三种保偏光纤几何参数测量算法
|
摘要
介绍了利用光纤端面图像采集和图像处理技术进行熊猫型保偏光纤几何参数测量的方法,重点介绍了三种图像处理算法——基于灰度分割和直方图分布的算法、基于Canny边缘检测和改进Hough变换的算法、基于Canny边缘检测和圆边界点选取的算法及其实现。利用三种方法对不同的光纤端面图像进行了处理,并对处理结果进行了对比分析。结果显示:三种算法均可用于熊猫型保偏光纤几何参数测量,针对不同的光纤端面图像选择合适的处理算法,可以实现快速、准确的测量。通过提高端面图像的分辨率,可以提高保偏光纤几何参数测量精度,基于该方法开发的自动化测量系统已成功应用于保偏光纤的质量与一致性实验。
Methods of image acquisition and processing technology of fiber end face to measure geometric parameters of panda-type polarization maintaining fiber was introduced,especially three image processing algorithms-gray-level segmentation and histogram distribution algorithm,Canny edge detection and improved Hough transform algorithm,Canny edge detection and circular boundary points selection algorithm and their implementation.The fiber end face images were processed using three algorithms to obtain centers and the diameters of the cladding layer,the core and two panda eyes.These results were analyzed and compared,which show that three algorithms have wide applicability on the measurement of panda-type polarization maintaining fiber geometric parameters.They can realize quick and accurate measurement by selecting the appropriate algorithm according to the fiber end face image.Accuracy of measurment can be improved through improving the image resolution.Automatic measurement system developed based on the method has been successfully applied in practice to test the quality and consistency of the panda-type polarization maintaining fiber.
Methods of image acquisition and processing technology of fiber end face to measure geometric parameters of panda-type polarization maintaining fiber was introduced,especially three image processing algorithms-gray-level segmentation and histogram distribution algorithm,Canny edge detection and improved Hough transform algorithm,Canny edge detection and circular boundary points selection algorithm and their implementation.The fiber end face images were processed using three algorithms to obtain centers and the diameters of the cladding layer,the core and two panda eyes.These results were analyzed and compared,which show that three algorithms have wide applicability on the measurement of panda-type polarization maintaining fiber geometric parameters.They can realize quick and accurate measurement by selecting the appropriate algorithm according to the fiber end face image.Accuracy of measurment can be improved through improving the image resolution.Automatic measurement system developed based on the method has been successfully applied in practice to test the quality and consistency of the panda-type polarization maintaining fiber.
引文
[1]Bai Chongben,Liu Youxin.Fiber Test[M].Beijing:Posts&Telecom Press,1988.(in Chinese)
[2]Zhao Xinyan.Fiber geometric parameters measurement withdigital image processing[J].Journal of Nanjing University ofPosts and Telecommunications(Natural Science),2009,29(2):55-58.(in Chinese)
[3]Zhang Bin,He Saixian.Improved edge-detection methodbased on Canny algorithm[J].Infrared Technology,2006,28(3):265-169.(in Chinese)
[4]Tang Lulu,Zhang Qican,Hu Song.An improved algorithmfor Canny edge detection with adaptive threshold[J].Opto-Electronic Engineering,2011,38(5):127-132.(in Chinese)
[5]Shao Xiaofang,Sun Jixiang,Wang Liangliang,et al.Animproved Canny edge detection algorithm[J].ElectronicsOptics&Control,2006,13(6):53-55.(in Chinese)
[6]Sun Yinghui.The study of edge detection based on Cannyoperator[J].Journal of Anshan Teachers College,2010,12(2):48-50.(in Chinese)
[7]Cui Jiwen,Tan Jiubin.Circle contour measurement techniquebased on randomized Hough transform using gradientinformation[J].Infrared and Laser Engineering,2006,35(4):419-424.(in Chinese)
[8]Huang Yonglin,Ye Yutang,Chen Zhenlong,et al.Newmethod of fast Hough transform for circle detection[J].Journal of Electronic Measurement and Instrument,2010,24(9):837-841.(in Chinese)
[9]Shang Lu,Li Rui,Song Xinyu.Improved Hough transformcircle detection algorithm[J].International ElectronicElements,2011,19(4):168-171.(in Chinese)
[2]Zhao Xinyan.Fiber geometric parameters measurement withdigital image processing[J].Journal of Nanjing University ofPosts and Telecommunications(Natural Science),2009,29(2):55-58.(in Chinese)
[3]Zhang Bin,He Saixian.Improved edge-detection methodbased on Canny algorithm[J].Infrared Technology,2006,28(3):265-169.(in Chinese)
[4]Tang Lulu,Zhang Qican,Hu Song.An improved algorithmfor Canny edge detection with adaptive threshold[J].Opto-Electronic Engineering,2011,38(5):127-132.(in Chinese)
[5]Shao Xiaofang,Sun Jixiang,Wang Liangliang,et al.Animproved Canny edge detection algorithm[J].ElectronicsOptics&Control,2006,13(6):53-55.(in Chinese)
[6]Sun Yinghui.The study of edge detection based on Cannyoperator[J].Journal of Anshan Teachers College,2010,12(2):48-50.(in Chinese)
[7]Cui Jiwen,Tan Jiubin.Circle contour measurement techniquebased on randomized Hough transform using gradientinformation[J].Infrared and Laser Engineering,2006,35(4):419-424.(in Chinese)
[8]Huang Yonglin,Ye Yutang,Chen Zhenlong,et al.Newmethod of fast Hough transform for circle detection[J].Journal of Electronic Measurement and Instrument,2010,24(9):837-841.(in Chinese)
[9]Shang Lu,Li Rui,Song Xinyu.Improved Hough transformcircle detection algorithm[J].International ElectronicElements,2011,19(4):168-171.(in Chinese)