基于机器视觉的编码器码盘零点检测系统研究

发布时间：2018-01-10 05:29

本文关键词：基于机器视觉的编码器码盘零点检测系统研究　出处：《长春工业大学》2015年硕士论文　论文类型：学位论文

【摘要】：目前,光电编码器在精密角位置的测量、精密定位、数显及数控等自动化设备当中广泛应用,企业对编码器的精度、稳定性和抗干扰性提出了更高层次的标准。编码器安装是否精准,直接影响其精确度。增量式光电编码器的码盘上,有一组特殊窄缝,用于产生零位信号,在增量式编码器安装时,码盘上的零点与检测光栅上的零点对齐是否精准至关重要,能够影响编码器检测精度、稳定性与可靠性。近年来,伴随着制造行业不断进步,在机器视觉领域开展了大量应用研究。在某些领域机器视觉几乎可以代替人类操作,应用在任何需要人眼工作的场合,甚至完成人类不能完成的工作,类似感知不可见物体、危险场景等,利用计算机视觉这一计算机产业化的技术,可以充分显示出它的优势。针对增量式光电编码器,传统的安装方法是由厂家工人在光电编码器出厂前完成目测校对,由于人工检测工作效率低,准确性不高,影响了光电编码器的准确程度且无法满足实际需求。本研究结合企业实际项目,设计一种编码器码盘零点视觉检测系统。利用摄像机、计算机、以及图像采集卡搭建了增量式光电编码器码盘零点检测平台,通过摄像机采集不同位置时刻编码器码盘以及检测光栅图像,利用NI视觉软件、通过LabVIWE编写处理程序,对编码器码盘和检测光栅图像进行预处理、形态学处理、边缘检测等操作,利用自动目标识别霍夫变换圆弧检测原理确定码盘位置,然后用模版匹配方法基于零点特征点进行匹配识别,进而完成检测光栅的固定,最后使编码器码盘零点与检测光栅零点对齐。实验证明,本文应用机器视觉技术,针对编码器安装过程中零点对齐问题设计开发了一套视觉识别系统,基于机器视觉,可以实现编码器安装过程中自动检测零点,能够提高光电编码器安装的精度和效率,增强光电编码器的稳定性与可靠性,降低人工操作工作量大和判断不准确等问题,该系统具有良好的人机交互界面,可以对不同类型的增量编码器进行实时检测识别。
[Abstract]:At present, the photoelectric encoder is widely used in the precision angle position measurement, precision positioning, digital display and numerical control and other automation equipment, the accuracy of the encoder. Stability and anti-jamming put forward a higher standard. Whether the encoder is accurate or not has a direct impact on its accuracy. There is a special set of narrow slits on the code plate of incremental photoelectric encoder to generate zero signal. When the incremental encoder is installed, the accuracy of 00:00 on the code plate and 00:00 on the detection grating is very important, which can affect the accuracy, stability and reliability of the encoder. With the continuous progress of the manufacturing industry, a large number of applications have been carried out in the field of machine vision. In some areas, machine vision can almost replace human operation, and it can be used in any situation where the human eye is required to work. Even to complete the work that human beings can't do, such as perceiving invisible objects, dangerous scenes and so on, using computer vision, the technology of computer industrialization. For incremental photoelectric encoder, the traditional installation method is for factory workers to complete visual proofreading before the photoelectric encoder leaves the factory. Because of the low efficiency of manual detection, the accuracy is not high. The accuracy of the photoelectric encoder is affected and can not meet the actual demand. This research combines the actual project of the enterprise, designs a kind of encoder code disk 00:00 vision detection system, uses the camera, the computer. And the image acquisition card set up an incremental photoelectric encoder code plate 00:00 detection platform, through the camera to collect different positions of time encoder code disk and detection grating image, using NI vision software. Through the LabVIWE program, the encoder and the detection grating image preprocessing, morphology processing, edge detection and other operations. Using the automatic target recognition Hough transform arc detection principle to determine the position of the code disk, then using template matching method based on 00:00 feature points to identify, and then complete the fixed detection grating. Finally, the encoder code disk is aligned with the detection grating 00:00 at 00:00. The experiment proves that a vision recognition system is designed and developed by using machine vision technology, aiming at the alignment problem of 00:00 in the course of encoder installation. Based on machine vision, it can automatically detect 00:00 in the process of encoder installation. It can improve the accuracy and efficiency of photoelectric encoder installation, and enhance the stability and reliability of photoelectric encoder. The system has a good man-machine interface and can detect and identify different types of incremental encoders in real time.
【学位授予单位】：长春工业大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TN762;TP391.41

【参考文献】