A method for detecting spot defects of an airborne optical fiber end face, a storage medium and an equipment

Abstract

The application discloses a kind of spot defect detection methods of airborne optical fiber end face, first to optical fiber image acquisition is carried out, then grey levelization obtains image Img1, then binaryzation is carried out.Edge detection is carried out to image, the outer edge contour of optical fiber is extracted, and the outer edge contour extracted is used as ROI region of interest, and the grey image in the region is operated with mask, and image Img2 is obtained.Automatic threshold binaryzation processing is carried out to image Img2, and image Img2 is carried out binaryzation segmentation;Automatic extraction is carried out to highlight spot in image Img2, to detect the defect area S' of highlight defect part of optical fiber.The ratio F of defect area S' and the area S of image Img2 is used as defect rate.The application can accurately extract highlight defect part, realizes automatic defect detection, with the advantages of high efficiency, simple, stable and reliable.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of optical fiber end face defect detection, and particularly relates to a method for detecting spot defects of an airborne optical fiber end face, a storage medium and equipment. BACKGROUND

[0002] Existing measurement and detection of optical fibers are mostly positioned on the position, sealing property, transmission performance, flow rate and the like of the optical fiber, and the detection methods for the end face of the optical fiber are mostly general, and mostly a device, apparatus and the like, and the structure and method thereof are relatively complex, errors are prone to occur, the precision is not enough, and the automation degree is not high, and the efficiency is also low. In the production process of the end face of the optical fiber, due to the deficiency of the process flow, spot defects prone to occur on the end face of the optical fiber are difficult to detect manually, the method used for detecting the end face of the optical fiber has a low automation degree, and has insufficient adaptability and the like.

[0003] The existing patent CN106197946B discloses a method, device and system for optical fiber detection. The waveform graph generated by the optical time domain reflectometer for detecting the optical fiber link is analyzed and processed to determine the health state of the optical fiber. However, the end face state is not detected. The existing patent CN105652396B discloses a device for optical fiber detection and an optical fiber detection device using the device. The light source is uniformly conveyed through the channel of the connecting seat sleeve body to realize the function of simultaneously detecting multiple optical fibers. However, the defect detection of the end face of the optical fiber is also not realized.

[0004] The existing patent CN104536090A discloses an optical fiber end face detection method and an optical fiber end face polishing and detection device. The damage defect detection method for the end face of the optical fiber is ordinary binary processing. However, the ordinary binary processing cannot realize automatic and intelligent detection, needs to set a threshold value by itself, needs to change parameters for different working environments, has low use efficiency, and is difficult to use. The existing patent CN208672278U discloses an optical fiber end face detection device and an optical fiber end face detection gripper. Multiple optical fibers are gripped and detected through the feeder, gripper and discharger, but does not mention what method is used to determine whether it is qualified, and the structure is complex, the reliability is not high, and errors are prone to occur. SUMMARY

[0005] The application aims to provide a method for detecting spot defects of an airborne optical fiber end face, a storage medium and equipment, and aims to realize automatic detection of defects of the airborne optical fiber end face by using image processing technology, to feed back and optimize the manufacturing and production process flow, and to further guarantee the quality stability and reliability of the optical fiber.

[0006] The application mainly realizes the following technical scheme:

[0007] A method for detecting spot defects of an airborne optical fiber end face, comprising the following steps:

[0008] Step S1: image acquisition is performed on the optical fiber, and the acquired image is preprocessed;

[0009] Step S2: the image after preprocessing is grayed to obtain an image Img1, and then binarization is performed;

[0010] Step S3: edge detection is performed on the image, the outer edge contour of the optical fiber is extracted, and the extracted outer edge contour is taken as a ROI (Region of Interest) region, a mask operation is performed on the gray image in the region to obtain an image Img2;

[0011] Step S4: automatic threshold binarization processing is performed on the image Img2, the image Img2 is binarized and segmented, and the high-light spots in the image Img2 are automatically extracted to detect the defect area S' of the high-light defect part of the optical fiber;

[0012] Step S5: the ratio F of the defect area S' to the area S of the image Img2 is taken as a defect rate.

