基于虚拟传感器的结构位移视觉测量误差抑制方法

Virtual sensor-based error suppression method for visual measurement of structural displacement

温度、热浪和振动等环境因素给结构位移监测造成不可忽视的误差。虽然利用静止标志物降低视觉测量误差的原理可行,但是其适用范围却因背景标志物不易获得和控制而受到很大限制。利用视觉测量系统的现有条件,在近场布置主动光源作为误差检测虚拟传感器,摆脱对天然标志物的依赖。针对焦外成像产生的弥散圆等伴生问题,采用图像滤波、伽马变换、图像质量评估、图像二值化、边缘检测和目标检测等算法,形成一套面向弥散圆光斑的自适应图像增强算法;利用虚拟传感器的虚假位移,实现结构位移测量误差补偿,开展了仿真实验和室外结构位移视觉测量实验。研究结果表明:该方法可以显著降低结构位移的初始误差,残留误差满足测量精度要求。

Environmental factors such as temperature,heat waves,and vibration cause non-negligible errors in structural displacement monitoring.The principle of using stationary markers to suppress visual measurement errors is feasible,but its applicability is greatly limited by the fact that background markers are not easy to obtain and control.With the readily available conditions of the visual measurement system,an active light source is arranged in the near field as a virtual sensor for error detection,freeing it from the dependence on natural markers.For the accompanying problems such as diffuse circles generated by out-of-focus imaging,a set of adaptive image enhancement algorithms oriented to diffuse circle spots is formed by using image filtering,gamma transformation,image quality assessment,image binarization,edge detection,and target detection algorithms.The false displacement of the virtual sensor is used to achieve structural displacement measurement error compensation.Simulation tests and outdoor structural displacement visual measurement tests are carried out,and the results show that the method can significantly reduce the initial error of structural displacement and the residual error meets the measurement accuracy requirements.