补偿(心理学)
计算机科学
校准
热的
结构光
观测误差
轮廓仪
点(几何)
温度测量
过程(计算)
测量不确定度
投影(关系代数)
控制理论(社会学)
算法
数学
计算机视觉
机械工程
物理
工程类
人工智能
统计
操作系统
几何学
气象学
表面光洁度
量子力学
心理学
控制(管理)
精神分析
作者
Chenwei Hsu,Gang Rao,Chengye He,Jing Xu
标识
DOI:10.1109/cyber53097.2021.9588241
摘要
Fringe Projection Profilometry (FPP) is considered to be one of the most reliable and accurate 3D measurement methods. However, in many applications, changes in ambient temperature or self-heating effects will cause different components expansion and contraction in optical systems. Owing to the the fact that parameters of structured light system are usually considered to be fixed after calibrated, the precision of high-accuracy measurement will be greatly reduced if the influence related to thermal error is ignored. In this study, we analyze the thermal error model of the structured light system, and verify our theory by simulation and experiment. Secondly, we propose an improved calibration method, based on traditional method with an additional virtual estimation points in the calibration process. The parameters are compensated at each temperature, and the experiments demonstrated that the thermal error can be greatly reduced after compensation, and the changes of parameters tend to be linear, it is in accordance with the physical phenomenon. Finally, our proposed method could effective compensate the error caused by temperature drift by over 85%.
科研通智能强力驱动
Strongly Powered by AbleSci AI