CN103115577A - Optic fiber dimension measurement algorithm based on machine vision - Google Patents

Optic fiber dimension measurement algorithm based on machine vision Download PDF

Info

Publication number
CN103115577A
CN103115577A CN2013100274149A CN201310027414A CN103115577A CN 103115577 A CN103115577 A CN 103115577A CN 2013100274149 A CN2013100274149 A CN 2013100274149A CN 201310027414 A CN201310027414 A CN 201310027414A CN 103115577 A CN103115577 A CN 103115577A
Authority
CN
China
Prior art keywords
obtains
image
circle
edge
fibre core
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN2013100274149A
Other languages
Chinese (zh)
Inventor
丛媛
陈晓荣
刘晓东
奚传立
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
University of Shanghai for Science and Technology
Original Assignee
University of Shanghai for Science and Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by University of Shanghai for Science and Technology filed Critical University of Shanghai for Science and Technology
Priority to CN2013100274149A priority Critical patent/CN103115577A/en
Publication of CN103115577A publication Critical patent/CN103115577A/en
Pending legal-status Critical Current

Links

Images

Abstract

The invention relates to an optical fiber dimension measurement algorithm based on machine vision. An image which needs to be measured is read, a gray level histogram of the image is obtained, then a threshold value of a boundary of a wrapping layer and a fiber core is obtained through the gray level histogram, the wrapping layer and the fiber core are separated according to the obtained threshold value, and a notch part is found out and cut according to the threshold value; the fiber core area and the wrapping layer area are extracted respectively for multiple measurement of relevant parameters, a mean value is obtained, circle centers of the fiber core and the wrapping layer are obtained, and concentricity of the fiber core and the wrapping layer is calculated and obtained. According to the method, a machine vision technology is used, the image of the end face of an optical fiber is subjected to threshold value division, interference removal, filtering, fitting and other processes, various geometric dimension parameters of the optical fiber are measured efficiently and intelligently. Compared with a traditional optical fiber measurement method, the method has the advantages of being fast in processing speed and high in measuring precision, and operation level is not depended on excessively. The method can be popularized and applied in dimension measurement of other small linear objects.