[0013] In order to better realize the present application, further, in the step S4, the automatic threshold binarization processing on the image Img2 comprises the following steps:

[0014] Step S41: since the defect area will appear high-light abnormal spots, the cluster number K value and the initial normal area cluster center point gray value C 11 and the abnormal area cluster center point gray value C 12 are set in the 0-255 gray space;

[0015] Step S42: the distance D i of each pixel point in the image Img2 from the cluster center is calculated respectively k1 , k2 p

[0016] ;

[0017] According to the distance size, the Img2 pixel gray value is divided into a normal set Z1 and an abnormal set Z2; the pixel gray value close to C 11 is added to the normal set Z1, and vice versa;

[0018] Step S43: the pixel gray value of the nearest 80% of the pixel points in the Z1 and Z2 sets from the cluster center is taken respectively, and the mean value thereof is obtained to obtain new cluster centers C 21 and C 22 ;

[0019] ​Step S44: repeat the iteration of step S42 and step S43 until the iteration termination condition is met.

[0020] Step S45: calculate the binarization threshold value V:

[0021] ;

[0022] Wherein: t is the iteration number;

[0023] Step S46: based on the binarization threshold value V, the image Img2 is binarized and segmented.

[0024] In order to better realize the present application, further, in the step S41, the clustering number K value is 2, the initial C 11 =125, C 12 =190.

[0025] In order to better realize the present application, further, in the step S44, the iteration termination condition is:

[0026] .

[0027] In order to better realize the present application, further, in the step S1, any one or more of mean filtering, Gaussian filtering, median filtering and bilateral filtering is used to perform noise reduction preprocessing on the collected image.

[0028] In order to better realize the present application, further, in the step S2, the binarization method is any one of maximum entropy value segmentation method, mean segmentation method, Otsu method and triangular threshold segmentation method.

[0029] In order to better realize the present application, further, in the step S3, the edge detection method is any one of first-order differential operator, Roberts operator, Prewitt operator, Sobel operator and Canny operator method.

[0030] A computer readable storage medium, having stored thereon a computer program, which is executed by a processor to implement the above method.

[0031] An electronic device, comprising a memory and a processor; the memory has stored thereon a computer program; the processor is configured to execute the computer program in the memory to implement the above method.

[0032] The beneficial effects of the present application are as follows:

[0033] The image processing technology and one-dimensional K-means clustering method are used for automatic defect detection, the threshold segmentation boundary blur problem caused by general image binarization can be effectively avoided, the highlighted defect part can be more accurately extracted, and the method has the advantages of high efficiency, simplicity, stability and reliability. The method can effectively reduce the difficulty and fatigue intensity of manual detection, and automatic detection of the fiber end face defect is achieved. The method can effectively optimize and upgrade the production process from the fiber end face defect detection, and ensure the product quality. BRIEF DESCRIPTION OF DRAWINGS

[0034] Figure 1 A flowchart of the spot defect detection method of the on-board fiber end face in embodiment 1. DETAILED DESCRIPTION

[0035] Embodiment 1:

[0036] A spot defect detection method of an on-board fiber end face, as shown in Figure 1 , comprising the following steps:

[0037] Step 1: The fiber is installed and fixed on the clamp, the clamp is fixed on the conveying belt, and the conveying belt is driven by the rotating wheel to move.

[0038] Step 2: After completing a single displacement stroke, the image acquisition is performed when the fiber material reaches the image acquisition station.

[0039] Specifically, the measured fiber is placed in the clamping end of the clamp at the end of the clamp, then the servo motor drives the rotating wheel to rotate, the rotating wheel drives the conveying belt to convey the next fiber material, and the conveying stroke is uniform between two clamps, which ensures that the detection of the previous fiber is completed and the next clamp conveying fiber can be moved to the image acquisition station for photographing and processing detection.

