基于机器视觉的玻璃温度计自动识别系统的研究
发布时间：2018-09-19 19:40
【摘要】：玻璃液体温度计因其测温精度高、携带方便等优点而被广泛应用于各个领域,因此玻璃液体温度计的使用量比较大。玻璃液体温度计需要进行定期的检定,当地的计量部门负责该检定工作,导致计量部门的工作任务繁重,通过之前对计量部门的研究,发现大多数计量部门通过人工读数的方式对玻璃温度计进行检定。人工读数的方式容易导致以下几种结果:温度计数量过多使得检定人员的疲倦而导致检定结果不准确;检定人员的主观因素影响温度计的读数而导致检定结果不准确;不能及时储存任意温度计的数据等原因导致检定速度慢。因此对玻璃液体温度计的自动检定的研究是非常必要的。随着科技的进步,特别是智能化、自动化的不断发展,机器视觉用于玻璃温度计的自动检定就受到了众多研究者的广泛重视。本文在机器视觉的基础上,利用恒温水浴锅保持温度计的温度稳定,通过一定分辨率的工业相机对温度计图像进行获取,对整个系统中的重要部分进行了分析,并通过一定的图像处理技术对目标物进行特征提取,以便进一步分析研究。对温度计的图像分别提取了温度计的液柱、刻度线以及数字,最后计算出了温度计的示数。论文研究的主要内容包括:对基于机器视觉的温度计自动识别系统进行了分析和研究,简述了采集到的图像的处理方以及该系统的恒温装置的实现方式,分析了工业相机的相关参数,并采用张正友标定发对工业相机进行标定,使得采集到的图像没有畸变的干扰。按照之前设计的温度计自动识别系统进行了现场的模拟实验,调节恒温水浴装置设置不同的温度,待温度稳定后利用工业相机采集图像。采用MATLAB软件平台对采集到的图像进行处理,处理的过程包括:图像的降噪去除获取图像过程中产生的各种噪声;图像的对比度的拉伸以突出图像中的温度计部分;图像的分割、特征提取以分别提取出温度计的液柱与刻度线;图像的倾斜校正以提高温度计识别精度;并且对温度计上的数字进行分割提取再进行模板匹配、识别。通过之前对图像的处理计算出温度计的示数,并采用Visual Studio结合HALCON机器视觉软件设计出相应的用户界面来显示图像及结果信息。
[Abstract]:Glass liquid thermometer is widely used in various fields because of its high temperature measuring accuracy and convenient carrying, so the use of glass liquid thermometer is relatively large. The glass liquid thermometer needs regular verification, and the local metrology department is responsible for the verification work, resulting in the heavy task of the measurement department. It is found that most measurement departments calibrate glass thermometers by manual reading. The way of manual reading is easy to lead to the following results: the excessive number of thermometers makes the verification result inaccurate, the subjective factors of the calibrator influence the reading of the thermometer, and the verification result is inaccurate; The speed of verification is slow due to the failure to store the data of arbitrary thermometers in time. Therefore, it is necessary to study the automatic verification of glass liquid thermometer. With the development of science and technology, especially the development of intelligence and automation, the automatic verification of glass thermometers by machine vision has been paid more and more attention by many researchers. On the basis of machine vision, this paper uses a constant temperature water bath pot to keep the thermometer temperature stable, obtains the thermometer image by a certain resolution industrial camera, and analyzes the important part of the whole system. And through a certain image processing technology to extract the features of the object for further analysis and research. The liquid column, scale and figure of the thermometer are extracted from the image of the thermometer. Finally, the indicator number of the thermometer is calculated. The main contents of this paper are as follows: the automatic recognition system of thermometer based on machine vision is analyzed and studied, and the processing side of the collected image and the realization of the constant temperature device of the system are briefly described. The related parameters of the industrial camera are analyzed, and the industrial camera is calibrated by Zhang Zhengyou, which makes the collected image without distortion interference. According to the automatic recognition system of thermometer designed before, the field simulation experiment was carried out to adjust the different temperature of the constant temperature water bath device, and to collect the image by the industrial camera after the temperature was stabilized. The MATLAB software platform is used to process the collected images. The process of processing includes: noise reduction and noise removal; image contrast stretching to highlight the thermometer part of the image; image segmentation, and image segmentation. The feature extraction is used to extract the liquid column and the scale line of the thermometer; the tilt correction of the image is made to improve the recognition accuracy of the thermometer; and the number on the thermometer is segmented and extracted and then the template matching is carried out to identify the thermometer. The thermometer is calculated by image processing before, and the corresponding user interface is designed by using Visual Studio and HALCON machine vision software to display image and result information.
