Reports indicate that high-cost, insecurity, and difficulty in complex environments hinder the traditional urban road inundation monitoring approach. This work proposed an automatic monitoring method for experimental urban road inundation based on the YOLOv2 deep learning framework. The proposed method is an affordable, secure, with high accuracy rates in urban road inundation evaluation. The automatic detection of experimental urban road inundation was carried out under both dry and wet conditions on roads in the study area with a scale of a few m². The validation average accuracy rate of the model was high with 90.1% inundation detection, while its training average accuracy rate was 96.1%. This indicated that the model has effective performance with high detection accuracy and recognition ability. Besides, the inundated water area of the experimental inundation region and the real road inundation region in the images was computed, showing that the relative errors of the measured area and the computed area were less than 20%. The results indicated that the proposed method can provide reliable inundation area evaluation. Therefore, our findings provide an effective guide in the management of urban floods and urban flood-warning, as well as systematic validation data for hydrologic and hydrodynamic models.

报告表明，复杂环境中的高成本、不安全性和困难性阻碍了传统的城市道路淹没监测方法。本工作提出了一种基于YOLOv2深度学习框架的实验性城市道路淹没自动监测方法。所提出的方法在城市道路淹没评估中是一种经济实惠、安全且准确率高的方法。在研究区道路干湿条件下，以几米²为尺度进行了试验性城市道路淹没的自动检测. 该模型的验证平均准确率较高，淹没检测率为90.1%，而其训练平均准确率为96.1%。这表明该模型具有有效的性能，具有较高的检测精度和识别能力。此外，计算了图像中实验淹没区和实际道路淹没区的淹没水域面积，表明实测面积与计算面积的相对误差小于20%。结果表明，所提出的方法可以提供可靠的淹没区评估。因此，我们的研究结果为城​​市洪水管理和城市洪水预警提供了有效的指导，并为水文和水动力模型提供了系统的验证数据。