Description

A kind of fiber size Measurement Algorithm based on machine vision
Technical field
The present invention relates to a kind of measuring technique, particularly a kind of fiber size Measurement Algorithm based on machine vision.
Background technology
In recent years along with the development of optical fiber technology, the precision measurement of the various parameters of optical fiber itself is to the manufacturing of optical fiber and use all that tool is of great significance.For guaranteeing its quality, production and application department all strictly carry out multiple check by national standard to optical fiber, and nearly 4 kinds of its method of inspection is respectively: artificial cognition method, image cut method, pulse counting method, digital image processing method.The edge feature that extracts the reflection grey scale change in the image is an important branch of Digital Image Processing.Method based on Digital Image Processing is measured automatically to the geometric parameter of fiber end face, has the advantage that processing speed is fast, measuring accuracy is high and too do not rely on operant level, and the method is the trend of industry.
Summary of the invention
The present invention be directed to digital image processing method and meet the problem that present optical fiber quality detects needs, a kind of new fiber size Measurement Algorithm based on machine vision has been proposed, the applied for machines vision technique, measuring accuracy and speed have greatly been improved, be fit to the dimensional measurement of fine object, relate in particular to the measurement of the physical dimension parameter of large aperture optical fiber.
Technical scheme of the present invention is: a kind of fiber size Measurement Algorithm based on machine vision specifically comprises the steps:
1) image pre-service: read the image that to measure, obtain its grey level histogram;
2) remove disturbing factor: the average gray value that obtains image according to grey level histogram, the relation of then closing marginal gray-scale value and average gray value according to covering and fibre core obtains threshold value, according to the threshold value that obtains covering and core segment are distinguished, then utilize this threshold value cut out portion to be found out and with its excision;
3) obtain the cladding regions of needs rim detection: according to step 2), graph transformation is carried out in the zone that obtains, obtain the circle with regional approximate size, the border circular areas that utilization obtains intercepts away the image of cladding regions from original image, then the correlation parameter of fibre core is measured, obtain the edge of fibre core according to function operators, edge is selected, combination, obtains required edge;
4) match is measured: the edge that obtains is carried out the shape match, match obtains circle and ellipse respectively, can obtain diameter and the center of circle of fibre core, change different edges and select parameter, take multiple measurements, obtain rational average, can confirm as diameter and the center of circle of fibre core, and the major and minor axis radius of determining the ellipse that match obtains R Long And R Short, according to following formula:
Figure 2013100274149100002DEST_PATH_IMAGE002
Can be in the hope of fiber core non circularity;
5) match of covering is measured: after the fibre core measurement is complete, the zone of core segment is clipped from original image, simultaneously cut out portion is removed from former figure, step is with 1) and 2), carry out the measurement of clad section dimensional parameters with remaining image section, obtain the Edge detected of covering, the edge that needs, and fitting circle, fitted ellipse, cladding diameter, the center of circle and out-of-roundness obtained through repeated experiments;
6) center of circle that step 4) and 5) measures fibre core and covering is respectively: A (X 1, Y 1), B (X 2, Y 2), then concentricity can be calculated by following formula:
Figure 2013100274149100002DEST_PATH_IMAGE004
Beneficial effect of the present invention is: the fiber size Measurement Algorithm that the present invention is based on machine vision, the applied for machines vision technique, the fiber end face image is carried out Threshold segmentation, removes the processing such as interference, filtering, match, every physical dimension parameter of high efficiency smart ground measuring optical fiber.Compare with traditional optical fibre measuring method, the method has the advantage that processing speed is fast, measuring accuracy is high and too do not rely on operant level.Can be applied to the dimensional measurement of other small class thread-shaped bodies.
Description of drawings
Fig. 1 is the fiber size Measurement Algorithm process flow diagram that the present invention is based on machine vision;
Fig. 2 is the original image figure that the embodiment of the invention is read in;
Fig. 3 is the grey level histogram of embodiment of the invention original image;
Fig. 4 is the figure after the embodiment of the invention is clipped otch;
Fig. 5 is the figure that the embodiment of the invention is removed the covering edge effect;
Fig. 6 is the fibre core edge that obtains of the embodiment of the invention and the fibre core outline map of selection;
Fig. 7 is match circle and the fitted ellipse figure at embodiment of the invention fibre core edge;
Fig. 8 is the figure after the embodiment of the invention is removed fibre core edge and otch impact;
Fig. 9 is the covering edge that obtains of the embodiment of the invention and the covering outline map of selection;
Figure 10 is fitting circle and the fitted ellipse figure at embodiment of the invention covering edge.
Embodiment
Based on the fiber size Measurement Algorithm process flow diagram of machine vision, specifically comprise the steps: as shown in Figure 1
1, image pre-service: the image that needs are measured reads in, and reads original image as shown in Figure 2; Obtain its grey level histogram, as shown in Figure 3.
2. removal disturbing factor: obtain the average gray value of image according to grey level histogram, then obtain the threshold value of needs according to the mutual relationship of border gray-scale value and average gray value among the figure.Can see that such as accompanying drawing 2 image of incision no longer is the original circular arc of optical fiber, but become the irregular range of linearity, correct result can be disturbed in the border of obtaining like this, therefore will first irregular incision tract be cut away, to get rid of it to the impact of diameter measurement.Can covering and core segment be distinguished according to the threshold value that obtains, then utilize this threshold value cut out portion to be found out and with its excision.Image behind the excision otch as shown in Figure 4.
3. rim detection: utilize the threshold value of trying to achieve previously to obtain the cladding regions that needs, graph transformation is carried out in the zone that obtains, obtain the circle with regional approximate size, the border circular areas that utilization obtains intercepts away the image of cladding regions from original image, so just removed the impact of covering outer boundary, then image after the intercepting can be measured the correlation parameter of fibre core as shown in Figure 5.Obtain the edge of fibre core according to function operators, edge is selected, combination, obtains an optimal edge the longest, has shown the edge (left figure) that detects in the accompanying drawing 6, and select, in conjunction with after edge (right figure):
4. match is measured: the edge that obtains is carried out the shape match, and match obtains circle and ellipse respectively.As shown in Figure 7 (left figure is that fibre core fitting circle and the center of circle, right figure are fitted ellipse).So far, can obtain diameter and the center of circle of fibre core, change different edge and select parameter, take multiple measurements, obtain rational average, can confirm as diameter and the center of circle of fibre core, and the major and minor axis radius of determining the ellipse that match obtains R Long And R Short, according to following formula:
Can be in the hope of fiber core non circularity.
5. the match of covering is measured: after the fibre core measurement is complete, the zone of core segment is clipped from original image, simultaneously cut out portion is also removed from former figure, provide above the method.Carry out the measurement of clad section dimensional parameters with remaining image section, the image after the intercepting as shown in Figure 8.In like manner, can utilize said method to detect, obtain the Edge detected of covering, the edge that needs; And fitting circle, fitted ellipse.Shown in result such as the accompanying drawing 9,10.Obtain cladding diameter, the center of circle and out-of-roundness through repeated experiments.Suppose that the center of circle that measures fibre core and covering is respectively: A (X 1, Y 1), B (X 2, Y 2).Then concentricity can be calculated by following formula:
Figure 313410DEST_PATH_IMAGE004