【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】：TH811;TP391.41
本文编号：2251133
[Abstract]:Glass liquid thermometer is widely used in various fields because of its high temperature measuring accuracy and convenient carrying, so the use of glass liquid thermometer is relatively large. The glass liquid thermometer needs regular verification, and the local metrology department is responsible for the verification work, resulting in the heavy task of the measurement department. It is found that most measurement departments calibrate glass thermometers by manual reading. The way of manual reading is easy to lead to the following results: the excessive number of thermometers makes the verification result inaccurate, the subjective factors of the calibrator influence the reading of the thermometer, and the verification result is inaccurate; The speed of verification is slow due to the failure to store the data of arbitrary thermometers in time. Therefore, it is necessary to study the automatic verification of glass liquid thermometer. With the development of science and technology, especially the development of intelligence and automation, the automatic verification of glass thermometers by machine vision has been paid more and more attention by many researchers. On the basis of machine vision, this paper uses a constant temperature water bath pot to keep the thermometer temperature stable, obtains the thermometer image by a certain resolution industrial camera, and analyzes the important part of the whole system. And through a certain image processing technology to extract the features of the object for further analysis and research. The liquid column, scale and figure of the thermometer are extracted from the image of the thermometer. Finally, the indicator number of the thermometer is calculated. The main contents of this paper are as follows: the automatic recognition system of thermometer based on machine vision is analyzed and studied, and the processing side of the collected image and the realization of the constant temperature device of the system are briefly described. The related parameters of the industrial camera are analyzed, and the industrial camera is calibrated by Zhang Zhengyou, which makes the collected image without distortion interference. According to the automatic recognition system of thermometer designed before, the field simulation experiment was carried out to adjust the different temperature of the constant temperature water bath device, and to collect the image by the industrial camera after the temperature was stabilized. The MATLAB software platform is used to process the collected images. The process of processing includes: noise reduction and noise removal; image contrast stretching to highlight the thermometer part of the image; image segmentation, and image segmentation. The feature extraction is used to extract the liquid column and the scale line of the thermometer; the tilt correction of the image is made to improve the recognition accuracy of the thermometer; and the number on the thermometer is segmented and extracted and then the template matching is carried out to identify the thermometer. The thermometer is calculated by image processing before, and the corresponding user interface is designed by using Visual Studio and HALCON machine vision software to display image and result information.
【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】：TH811;TP391.41
【参考文献】
相关期刊论文 前10条
1 苑玮琦;李坤鹏;;玻璃管温度计示值视觉检测系统研制[J];化工自动化及仪表;2016年01期
2 魏莹;;高斯滤波OpenMP并行化[J];通讯世界;2015年10期
3 张宏伟;成宝芝;;基于MATLAB图像边缘检测方法的研究[J];大庆师范学院学报;2015年03期
4 任彪;;机器视觉中立视技术的发展与现状[J];信息系统工程;2014年11期
5 黎小琴;;Halcon软件在机器视觉课程实验教学中的应用[J];科技资讯;2014年30期
6 辛光泽;侯宏录;李飞;齐晶晶;;基于Camera Link接口的高速视频图像采集系统[J];国外电子测量技术;2014年03期
7 吴进;;机器视觉中快速模版匹配算法研究[J];新型工业化;2014年01期
8 刘伟;;一种基于智能相机的机器视觉系统的设计[J];重庆工商大学学报(自然科学版);2013年11期
9 董锐;阮社;穆蕾;尹跃;李颖;;玻璃液体温度计计算机辅助检定系统原理简述和思路拓展[J];现代制造技术与装备;2013年04期
10 张法全;王国富;曾庆宁;叶金才;;利用重心原理的图像目标最小外接矩形快速算法[J];红外与激光工程;2013年05期
，本文编号：2251133
本文链接：https://www.wllwen.com/kejilunwen/ruanjiangongchenglunwen/2251133.html
