Image Recognition and Early Warning System of Urban Waterlogging Based on Tensorflow

In recent years, heavy rainfall caused by low-lying roads, under-worn overpasses and tunnels to produce a large amount of water phenomenon occurs from time to time, to people’s travel brought great inconvenience, serious even will cause people’s lives, property and major losses. With the rapid development ofIo T technology and 5G, smart street light poles have become a hub for urban data transmission due to their wide range and integrated functions such as lighting control, environmental monitoring, 5G micro-base stations and meteorological monitoring. Therefore, with the powerful function of smart street light pole, the whole process of rain and sewage supervision can be realized in urban areas. Water conservancy departments can use the system as a whole to grasp the flood situation in the entire urban area, timely drainage scheduling, and can use TensorFlow’s original facilities to achieve early warning information LED screen display, video surveillance, sound column alarm and other functions to remind pedestrians to pay attention to safety.