Self-feeding optical fiber damage detection method based on optical contrast edge detection algorithm

Abstract

The invention provides a self-feeding optical fiber damage detection method based on an optical contrast edge detection algorithm. The self-feeding optical fiber damage detection method comprises the steps that an automatic feeding mechanism is used for conveying an optical fiber to be detected to an optical fiber detection area; in the optical fiber detection area, acquiring a contrast image of the optical fiber to be detected in real time by using a microscopic imaging system; and processing the optical contrast image by using a specific edge detection algorithm, and identifying a damaged area of the to-be-detected optical fiber. The self-feeding optical fiber damage detection method provided by the invention has the advantages of automation, high efficiency and capability of accurately determining the damage position.

Description

Technical Field

[0001] The present invention relates to the fields of optical fiber damage detection and image processing, and more specifically, to a self-feeding optical fiber damage detection method based on an optical contrast edge detection algorithm. Background Art

[0002] From the perspective of cross section, the optical fiber generally includes a core layer, a cladding layer, and a coating layer from the inside to the outside. During use, due to bending, scratching, local impact, etc., each layer of the optical fiber may be damaged, resulting in consequences of varying degrees. Damage to the core layer and cladding will directly affect the transmission performance of the optical fiber, causing unexpected loss of the transmitted light at the damaged location and local heating of the optical fiber, thereby reducing the overall performance and transmission efficiency of the optical fiber, and easily causing the burning of optical fiber connectors. Although obvious damage to the coating or slight damage to other parts will not significantly affect the transmission performance of the optical fiber in a short period of time, it will reduce the reliability of the optical fiber for a long time, becoming a quality risk in the manufacturing process of optical fiber amplifiers, optical fiber lasers, etc., and also has a great impact on the life and reliability of optical fiber equipment.

[0003] At present, optical fiber damage detection generally adopts methods such as red light pen flaw detection, optical fiber transmission performance detection and / or manual observation. For example, red light pen flaw detection uses red laser to inject into the optical fiber core layer, and the red light will be scattered from the optical fiber to form bright spots at the parts where the core layer or cladding is obviously damaged. Optical fiber transmission performance detection uses a certain characteristic of the transmission light in the optical fiber, such as optical power, light polarization, etc., to test whether the transmission optical fiber is normal. Manual observation uses the naked eye or tools such as microscopes and magnifying glasses to observe the end face or the whole of the optical fiber. However, for slight damage to the optical fiber and coating damage, as well as other damage that does not change the detection performance of the optical fiber, red light pen flaw detection and optical fiber transmission performance detection methods cannot meet the detection requirements. In addition, the observation method is mainly completed manually, its efficiency is limited, and the observation itself has uncertainties that vary from person to person. Therefore, an effective optical fiber damage detection method is needed to reliably and efficiently detect the optical fiber and determine the specific location of the damage so that relevant technicians can take targeted measures. Summary of the invention

[0004] In view of the above problems in the prior art, the present invention provides a self-feeding optical fiber damage detection method based on an optical contrast edge detection algorithm, which has the advantages of automation, high efficiency, and the ability to accurately determine the damage location.

[0005] One aspect of the present invention provides a self-feeding optical fiber damage detection method based on an optical contrast edge detection algorithm, comprising: using an automatic feeding mechanism to deliver the optical fiber to be inspected to an optical fiber detection area; in the optical fiber detection area, using a microscopic imaging system to collect a contrast image of the optical fiber to be inspected in real time; using a specific edge detection algorithm to process the optical contrast image to identify the damaged area of ​​the optical fiber to be inspected.

[0006] In some exemplary embodiments of the present invention, the automatic feeding mechanism includes: a rotatable feeding fiber optic reel, used to transport the optical fiber to be inspected wound on the circumference of the feeding fiber optic reel to the optical fiber detection area; a rotatable receiving fiber optic reel, used to wind and recycle the inspected optical fiber, the receiving fiber optic reel and the feeding fiber optic reel have the same rotation direction and rotation speed, and the receiving fiber optic reel and the feeding fiber optic reel are located in the same rotation plane; a synchronous belt, the two ends of the synchronous belt are respectively connected to the rotating shafts of the feeding fiber optic reel and the receiving fiber optic reel, used to drive the feeding fiber optic reel and the receiving fiber optic reel to rotate synchronously and in the same direction; wherein the optical fiber detection area is arranged in parallel and directly above the synchronous belt, and maintains a specific distance from the synchronous belt.

