The effectiveness of monitoring and early-warning systems for ground deformation phenomena, such as sinkholes, depends on their ability to accurately resolve the ongoing ground displacement and detect the subtle deformation preceding catastrophic failures. Sagging sinkholes with a slow subsidence rate and diffuse edges pose a significant challenge for subsidence monitoring due to the low deformation rates and limited lateral strain gradients. In this work, we satisfactorily illustrate the practicality of the Brillouin optical time domain analysis (BOTDA) to measure the spatial-temporal patterns of the vertical displacement in such challenging slow-moving sagging sinkholes. To assess the performance of the approach, we compare the strain recorded by the distributed optical fiber sensor with the vertical displacement measured by high-precision leveling. The results show a good spatial correlation with the ability to identify the maximum subsidence point. There is also a good temporal correlation with the detection of an acceleration phase in the subsidence associated with a flood event.

为解决直线电机动子位置测量中多帧图像位移叠加而产生的误差累积问题，提出一种加阈值变换基准图（TTBM）的机器视觉累积误差消减算法。首先，根据直线电机动子的一维刚体平动特点，生成并优选灰度变化更加平滑的一维散斑图像作为目标拍摄图。其次，采用相位相关算法（PCA）改进灰度梯度方法，对配准图像进行整像素平移后再计算亚像素位移，以增大基准图像与配准图像之间的位移测量范围。最后，提出自调整基准图方法，根据图像位移的测量范围设定基准图像变换阈值，降低位移叠加次数，减小累积误差，实现直线电机长行程精密位移测量。同时，研究了基准图像变换阈值对测量精度的影响，给出了基准图变换的最佳阈值选取范围。仿真和实验表明，所提方法可以有效地减小直线电机动子测量中叠加而产生的累积误差。

为了解决外差干涉相位测量中多通道采样信号间的串扰误差对相位测量精度的影响，提出了一种基于采样信号频谱分析的预补偿方法来实现信号串扰误差的补偿和消除。首先建立基于锁相放大的正交鉴相法的信号串扰误差理论模型，阐明了串扰系数、输入信号幅值比和串扰信号相位偏移对相位测量误差的影响；设计仿真实验验证了该误差模型和补偿方法的有效性；然后基于紧凑型FPGA开发平台设计了相位测量实验，结果表明该补偿方法能够有效消除信号串扰误差的影响，补偿后相位测量的最大误差从0.34°下降到0.01°；最后搭建了外差干涉仪并与高精度的压电位移平台进行比对，实验结果表明补偿后的信号处理系统能够满足外差干涉测量的应用需求。