[0040] Step 3: The collected image is preprocessed for noise reduction. Specifically, the image filtering and noise reduction processing can be selected from mean filtering, Gaussian filtering, median filtering, bilateral filtering and the like, and the selection is based on the characteristics of the actual detection object.

[0041] Step 4: The preprocessed image is grayed to obtain an image Img, and then binarized. Specifically, the binarization method can be selected from maximum entropy value segmentation method, mean segmentation method, Otsu method, triangular threshold segmentation method, and the selection is based on the characteristics of the actual detection object.

[0042] Step 5: Perform edge detection on the image to extract the edge features of the fiber's outer edge. Specifically, after binarization, perform edge detection on the image. Optional methods include the first-order differential operator, Roberts operator, Prewitt operator, Sobel operator, Canny operator, etc. The selection depends on the characteristics of the detection object in the actual situation.

[0043] Step 6: Using the extracted outer edge as the ROI (region of interest), mask the grayscale image Img within this region to obtain image Img2. The fiber's outer edge contour is then extracted based on its length or circular features. The area enclosed by this contour is then used as the ROI, and mask the grayscale image Img within this region to obtain the target region grayscale image Img2.

[0044] Step 7: Since optical fiber defects often appear as abnormally bright spots, automated extraction of these bright spots in image Img2 can detect the defective portion of the fiber. Therefore, image Img2 is automatically binarized using the proposed K-means local clustering binarization method based on regional grayscale values. This automated thresholding method is described below.

[0045] 1) Since there will be bright abnormal spots in the defective area, the cluster number K value and the initial normal area cluster center gray value C are set in the 0-255 gray space. 11 Gray value C of the center point of the abnormal regional cluster 12 .

[0046] Preferably, the K value of the cluster center is 2, and a binary classification based on the one-dimensional gray value is performed. Due to the highlight feature of the defect point, the normal area cluster center C of the initial cluster center is 11 =125, abnormal regional cluster center C 12 =190.

[0047] 2) Calculate the distance D between each pixel in Img2 and the cluster center k1 ,D k2 According to the distance, the grayscale values ​​of the pixels Img2 are divided into two sets: normal set Z1 and abnormal set Z2.

[0048] Specifically,

[0049] ;

[0050] Compare the two distances and divide the grayscale value of each pixel into two sets according to the closest cluster center. 11 Add Z1 and vice versa.

[0051] 3) Take the gray value of the 80% of the pixel points closest to the clustering center in Z1 and Z2 respectively, and get the new clustering center C by averaging 21 With C 22 .

[0052] 4) Because the gray value at the junction of the two sets is very close, the boundary is blurred, which is greatly affected by the clustering center. Therefore, when calculating the clustering center, take the gray value of the 80% of the pixel points closest to the clustering center in the set, form a new set Z1' and Z2', and then calculate the new clustering center C 21 With C 22 .

[0053] Specifically,

[0054] ;

[0055] ;

[0056] ;

[0057] .

[0058] 5) Repeat the iteration of the above steps 2) and 3). After a certain number of iterations, when the clustering center changes little in the adjacent number of iterations, the iteration loop is terminated.

[0059] 6) Calculate the threshold value V obtained by automatic operation.

[0060] After iteration t times, according to the actual demand control termination condition, for example, when the size of the clustering center calculated by the adjacent two times changes little, the binary threshold value V is calculated as the average of the last clustering center:

[0061]

[0062] .

[0063] Step 8: Take V as the binary threshold value, and perform binary segmentation on the image Img2. The ratio F of the area S' of the highlighted defect part to the area S of the image Img2 is the defect rate. According to the process requirement, set the threshold value of the ratio to determine whether the optical fiber is qualified. When the defect rate is greater than the process qualified requirement value, the computer upper computer software outputs an unqualified label and displays it on the display screen to remind. Otherwise, it indicates that the optical fiber quality is qualified.

[0064] Step 9: For the optical fiber with unqualified detection result, when it is transported above the unqualified material box, it can be released by loosening the clamping end of the clamp, and then it is stored in the unqualified material box. Otherwise, it will enter the qualified material box in the same way.