[0007] In some exemplary embodiments of the present invention, the microscopic imaging system is arranged on a side of the optical fiber detection area away from the automatic feeding mechanism, and includes: a CCD camera, used to collect the contrast image of the optical fiber to be tested in the optical fiber detection area in real time, the imaging direction of the CCD camera is perpendicular to the feeding direction of the optical fiber to be tested, and the resolution of the collected image is greater than 5μm / pix; a white light source, used to provide illumination for the optical fiber to be tested in the optical fiber detection area, the light emitting direction of the white light source is perpendicular to the feeding direction of the optical fiber to be tested; wherein the center line of the illumination area of ​​the white light source coincides with the center line of the imaging area of ​​the CCD camera, is perpendicular to the axis of the optical fiber to be tested in the optical fiber detection area, and is in the same plane.

[0008] In some exemplary embodiments of the present invention, a specific edge detection algorithm is used to process the optical contrast image, including: performing grayscale processing on the optical contrast image to obtain a grayscale image of the optical contrast image; gradually performing filtering processing, gradient processing and binarization processing on the grayscale image to sequentially obtain filtering results, gradient results and edge information of the grayscale image, and establishing a connected domain of the grayscale image based on the edge information; and, searching the continuity of the connected domains one by one, judging the damage position of the optical fiber to be inspected based on the continuity, and obtaining the coordinate information corresponding to the damage position through reverse retrieval.

[0009] In some exemplary embodiments of the present invention, the filtering process includes: setting a first coordinate direction and a second coordinate direction perpendicular to each other; setting a first threshold, and filtering the original coordinates of the grayscale image according to the first threshold to obtain a filtering result of the grayscale image:

[0010]

[0011] Wherein, x is the first coordinate direction, y is the second coordinate direction, T1 is the first threshold, G0(x, y) is the original coordinate of the grayscale image, and G1(x, y) is the filtering result of the grayscale image.

[0012] In some exemplary embodiments of the present invention, the gradient processing includes: performing gradient processing on the filtering result of the grayscale image in the second coordinate direction to obtain a gradient processing result of the grayscale image:

[0013]

[0014] Wherein, x is the first coordinate direction, y is the second coordinate direction, G1(x, y) is the filtering result of the grayscale image, and G2(x, y) is the gradient result of the grayscale image.

[0015] In some exemplary embodiments of the present invention, the binarization process includes: setting a second threshold, and binarizing the gradient result of the grayscale image according to the second threshold to obtain edge information of the grayscale image:

[0016]

[0017] Wherein, T2 is the second threshold, x is the first coordinate direction, y is the second coordinate direction, G2(x, y) is the gradient result of the grayscale image, and G3(x, y) is the edge information of the grayscale image.

[0018] In some exemplary embodiments of the present invention, the first threshold is determined by the light intensity in the optical fiber detection area and is used to filter out the background noise of the grayscale image.

[0019] In some exemplary embodiments of the present invention, the second threshold is determined by a gradient value in a second coordinate direction corresponding to the light intensity in the optical fiber detection area, and is used to filter out pixels other than edges and cracks in the grayscale image.

[0020] In some exemplary embodiments of the present invention, after identifying the damaged area of ​​the optical fiber to be inspected, the method further includes: marking the damaged area and saving and outputting the damage information of the optical fiber to be inspected; and returning to the operation of using the automatic feeding mechanism to transport the optical fiber to be inspected to the optical fiber detection area for the next section of the optical fiber to be inspected.

[0021] Compared with traditional optical fiber detection methods, the present invention can equally identify both obvious damage and minor damage through a unified edge recognition algorithm, and automatically mark them through multimedia data recording, with high detection efficiency, overcoming the uncertainty of manual detection. The present invention adopts the design of automatic lossless feeding, which can realize automatic and continuous detection of optical fibers and has the scalability of access to automated production lines. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 A self-feeding optical fiber damage detection method based on an optical contrast edge detection algorithm according to an embodiment of the present invention is schematically shown;

[0023] Figure 2 An automatic feeding mechanism according to an embodiment of the present invention is schematically shown;

[0024] Figure 3 A microscopic imaging system according to an embodiment of the present invention is schematically shown;

[0025] Figure 4 The edge detection algorithm process according to an embodiment of the present invention is schematically shown;

[0026] Figure 5 The effect diagram of detecting optical fiber damage based on the optical contrast edge detection algorithm according to an embodiment of the present invention is schematically shown. DETAILED DESCRIPTION

[0027] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments and with reference to the accompanying drawings.

