基于机器视觉的光纤环绕制缺陷检测系统研究

发布时间：2018-06-11 12:50

本文选题：机器视觉 + 光纤环绕制　；参考：《中北大学》2017年硕士论文

【摘要】：光纤环是由光纤丝按照一定的规则缠绕而成的圆环形线圈,它是干涉型光纤陀螺仪的核心部件,光纤环的绕制品质是影响光纤陀螺性能的关键因素。由于光纤直径误差、静电以及内应力等随机因素的影响,光纤环在实际的绕制过程中会偶尔出现一些绕制缺陷。对光纤环绕制缺陷进行检测既是光纤环生产过程中的一个重要环节,也是保证光纤环绕制质量的必要条件,因此开展基于机器视觉的光纤环绕制缺陷检测研究具有重要的理论意义和工程实用价值。结合光纤绕制的特点和缺陷检测的要求,完成了光纤绕制缺陷检测系统的整体设计,其中对采集系统中的摄像机和镜头的选取以及对照明系统中光源的照明方式进行了科学分析,利用清晰度评价函数分析了不同光强对光纤环绕制图像质量的影响,为照明系统中光强的选择提供了理论依据。为了消除光纤环绕制图像中噪声对后期的缺陷检测的干扰,本文对图像预处理展开了研究。通过实验对比以及分析确定了最佳的光纤环绕制图像预处理方法,然后经过高斯滤波、灰度增强、图像分割处理得到了理想的光纤轮廓目标图像。对光纤环的绕制方法进行了介绍,并根据绕制的特点提出了一种通用的光纤环绕制缺陷动态检测算法,然后对其中的光纤轮廓边缘精确定位算法以及缺陷判别算法展开了深入的研究。通过实验分析了不同光照强度对系统检测误差的影响,并验证了照明系统搭建的合理性。从误检率、检测精度以及检测速度三个方面对系统进行了实验。实验结果表明:系统的检测精度可达0.03mm,误检率为8%,系统的检测速度可以满足实时性要求,从而证实了该系统具有一定的可行性和实用性。
[Abstract]:Fiber optic ring is a kind of ring coil which is made up of fiber optic wire winding according to certain rules. It is the core component of interferometric fiber optic gyroscope. The winding quality of fiber optic ring is the key factor to affect the performance of fiber optic gyroscope. Due to the influence of random factors, such as fiber diameter error, static electricity and internal stress, the fiber ring will occasionally appear some winding defects in the actual winding process. The detection of optical fiber ring defects is not only an important link in the production process of optical fiber ring, but also a necessary condition to ensure the quality of optical fiber ring making. Therefore, the research of optical fiber surround defect detection based on machine vision has important theoretical significance and practical value. According to the characteristics of optical fiber winding and the requirements of defect detection, the overall design of optical fiber wound defect detection system is completed. The selection of the camera and lens in the acquisition system and the illumination mode of the light source in the illumination system are analyzed scientifically, and the influence of different light intensity on the image quality of the optical fiber surround system is analyzed by using the definition evaluation function. It provides a theoretical basis for the choice of light intensity in lighting system. In order to eliminate the noise interference in the optical fiber surround image, the image preprocessing is studied in this paper. Through experimental comparison and analysis, the optimal preprocessing method of optical fiber surround mapping image is determined, and then the ideal optical fiber contour target image is obtained by Gao Si filtering, gray enhancement and image segmentation. In this paper, the winding method of optical fiber ring is introduced, and according to the characteristics of fiber winding, a universal algorithm for dynamic detection of defects in optical fiber encircling is proposed. Then, the optical fiber contour edge accurate location algorithm and defect discrimination algorithm are studied. The influence of different illumination intensity on the detection error of the system is analyzed experimentally, and the rationality of the lighting system is verified. The system is tested from three aspects: false detection rate, detection precision and detection speed. The experimental results show that the detection accuracy of the system can reach 0.03mm, the error detection rate is 8 and the detection speed of the system can meet the real-time requirements, which proves the feasibility and practicability of the system.
【学位授予单位】：中北大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP391.41

【参考文献】