相机链式路基沉降监测系统通过图像式测量的方法可以实现高精度、非接触的铁路路基沉降监测。图像式测量的精密性和灵敏性对准确快速评估系统性能提出了要求，需要变定期维护为动态维护，提出了利用光斑图像质量反映系统性能的策略。引入了边缘清晰度、相对亮度、图像信息熵、光线垂直程度4个参数用以评估光斑质量，应用层次分析法和熵权法确定上述参数所占权重，构建了光斑质量综合评价模型，完成了模拟实验和现场实验。模拟实验结果表明，该模型能很好地跟踪光斑图像质量分数变化，具有较好的灵敏性；现场实验结果表明，该模型评价光斑质量结果准确，分析得出设置质量分数阈值为0.95，定位误差小于1 mm（沉降测量规范中允许误差范围）。当光斑图像质量分数持续未达到阈值且超过静置时间时，系统需要维护，实现了系统状态实时监测以及动态维护功能。

Monitoring and measurement systems for railway subgrade settlement using a chain of cameras can realize high-precision, non-contact railway subgrade settlement monitoring with the image measurement method. The precision and sensitivity of image measurement requires accurate and rapid evaluation of system performance, and regular maintenance should be changed to dynamic maintenance. Herein we propose a strategy for determining system performance using the spot-image quality and introduce four parameters, viz. edge definition, relative brightness, image information entropy, and light vertical degree, to evaluate the quality of light spots. The weight of each parameter can be constructed using the analytic hierarchy process and the entropy method. A comprehensive evaluation model for spot quality is constructed. Simulation and field experiments are conducted. The simulation shows that the proposed model can track changes in the spot-image quality score with good sensitivity. The field experiment shows that this model can be used to accurately evaluate spot quality. In addition, when the quality score is set at 0.95, the positioning error exceeds 1 mm (the allowable error range in subsidence for settlement measurement). The system requires maintenance when the light spot-image quality score continuously misses the threshold and exceeds the resting time. This method realizes system-state real-time monitoring and dynamic maintenance.