摘要
为减小图像式路基沉降监测系统中温度变化对测量结果的影响,通过对其核心设备—CMOS工业相机的温度漂移进行研究,提出一种针对该监测系统的温度补偿方法。首先,分析说明了温度变化会对监测系统测量精度造成影响;然后,建立漂移参数补偿模型并设置实验进行验证。漂移参数补偿模型使相机的像点误差从1.0~2.5 pixel降低到0.1~0.4 pixel,验证了该模型的正确性;同时设置对比实验,与系统辨识的方法相比,两种方案补偿效果接近,但提出的补偿方法过程简单,更具实用性;最后,提出路基沉降监测系统的温度补偿方法并验证。结果表明,该方法能够提高监测系统测量精度。所提补偿方法,对于提高图像式路基沉降监测系统的测量精度具有重要意义。
To reduce the influence of temperature change on the measurement results in the image-based roadbed settlement monitoring system,this article proposes a temperature compensation method for this monitoring system by studying the temperature drift of its core device—CMOS industrial camera.First,it is analyzed and explained that the temperature variation can affect the measurement accuracy of the monitoring system.Then,a drift parameter compensation model is formulated and experiments are set up for evaluation.Experimental results show that the drift parameter compensation model reduces the image point error of the camera from 1.0~2.5 pixel to 0.1~0.4 pixel,which shows the correctness of the model.Meanwhile,a comparison experiment is set up,and the compensation effect of the drift parameter model schemes is close with the compared method of system identification.But,the proposed compensation method has a simple process and is more practical.Finally,the temperature compensation method of the roadbed settlement monitoring system is proposed and evaluated.Results show that the method can improve the measurement accuracy of the monitoring system.The proposed compensation method is of great significance to improve the measurement accuracy of the image-based roadbed settlement monitoring system.
作者
杨锋
闵永智
孙天放
Yang Feng;Min Yongzhi;Sun Tianfang(School of Automation and Electrical Engineering,Lanzhou Jiaotong University,Lanzhou 730070,China)
出处
《仪器仪表学报》
EI
CAS
CSCD
北大核心
2022年第8期18-25,共8页
Chinese Journal of Scientific Instrument
基金
国家自然科学基金(61663022,6206020128)
甘肃省自然科学基金(18JR3RA105)项目资助
关键词
沉降监测
CMOS工业相机
漂移参数
温度补偿
settlement monitoring
CMOS industrial cameras
drift parameters
temperature compensation
作者简介
杨锋,2018年于兰州交通大学获得学士学位,现为兰州交通大学硕士研究生,主要研究方向为机器视觉与数据处理。E-mail:750742910@qq.com;通信作者:闵永智,2004年于武汉大学获得硕士学位,2014年于兰州交通大学获得博士学位,现为兰州交通大学教授、博士生导师,主要研究方向为图像处理、机器视觉与人工智能测试。E-mail:minyongzhi@mail.lzjtu.cn
