the visual recognition unit, so that the origin of the subsequent virtual protractor can be accurately aligned with the vertex of the bevel feature in the image frame of the optical fiber end face to be measured, so as to accurately obtain the angle value on the virtual protractor corresponding to the bevel feature.

[0087] On the other hand, the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, wherein when the computer program is executed, the steps of any of the above-mentioned methods for measuring the optical fiber end face angle are implemented.

[0088] In summary, the optical fiber end face angle measurement method, system and storage medium provided by the present invention cleverly link the image recognition technology with the automation technology control, so as to realize the accurate capture and identification of the lateral end face profile of the measured optical fiber through a single camera, so as to use the virtual protractor technology, so that the entire measurement process is more visualized and intuitive. Compared with traditional edge detection and slope calculation algorithms that only output results but cannot intuitively evaluate or verify the accuracy of the results, the virtual protractor technology can enable testers to more intuitively observe the actual bevel of the measured optical fiber end face and the position of the scale on the virtual protractor, thereby facilitating the testers to verify the authenticity and accuracy of the measurement results, providing observable verification conditions for the measurement results, thereby effectively improving the reliability of the optical fiber end face angle measurement results.

[0089] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is only limited by the claims and their full scope and equivalents. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

[0090] Those skilled in the art can understand that, in addition to implementing the system, device, unit and its various modules provided by the present invention in a purely computer-readable program code, it is entirely possible to implement the same program in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers and embedded microcontrollers by logically programming the method steps. Therefore, the system, device and its various modules provided by the present invention can be considered as a hardware component, and the modules included therein for implementing various programs can also be regarded as structures within the hardware component; the modules for implementing various functions can also be regarded as both software programs for implementing the method and structures within the hardware component.

[0091] In addition, all or part of the steps in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a program, and the program is stored in a storage medium, including a number of instructions to enable a single-chip microcomputer, a chip or a processor to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, RandomAccess Memory), disk or optical disk and other media that can store program codes.

[0092] In addition, various implementation modes of the embodiments of the present invention may be arbitrarily combined, and as long as they do not violate the concept of the embodiments of the present invention, they should also be regarded as the contents disclosed by the embodiments of the present invention.

Claims

1. A method for measuring an optical fiber end face angle, comprising the steps of: Collect the image frames of the fiber end face with bevel features, build a teaching data set to train the visual recognition network, predict the coordinate points at both ends of the bevel features, obtain the recognition model, and establish the control strategy of the recognition results and the radial rotation of the fiber; The detection end of the optical fiber to be tested is placed in the imaging area, and the optical fiber is radially rotated according to the control strategy until an image frame of the end face of the optical fiber to be tested containing the bevel edge feature is identified and obtained; A virtual protractor is constructed, and the origin of the virtual protractor is aligned with the vertex of the bevel feature in the image frame of the optical fiber end face to be measured to obtain the angle value of the bevel feature on the virtual protractor corresponding to the bevel feature.

2. The optical fiber end face angle measurement method according to claim 1, wherein the step of establishing the identification result and the control strategy of the optical fiber radial rotation comprises: Set the axial rotation speed of the optical fiber to be tested to match the recognition frame rate of the visual recognition network; When the visual recognition network recognizes the bevel feature of the end face of the optical fiber to be tested, the rotation of the optical fiber to be tested is stopped.

3. The optical fiber end face angle measurement method according to claim 1, wherein the step of establishing the identification result and the control strategy of the optical fiber radial rotation comprises: Set the axial rotation speed of the optical fiber to be tested to match the recognition frame rate of the visual recognition network; Set the radial rotation pre-start time of the optical fiber to be tested; The rotation is started and after the pre-start time, when the visual recognition network recognizes the bevel edge feature of the end face of the optical fiber to be tested, the rotation of the optical fiber to be tested is stopped.

4. The method for measuring the optical fiber end face angle according to claim 1, wherein the step of constructing a virtual protractor and aligning the origin of the virtual protractor with the vertex of the bevel feature in the image frame of the optical fiber end face to be measured comprises: Compare the coordinate values ​​of the two end points of the bevel feature, determine the coordinates of the vertex of the bevel feature, and set them as the origin of the virtual protractor; According to the preset virtual protractor radius, the 0° and 90° points of the virtual protractor are determined at the origin respectively; According to the preset virtual protractor radius, draw a semicircle, and according to the obtained point position, position the scale to draw scale lines in equal parts, and add scale labels to the scale lines.

5. The method for measuring the optical fiber end face angle according to claim 4, wherein the steps further include: According to the image recognition result of the optical fiber to be tested, the distance between the coordinate points at both ends of the bevel feature is calculated to obtain the bevel feature length; According to the characteristic length of the hypotenuse, the radius of the virtual protractor is dynamically adjusted to scale the virtual protractor proportionally.

6. The optical fiber end face angle measurement method according to claim 5, wherein the step of obtaining the angle value on the virtual protractor corresponding to the bevel feature comprises: Determine the coincident scale line of the virtual protractor at which the bottom point of the hypotenuse feature is located, and read the value of the corresponding scale label as the angle value.

7. The optical fiber end face angle measurement method according to claim 4, wherein the step of obtaining the angle value on the virtual protractor corresponding to the bevel feature comprises: Identify the coincident scale lines of the virtual protractor where the hypotenuse feature is located, and read the value of the corresponding scale label as the angle value.

8. An optical fiber end face angle measuring device, comprising: A visual recognition unit, a fiber optic clamp, and a rotating device, wherein the fiber optic clamp is provided with a fiber optic groove for axially positioning the fiber to be tested and allowing the end face of the fiber to be tested to extend from one end of the fiber optic clamp to the camera area of ​​the visual recognition unit, the rotating device is matched with the fiber to be tested, and the visual recognition unit is control-connected to the rotating device, and the rotating device is controlled to rotate the fiber to be tested according to the steps of the fiber end face angle measurement method according to any one of claims 1 to 7 until the visual recognition unit obtains the complete bevel edge feature of the fiber to be tested, and the end face angle of the fiber to be tested is measured with a virtual protractor.

9. The optical fiber end face angle measuring device according to claim 8, further comprising: A light source unit, a display screen, and a scale, wherein the scale is arranged at one end of the optical fiber clamp to guide the optical fiber to be tested to extend to a position within the imaging area of ​​the visual recognition unit, the light emitting side of the light source unit faces the optical fiber to be tested within the imaging area, and the display screen is connected to the visual recognition unit to show the relative measurement posture of the end face of the optical fiber to be tested and the virtual protractor.

10. A computer-readable storage medium storing a computer program, wherein when the computer program is executed, the steps of the optical fiber end face angle measurement method according to any one of claims 1 to 7 are implemented.

Images