[0028] Figure 1 A self-feeding optical fiber damage detection method based on an optical contrast edge detection algorithm according to an embodiment of the present invention is schematically illustrated.

[0029] like Figure 1 As shown, in this embodiment, a self-feeding optical fiber damage detection method based on an optical contrast edge detection algorithm includes:

[0030] S1, using the automatic feeding mechanism 1 to transport the optical fiber 3 to be tested to the optical fiber testing area;

[0031] S2, in the optical fiber detection area, using the microscopic imaging system 2 to collect the contrast image of the optical fiber 3 to be detected in real time;

[0032] S3, using a specific edge detection algorithm to process the optical contrast image, and identifying the damaged area of ​​the optical fiber 3 to be inspected.

[0033] In some other embodiments, after the damaged area of ​​the optical fiber to be inspected is identified, the method further includes:

[0034] S4, marking the damaged area and saving and outputting the damage information of the optical fiber 3 to be inspected;

[0035] S5, for the next section of optical fiber to be inspected, the process returns to the operation of using the automatic feeding mechanism 1 to transport the optical fiber to be inspected to the optical fiber inspection area.

[0036] Figure 2 An automatic feeding mechanism according to an embodiment of the present invention is schematically shown.

[0037] like Figure 2 As shown, in this embodiment, the automatic feeding mechanism 1 includes a rotatable feeding fiber tray 11, a rotatable receiving fiber tray 12 and a synchronous belt 13. The present invention uses the synchronously rotating fiber feeding tray 11 and the fiber receiving tray 12 to perform stress-free feeding of the optical fiber 3 to be inspected, thereby solving the problem of damage-free automatic feeding. In addition, the design of automatic lossless feeding enables the present invention to realize unattended automatic continuous detection of optical fibers, and has the scalability of access to automated production lines.

[0038] Specifically, the rotatable feeding fiber optic disc 11 is used to transport the optical fiber 3 to be inspected, which is wound on the circumference of the feeding fiber optic disc 11, to the optical fiber inspection area; the rotatable receiving fiber optic disc 12 is used to wind and recycle the inspected optical fiber. The receiving fiber optic disc 12 has the same rotation direction and rotation speed as the feeding fiber optic disc 11, and the receiving fiber optic disc 12 and the feeding fiber optic disc 11 are located in the same rotation plane. The two ends of the synchronous belt 13 are respectively connected to the rotating shafts of the feeding fiber optic disc 11 and the receiving fiber optic disc 12, and are used to drive the feeding fiber optic disc 11 and the receiving fiber optic disc 12 to rotate synchronously and in the same direction. It should be noted that the transmission method of the feeding fiber optic disc 11 and the receiving fiber optic disc 12 in the present invention is not only the synchronous belt 13, and other transmission methods that can realize the synchronous and same-direction rotation of the feeding fiber optic disc 11 and the receiving fiber optic disc 12, such as chain transmission, should also be within the protection scope of the present invention. There are many driving methods for the automatic feeding mechanism 1, and the present invention does not make specific limitations on this.

[0039] Please continue to refer to Figure 2 A fiber detection area is arranged parallel to and directly above the synchronous belt 13, and a specific spacing is maintained between the synchronous belt 13 and the optical fiber detection area. The specific spacing is determined according to the actual situation. A reasonable spacing is convenient for the detection operation, and can make the angle between the optical fiber 3 to be detected and the synchronous belt 13 as small as possible during the conveying process, so as to avoid excessive bending of the optical fiber 3 to be detected from the feeding optical fiber reel 11 to the receiving optical fiber reel 12.

[0040] Figure 3 A microscopic imaging system according to an embodiment of the present invention is schematically shown.

[0041] like Figure 3 As shown, in this embodiment, Figure 3 The left side is a cross section of the optical fiber 3 to be inspected in the optical fiber inspection area, including the coating layer, cladding layer, and core layer. The microscopic imaging system 2 is located at Figure 3The right side is specifically arranged on the side of the optical fiber detection area away from the automatic feeding mechanism 1, and includes: a CCD camera 21 and a white light source 22. The CCD camera 21 is used to collect the contrast image of the optical fiber 3 to be tested in the optical fiber detection area in real time. Contrast refers to the difference in brightness between different areas on the image. In order to ensure that the CCD camera 21 can accurately and clearly collect the contrast images of each layer of the optical fiber 3 to be tested, the outer diameter of the entire optical fiber 3 to be tested is located within the range of the microscopic imaging system 2, and the imaging direction of the CCD camera 21 is required to be perpendicular to the feeding direction of the optical fiber 3 to be tested, and the resolution of the collected image is greater than 5μm / pix.