Claims (1)

1. the fiber size Measurement Algorithm based on machine vision is characterized in that, specifically comprises the steps:
1) image pre-service: read the image that to measure, obtain its grey level histogram;
2) remove disturbing factor: the average gray value that obtains image according to grey level histogram, the relation of then closing marginal gray-scale value and average gray value according to covering and fibre core obtains threshold value, according to the threshold value that obtains covering and core segment are distinguished, then utilize this threshold value cut out portion to be found out and with its excision;
3) obtain the cladding regions of needs rim detection: according to step 2), graph transformation is carried out in the zone that obtains, obtain the circle with regional approximate size, the border circular areas that utilization obtains intercepts away the image of cladding regions from original image, then the correlation parameter of fibre core is measured, obtain the edge of fibre core according to function operators, edge is selected, combination, obtains required edge;
4) match is measured: the edge that obtains is carried out the shape match, match obtains circle and ellipse respectively, can obtain diameter and the center of circle of fibre core, change different edges and select parameter, take multiple measurements, obtain rational average, can confirm as diameter and the center of circle of fibre core, and the major and minor axis radius of determining the ellipse that match obtains R Long And R Short, according to following formula:
Figure 2013100274149100001DEST_PATH_IMAGE002
Can be in the hope of fiber core non circularity;
5) match of covering is measured: after the fibre core measurement is complete, the zone of core segment is clipped from original image, simultaneously cut out portion is removed from former figure, step is with 1) and 2), carry out the measurement of clad section dimensional parameters with remaining image section, obtain the Edge detected of covering, the edge that needs, and fitting circle, fitted ellipse, cladding diameter, the center of circle and out-of-roundness obtained through repeated experiments;
6) center of circle that step 4) and 5) measures fibre core and covering is respectively: A (X 1, Y 1), B (X 2, Y 2), then concentricity can be calculated by following formula:
Figure 2013100274149100001DEST_PATH_IMAGE004
CN2013100274149A 2013-01-25 2013-01-25 Optic fiber dimension measurement algorithm based on machine vision Pending CN103115577A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN2013100274149A CN103115577A (en) 2013-01-25 2013-01-25 Optic fiber dimension measurement algorithm based on machine vision

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN2013100274149A CN103115577A (en) 2013-01-25 2013-01-25 Optic fiber dimension measurement algorithm based on machine vision

Publications (1)

Publication Number Publication Date
CN103115577A true CN103115577A (en) 2013-05-22

Family

ID=48413996

Family Applications (1)

Application Number Title Priority Date Filing Date
CN2013100274149A Pending CN103115577A (en) 2013-01-25 2013-01-25 Optic fiber dimension measurement algorithm based on machine vision

Country Status (1)

Country Link
CN (1) CN103115577A (en)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103438802A (en) * 2013-09-17 2013-12-11 侯俊 Method for measuring geometric parameters of optical fiber coating layer
CN105676356A (en) * 2016-03-15 2016-06-15 中国工程物理研究院激光聚变研究中心 Fiber core positioning method and fiber core alignment calibration method for optical fiber fusion
CN108876777A (en) * 2018-06-14 2018-11-23 重庆科技学院 A kind of visible detection method and system of wind electricity blade end face of flange characteristic size
CN110068278A (en) * 2019-04-22 2019-07-30 南京理工大学 Non-contact optical fiber preform size real-time measurement system and method based on FPGA
CN114986633A (en) * 2022-05-05 2022-09-02 四川大学 Machine vision-based automatic selection system and method for original bamboo splitting tool

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6661502B1 (en) * 1999-10-28 2003-12-09 Fitel Usa Corp. Method and apparatus for measuring the diameter and/or eccentricity of a coating layer of a coated optical fiber
US7162073B1 (en) * 2001-11-30 2007-01-09 Cognex Technology And Investment Corporation Methods and apparatuses for detecting classifying and measuring spot defects in an image of an object
CN101634603A (en) * 2009-05-21 2010-01-27 沈群华 Improved optical fiber geometric parameter tester
CN102519398A (en) * 2011-10-28 2012-06-27 广西师范大学 Optical fiber base pin concentricity measuring method based on visual measurement and system thereof

