We introduce a domain-speci c and late-fusion algorithm to cope with the challenge raised in The MediaEval 2017 Multimedia Satel- lite Task. Several known techniques are integrated based on domain- speci c criteria such as late fusion, tuning, ensemble learning, object detection using deep learning, and temporal-spatial-based event con rmation. Experimental results show that the proposed algo- rithm can overcome the main challenges of the proper discrimina- tion of the water levels in di erent areas as well as the consideration of di erent types of ooding events.