[0042] The white light source 22 is used to provide illumination for the optical fiber 3 to be inspected in the optical fiber detection area. Under a black background, when the optical fiber is irradiated with strong light, different layers will show different contrasts, and its inner layer structure can be clearly recorded by the microscope system, thereby solving the problem of optical fiber image acquisition. In order for the CCD camera 21 to clearly capture the structure of each layer of the optical fiber 3 to be inspected, the light emitting direction of the white light source 22 is required to be perpendicular to the feeding direction of the optical fiber to be inspected, and the center line of the illumination area of ​​the white light source 22 and the center line of the imaging area of ​​the CCD camera 21 coincide with each other, and are perpendicular to the axis of the optical fiber 3 to be inspected in the optical fiber detection area, and are in the same plane. The illumination intensity of the white light source 22 needs to be considered and selected according to the actual situation, and the present invention does not limit it here.

[0043] Figure 4 The edge detection algorithm process according to an embodiment of the present invention is schematically shown. Figure 5 The effect diagram of detecting optical fiber damage based on the optical contrast edge detection algorithm according to an embodiment of the present invention is schematically shown.

[0044] First, please refer to Figure 4 .like Figure 4 As shown, in this embodiment, a specific edge detection algorithm is used to process the light contrast image, including:

[0045] S31, performing grayscale processing on the light contrast image to obtain a grayscale image of the light contrast image;

[0046] S32, gradually performing filtering processing, gradient processing and binarization processing on the grayscale image, sequentially obtaining filtering results, gradient results and edge information of the grayscale image, and establishing a connected domain of the grayscale image according to the edge information;

[0047] S33, searching the continuity of the connected domains one by one, determining the damage position of the optical fiber 3 to be inspected according to the continuity, and obtaining the coordinate information corresponding to the damage position through reverse search.

[0048] Further, in S31, after the CCD camera 21 acquires the contrast image of the optical fiber 3 to be inspected, it is output to a processing unit external to the CCD camera 21. The processing unit preprocesses the color contrast image and converts the image into a grayscale image. The processing unit may be an application configured on a computer, and the present invention does not limit this.

[0049] Now Figure 4 Based on Figure 5 , the process of edge detection algorithm processing light contrast image is explained in detail.

[0050] like Figure 5 As shown in FIG. 1 , there are 6 sub-figures (a) to (f), in which a rectangle represents a cross-sectional view of a section of the optical fiber 3 to be tested along the axial direction, and a small gap in the rectangle represents a certain degree of damage to the optical fiber 3 to be tested. Figure 5 After the grayscale image shown in (a) is filtered, the purpose is to filter out the background noise of the grayscale image and retain the main image information. Set the first coordinate direction x and the second coordinate direction y perpendicular to each other; set the first threshold T1, and filter the original coordinates G0 (x, y) of the grayscale image according to the first threshold T1 to obtain the filtering result G1 (x, y) of the grayscale image, as shown in Figure 5 (b) as shown:

[0051]

[0052] Wherein, x is the first coordinate direction, y is the second coordinate direction, T1 is the first threshold, G0(x, y) is the original coordinate of the grayscale image, and G1(x, y) is the filtering result of the grayscale image.

[0053] Next, the grayscale image filtering result G1(x, y) is gradient-processed in the second coordinate direction y to obtain the grayscale image gradient result G2(x, y), as shown in FIG. Figure 5 (c) as shown:

[0054]

[0055] Wherein, x is the first coordinate direction, y is the second coordinate direction, G1(x, y) is the filtering result of the grayscale image, and G2(x, y) is the gradient result of the grayscale image.

[0056] Then, a second threshold T2 is set, and the gradient result G2(x, y) of the grayscale image is binarized according to the second threshold T2 to obtain the edge information G3(x, y) of the grayscale image, as shown in FIG. Figure 5 (d) as shown:

[0057]

[0058] Wherein, T2 is the second threshold, x is the first coordinate direction, y is the second coordinate direction, G2(x, y) is the gradient result of the grayscale image, and G3(x, y) is the edge information of the grayscale image.

