基于机器视觉的飞机铆钉尺寸测量和缺陷检测系统的研究

发布时间：2018-04-29 22:39

本文选题：机器视觉 + 飞机铆钉　；参考：《陕西科技大学》2017年硕士论文

【摘要】：随着我国飞机制造业的快速发展,飞机铆钉作为飞机装配的必要零件被大量需求。为了保证飞机铆钉的质量,并区分产品的规格等级,需要运用一定的技术手段对铆钉产品进行尺寸测量和表面缺陷检测。然而目前大多飞机铆钉的生产厂家仍然使用传统的人工检测方法对产品进行检测,该检测方法存在着检测效率低、检测成本高的问题,已经不能满足现代工业快速、精密和稳定的测量及检测要求。基于机器视觉的工业检测技术因其具有非接触、在线检测、实时分析和判断的特点以及速度快、精度高、效率高的优点,目前已广泛应用于现代工业大生产的检测领域。本文将飞机铆钉作为检测对象,研制了一种基于机器视觉的飞机铆钉尺寸测量和表面缺陷检测系统。通过分析飞机铆钉的形状结构及表面缺陷的特征信息,对检测系统的硬件部分进行选型和设计,对飞机铆钉尺寸测量算法和表面缺陷检测算法进行研究,并对检测系统的软件部分进行了设计。本文的研究工作主要包括以下几个部分:(1)根据铆钉生产企业提出的对飞机铆钉检测的性能指标,结合工艺条件,设计了一个包括运动控制部分、图像采集和处理部分以及光源照明系统的机器视觉检测的硬件系统。选择合适的工业相机和光学镜头,对光源照明系统进行选型和设计,并对光源照明系统进行优化,以保证待测铆钉图像的质量。(2)对飞机铆钉尺寸测量算法进行了研究。根据飞机铆钉的测量需要及铆钉特征,确定的算法流程为:首先对原始铆钉图像进行预处理并对存在畸变的铆钉图像进行畸变补偿,确定铆钉图像的边缘图像,最后对铆钉图像的圆形部分采用最小二乘法拟合圆的方法进行测量,同时对铆钉图像的线型特征采用Hough变换的方法确定铆钉的尺寸参数。(3)对飞机铆钉表面缺陷检测算法进行了研究。表面缺陷检测算法主要由缺陷补偿算法和缺陷判断算法组成。缺陷补偿算法采用稀疏分解和聚类分析的方法获得飞机铆钉的模板图像,与待测飞机铆钉进行差分运算后提取铆钉缺陷图像,缺陷判断算法为根据铆钉缺陷图像的几何特征和形状特征判断缺陷的类型。(4)在Visual Studio 2010环境下,结合MFC和Open CV库实现了软件系统的设计。运用MFC设计友好的人机交互界面,采用模块化编程的思想,利用Open CV软件实现了飞机铆钉尺寸测量算法和表面缺陷检测算法,并通过数据库对飞机铆钉尺寸测量和缺陷检测的信息进行保存。最后,在完成系统构建、算法设计和软件实现的基础上,通过实验对本文所设计的检测系统的效果进行了验证。实验结果表明,本文所设计的基于机器视觉的飞机铆钉尺寸测量和表面缺陷检测系统能够实现飞机铆钉的高精度测量以及准确的缺陷判断,满足各方面的检测要求。
[Abstract]:With the rapid development of China's aircraft manufacturing industry, aircraft rivets are required as necessary parts for aircraft assembly. In order to ensure the quality of aircraft rivets and distinguish the grade of products, it is necessary to use certain technical means to measure the size and detect the surface defects of the rivets. However, at present, most manufacturers of aircraft rivets still use the traditional manual testing method to detect the products. The detection method has the problems of low efficiency and high cost, and can not meet the rapid development of modern industry. Precise and stable measurement and testing requirements. The industrial detection technology based on machine vision has been widely used in the field of modern industrial production because of its characteristics of non-contact, on-line detection, real-time analysis and judgment, as well as the advantages of high speed, high precision and high efficiency. In this paper, an aircraft rivet size measurement and surface defect detection system based on machine vision is developed. By analyzing the shape and structure of aircraft rivets and the characteristic information of surface defects, the hardware part of the detection system is selected and designed, and the algorithm of measuring the size of aircraft rivets and the algorithm of detecting surface defects are studied. And the software part of the detection system is designed. The research work of this paper mainly includes the following parts: 1) according to the performance index of the rivet inspection proposed by the rivet manufacturing enterprise and combining with the technological conditions, a motion control part is designed. The part of image acquisition and processing and the hardware system of machine vision detection of light source lighting system. Selecting the appropriate industrial camera and optical lens, selecting and designing the lighting system of light source, and optimizing the lighting system of light source to ensure the quality of rivet image to be tested, the algorithm of measuring the size of aircraft rivet is studied. According to the measurement requirements and rivet characteristics of aircraft rivets, the algorithm flow is as follows: first, preprocess the original rivets image and compensate the distortion of the rivet image with distortion, and determine the edge image of the rivet image. Finally, the circular part of the rivet image is measured by using the least square method to fit the circle. At the same time, the method of Hough transform is used to determine the size parameter of rivets, and the algorithm for detecting the surface defects of aircraft rivets is studied. The surface defect detection algorithm is mainly composed of defect compensation algorithm and defect judgment algorithm. The defect compensation algorithm uses sparse decomposition and clustering analysis to obtain the template image of the aircraft rivets, and then extracts the rivet defect image after differential operation with the aircraft rivet to be tested. The defect judgment algorithm is based on the geometric and shape features of rivets defect image to judge the type of defect. In Visual Studio 2010 environment, the software system is designed by combining MFC and Open CV library. Using MFC to design friendly man-machine interface, using the idea of modular programming, using Open CV software to realize the aircraft rivet size measurement algorithm and surface defect detection algorithm. The information of size measurement and defect detection of aircraft rivets is saved by database. Finally, on the basis of the system construction, algorithm design and software implementation, the effect of the detection system designed in this paper is verified by experiments. The experimental results show that the aircraft rivet size measurement and surface defect detection system based on machine vision can realize the high precision measurement and accurate defect judgment of aircraft rivets, and meet the requirements of all aspects of the detection.
【学位授予单位】：陕西科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：V262.4;TP391.41

【参考文献】