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6661502B1 (en) * 1999-10-28 2003-12-09 Fitel Usa Corp. Method and apparatus for measuring the diameter and/or eccentricity of a coating layer of a coated optical fiber
US7162073B1 (en) * 2001-11-30 2007-01-09 Cognex Technology And Investment Corporation Methods and apparatuses for detecting classifying and measuring spot defects in an image of an object
CN101634603A (en) * 2009-05-21 2010-01-27 沈群华 Improved optical fiber geometric parameter tester
CN102519398A (en) * 2011-10-28 2012-06-27 广西师范大学 Optical fiber base pin concentricity measuring method based on visual measurement and system thereof

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
杨远: "基于机器视觉的光纤几何参数检测研究", 《中国优秀硕士学位论文全文数据库-信息科技辑》, no. 6, 15 June 2012 (2012-06-15) *
连国栋: "利用图像处理技术进行空芯光纤几何参数的测量", 《中国优秀硕士学位论文全文数据库-信息科技辑》, no. 2, 15 June 2003 (2003-06-15) *
陈磊;陈进榜;陆润华: "光纤几何参数的自动检测仪", 《光学学报》, vol. 21, no. 10, 31 October 2001 (2001-10-31) *

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103438802A (en) * 2013-09-17 2013-12-11 侯俊 Method for measuring geometric parameters of optical fiber coating layer
CN103438802B (en) * 2013-09-17 2016-04-20 上海理工大学 Optical fiber coating geometric parameter measurement method
CN105676356A (en) * 2016-03-15 2016-06-15 中国工程物理研究院激光聚变研究中心 Fiber core positioning method and fiber core alignment calibration method for optical fiber fusion
CN105676356B (en) * 2016-03-15 2019-03-01 中国工程物理研究院激光聚变研究中心 A kind of localization method of fibre core and the fibre core of fused fiber splice align calibration method
CN108876777A (en) * 2018-06-14 2018-11-23 重庆科技学院 A kind of visible detection method and system of wind electricity blade end face of flange characteristic size
CN110068278A (en) * 2019-04-22 2019-07-30 南京理工大学 Non-contact optical fiber preform size real-time measurement system and method based on FPGA
CN114986633A (en) * 2022-05-05 2022-09-02 四川大学 Machine vision-based automatic selection system and method for original bamboo splitting tool

Similar Documents

Publication Publication Date Title
US11200351B2 (en) Method for constructing curve of robot processing path of part with small curvature based on point cloud boundary
CN103115577A (en) Optic fiber dimension measurement algorithm based on machine vision
CN102589435B (en) Efficient and accurate detection method of laser beam center under noise environment
CN105279756B (en) Notch circular arc accessory size visible detection method based on adaptive region segmentation
CN106824806B (en) The detection method of low module plastic gear based on machine vision
CN109685760B (en) MATLAB-based SLM powder bed powder laying image convex hull depression defect detection method
CN109658402A (en) Industry profile geometric dimension automatic testing method based on computer vision imaging
CN104949925A (en) Pearl grading method
CN103606169A (en) Method for detecting defects of bottle cap
JP5960125B2 (en) Method and apparatus for optically measuring the thickness of an object by interferometry
CN108896278A (en) A kind of optical filter silk-screen defect inspection method, device and terminal device
CN107545247B (en) Stereo cognition method based on binocular recognition
CN112330663B (en) Method for detecting width of cut tobacco based on reducing circle by computer vision
CN114279357A (en) Die casting burr size measurement method and system based on machine vision
CN113865508B (en) Automatic detection device and method for through hole rate of sound lining of honeycomb sandwich composite material
JP2005538473A (en) A method for minimizing the influence of interference signals in the calculation of shape elements from coordinate points.
CN105184792B (en) A kind of saw blade wear extent On-line Measuring Method
CN104614386A (en) Lens defects type identification method
CN115526885A (en) Product image defect detection method, system, device and medium
CN104259110A (en) Detection method of ceramic valve element
CN105992930B (en) Data processing method and data processing equipment
CN103163156A (en) Automatic grading method of lens defects based on machine vision technology
CN103493096B (en) The analysis of the digital picture of tire inner surface and the processing of pseudo-measurement point
CN111815575A (en) Bearing steel ball part detection method based on machine vision
CN109584212B (en) MATLAB-based SLM powder bed powder laying image scratch defect identification method

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C02 Deemed withdrawal of patent application after publication (patent law 2001)
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20130522