[0059] Then, the connected domain G4(x, y) of the grayscale image is established based on the edge information G3(x, y), as shown in Figure 5 As shown in (e), consecutive different numbers 1, 2, 3, 4, 5, 6 are used to represent the connectivity of some edge information.

[0060] Finally, by searching the continuity of the connected domain G4 (x, y) one by one along the first coordinate direction x, the breakpoint in the optical fiber 3 to be inspected is found, and its corresponding coordinate position is obtained by reverse search, such as Figure 5 As shown in (f), it can be seen that among the layers of numbers, numbers 1, 2, and 3 are continuous, numbers 4, 5, and 6 are discontinuous, and the discontinuous number 5 is marked with a circle, indicating that a breakpoint occurs here, and this position corresponds to a certain joint between the cladding and the coating layer of the optical fiber 3 to be tested.

[0061] Please continue to refer to Figure 5 In this embodiment, the setting of the first threshold value T1 and the second threshold value T2 directly affects the effect of the edge detection algorithm in the present invention on the light contrast image processing, such as accuracy.

[0062] To ensure that the image is not over-filtered when the grayscale value population of the image is relatively concentrated under low light intensity, a first threshold T1 and a second threshold T2 are set respectively, as described in detail below.

[0063] In image processing, different light intensities correspond to different image contrasts and image background noises. The purpose of setting the first threshold T1 is to filter out unnecessary background noise in the image while retaining the main image information. Therefore, the first threshold T1 is mainly determined by the light intensity in the optical fiber detection area.

[0064] Furthermore, in this embodiment, the maximum illumination intensity that the white light source 22 can achieve is set to Lux_max, and the illumination intensity during actual operation is Lux_real. The grayscale statistics of the pixel points of the grayscale image filtering result G1 (x, y) are performed to obtain the population numbers corresponding to different grayscale values. The population numbers are accumulated from low grayscale to high grayscale. When the accumulated value exceeds (Lux_real) / (Lux_max)×20% of the total population number for the first time, the grayscale value corresponding to the accumulated population number is the first threshold value T1 corresponding to the threshold filter.

[0065] At the same time, in image processing, different illuminations correspond to different image contrasts and image gradients. The purpose of setting the first threshold T2 is to filter out pixels outside areas with large image contrasts such as edges and cracks as much as possible, so as to facilitate subsequent edge and crack recognition processing.

[0066] Furthermore, in this embodiment, the maximum illumination intensity that the white light source 22 can achieve is set to Lux_max, and the illumination intensity during actual operation is Lux_real. The gradient statistics of the gradient result G2 (x, y) of the grayscale image are performed to obtain the population corresponding to different gradient values. The population is accumulated from low gradient to high gradient. When the accumulated value exceeds (Lux_real) / (Lux_max)×50% of the total population for the first time, the gradient value corresponding to the accumulated population is the second threshold T2 corresponding to the threshold filter.

[0067] In summary, firstly, the present invention uses the synchronously rotating feeding fiber disk and the fiber receiving disk to perform stress-free feeding of the optical fiber to be inspected, thus solving the problem of damage-free automatic feeding. Then, through the microscopic imaging system, the contrast image of each section of the optical fiber to be inspected is collected, and its inner layer structure is clearly recorded, thus solving the problem of optical fiber image acquisition. Finally, based on the edge detection algorithm, the edge of each section of the optical fiber image to be inspected is identified, and the images of each layer of the optical fiber to be inspected in the image are segmented and taken out, and then the damage is identified according to the morphology of the edge, thus solving the problem of identifying the damaged optical fiber image. Compared with traditional optical fiber detection methods, the present invention is automated, efficient, and can accurately locate the damage position.

[0068] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A self-feeding optical fiber damage detection method based on optical contrast edge detection algorithm, It is characterized in that include: Using an automatic feeding mechanism (1) to transport the optical fiber to be tested (3) to the optical fiber testing area; In the optical fiber detection area, a microscopic imaging system (2) is used to collect a contrast image of the optical fiber to be detected (3) in real time; The optical contrast image is processed using a specific edge detection algorithm to identify the damaged area of ​​the optical fiber (3) to be inspected.