[0065] Step 10: When the qualified material box and the unqualified material box are full of optical fiber materials, the attached sensor sends a signal to the computer end, displays a prompt and timely cleaning information in the computer host software, and controls the servo motor to stop working.

[0066] Specifically, an optical sensor is provided in the fiber material box. When the optical fiber material is too full, the feedback signal stops the servo motor from working, and prompts the material to be full and cleaned in time in the display screen. After cleaning is completed, it can be restarted to resume work.

[0067] Step 11: After cleaning the material box, restart and continue to load the optical fiber for detection until all the optical fibers to be detected are detected.

[0068] The above is only a preferred embodiment of the present application, and does not limit the present application in any form. Any simple modification or equivalent change of the above embodiment according to the technical essence of the present application falls within the protection scope of the present application.

Claims

1. A method for detecting speckle defects on an airborne optical fiber end face, characterized in that: The following steps are involved: Step S1: collecting images of the optical fiber and preprocessing the collected images; Step S2: grayscale the preprocessed image to obtain image Img1, and then binarize it; Step S3: edge detection is performed on the image to extract the outer edge contour of the optical fiber, and the extracted outer edge contour is used as the ROI region of interest, and the grayscale image in the region is masked to obtain the image Img2; Step S4: performing automatic threshold binarization processing on the image Img2 and performing binary segmentation on the image Img2; Automatically extract the highlight spot in the image Img2 to detect the defect area S' of the highlight defect portion of the optical fiber; In step S4, performing automatic threshold binarization processing on the image Img2 includes the following steps: Step S41: Since the defective area will have bright abnormal spots, set the cluster number K value in the 0-255 grayscale space and the initial normal area cluster center grayscale value C 11 and the gray value C of the abnormal region cluster center 12 ; Step S42: Calculate the value of each pixel p in the image Img2 i The distance D from the cluster center k1 , D k2 : ； Among them, D k1 For each pixel p in image Img2 i Gray value C of the cluster center point from the normal area 11 The distance, D k2 For each pixel p in image Img2 i Gray value C of the cluster center point from the normal area 12 distance; According to the distance, the grayscale values ​​of the pixels Img2 are divided into the normal set Z1 and the abnormal set Z2; 11 Add to the normal set Z1, otherwise add to the abnormal set Z2; Step S43: Take the grayscale values ​​of 80% of the pixels closest to the cluster center in the Z1 and Z2 sets respectively, and calculate their average value to obtain the new cluster center C. 12 with C 22 ; Step S44: Repeat steps S42 and S43 until the iteration termination condition is met: ； Step S45: Calculate the binarization threshold V: ； Where: t is the number of iterations; Step S46: performing binary segmentation on the image Img2 based on the binary threshold V; Step S5: The ratio F of the defect area S' to the area S of the image Img2 is taken as the defect rate.

2. The method for detecting speckle defects on an airborne optical fiber end face according to claim 1, wherein: In step S41, the number of clusters K is 2, and the initial C 11 =125, C 12 =190.

3. The method for detecting speckle defects on an airborne optical fiber end face according to claim 1, wherein: In step S1, any one or more methods selected from the group consisting of mean filtering, Gaussian filtering, median filtering, and bilateral filtering are used to perform noise reduction preprocessing on the collected image.

4. The method for detecting speckle defects on an airborne optical fiber end face according to claim 1, wherein: In step S2, the binarization method is any one of the maximum entropy segmentation method, the mean segmentation method, the Otsu method, and the triangular threshold segmentation method.

5. The method for detecting speckle defects on an airborne optical fiber end face according to claim 1, wherein: In step S3, the edge detection method is any one of the first-order differential operator, Roberts operator, Prewitt operator, Sobel operator, and Canny operator methods.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

7. An electronic device, characterized in that: The method comprises a memory and a processor; the memory stores a computer program; the processor is configured to execute the computer program in the memory to implement the method according to any one of claims 1 to 5.

Images