【摘要】 随着全球气候的变化，部分城市时常发生极端降水事件。这将加剧城市内涝灾害的发生。目前有研究学者利用目标检测算法来测量城市内涝的积水深度。该种方法的关键在于对积水参照物的准确识别。本文选用You Only Look Once version5目标检测算法进行识别，而后通过改进注意力机制来提高检测准确性。实验结果显示，改进后的模型检测精度提高6.27%，召回率提高3%,mAP值提高2.31%。更多还原

【Abstract】 With the global climate change, some cities are experiencing extreme precipitation events from time to time, which will intensify the occurrence of urban flooding disasters. Currently, some researchers use target detection algorithms to measure the depth of waterlogging in cities, and the key to this method is the accurate identification of waterlogging references. In this paper, the YOLOv5(You Only Look Once version 5) target detection algorithm is used for identification, and then the attention mechanism is improved to improve the detection accuracy. The experimental results show that the improved model improves the detection accuracy by 6.27%, the recall rate by 3%, and the mAP(mean Average Precision) value by 2.31%.更多还原