2. The self-feeding optical fiber damage detection method according to claim 1, It is characterized in that The automatic feeding mechanism (1) comprises: A rotatable fiber feeding disc (11) for conveying the optical fiber to be inspected (3) wound on the circumference of the fiber feeding disc (11) to the optical fiber inspection area; A rotatable receiving optical fiber disc (12) is used to coil and recycle the inspected optical fiber, wherein the receiving optical fiber disc (12) and the feeding optical fiber disc (11) have the same rotation direction and rotation speed, and the receiving optical fiber disc (12) and the feeding optical fiber disc (11) are located in the same rotation plane; A synchronous belt (13), wherein both ends of the synchronous belt (13) are respectively connected to the rotating shafts of the feeding optical fiber disc (11) and the receiving optical fiber disc (12), and are used to drive the feeding optical fiber disc (11) and the receiving optical fiber disc (12) to rotate synchronously and in the same direction; wherein: The optical fiber detection area is arranged in parallel just above the synchronous belt (13) and maintains a specific distance from the synchronous belt (13).

3. The self-feeding optical fiber damage detection method according to claim 1, It is characterized in that The microscopic imaging system (2) is arranged on a side of the optical fiber detection area away from the automatic feeding mechanism (1), and comprises: A CCD camera (21) is used to collect a contrast image of the optical fiber to be tested (3) in the optical fiber detection area in real time, the imaging direction of the CCD camera (21) is perpendicular to the feeding direction of the optical fiber to be tested (3), and the resolution of the collected image is greater than 5 μm / pix; A white light source (22) is used to provide illumination for the optical fiber (3) to be inspected in the optical fiber inspection area, and the light emitting direction of the white light source (22) is perpendicular to the feeding direction of the optical fiber (3) to be inspected; wherein, The center line of the illumination area of ​​the white light source (21) coincides with the center line of the imaging area of ​​the CCD camera (22), is perpendicular to the axis of the optical fiber to be tested in the optical fiber detection area, and is located in the same plane.

4. The self-feeding optical fiber damage detection method according to claim 1, It is characterized in that The light contrast image is processed using a specific edge detection algorithm, including: Performing grayscale processing on the light contrast image to obtain a grayscale image of the light contrast image; Step by step, filtering, gradient processing and binarization processing are performed on the grayscale image to sequentially obtain filtering results, gradient results and edge information of the grayscale image, and establish a connected domain of the grayscale image according to the edge information; The continuity of the connected domains is retrieved one by one, the damaged position of the optical fiber (3) to be inspected is determined according to the continuity, and the coordinate information corresponding to the damaged position is obtained through reverse retrieval.

5. The self-feeding optical fiber damage detection method according to claim 4, It is characterized in that The filtering process includes: Setting a first coordinate direction and a second coordinate direction perpendicular to each other; A first threshold is set, and the original coordinates of the grayscale image are filtered according to the first threshold to obtain a filtering result of the grayscale image: Among them, x is the first coordinate direction, y is the second coordinate direction, T1 is the first threshold, G0(x, y) is the original coordinate of the grayscale image, and G1(x, y) is the filtering result of the grayscale image.

6. The self-feeding optical fiber damage detection method according to claim 5, It is characterized in that Gradient processing includes: Performing gradient processing on the filtering result of the grayscale image in the second coordinate direction to obtain a gradient processing result of the grayscale image: Among them, x is the first coordinate direction, y is the second coordinate direction, G1(x, y) is the filtering result of the grayscale image, and G2(x, y) is the gradient result of the grayscale image.

7. The self-feeding optical fiber damage detection method according to claim 6, It is characterized in that Binarization processing includes: A second threshold is set, and the gradient result of the grayscale image is binarized according to the second threshold to obtain edge information of the grayscale image: Wherein, T2 is the second threshold, x is the first coordinate direction, y is the second coordinate direction, G2(x, y) is the gradient result of the grayscale image, and G3(x, y) is the edge information of the grayscale image.

8. The self-feeding optical fiber damage detection method according to claim 5, wherein the first threshold is determined by the light intensity in the optical fiber detection area and is used to filter out the background noise of the grayscale image.

9. According to the self-feeding optical fiber damage detection method of claim 7, the second threshold is determined by the gradient value in the second coordinate direction corresponding to the light intensity in the optical fiber detection area, and is used to filter out pixels other than edges and cracks in the grayscale image.

10. The self-feeding optical fiber damage detection method according to claim 1, It is characterized in that After the damaged area of ​​the optical fiber (3) to be inspected is identified, the method further comprises: Marking the damaged area, and saving and outputting damage information of the optical fiber (3) to be inspected; For the next section of optical fiber to be inspected, the process returns to the operation of using the automatic feeding mechanism (1) to transport the optical fiber to be inspected (3) to the optical fiber inspection area.

Images