Abstract: Video surveillance and artificial intelligence technologies are increasingly widely used in the construction of smart water conservancy. Digital twin construction is the key and core of building smart water conservancy. In line with the actual needs of water conservancy digital twin construction, we developed a video intelligent recognition model system using video monitoring, image processing, artificial intelligence and other technical means. With this system, intelligent simulation could interact with physical element videos. We highlight the systematic design of main application scenarios such as water gauge recognition, floating object recognition, surface water body detection and others. According to actual operation, the video intelligent recognition system can monitor, analyze, identify, predict and early warn physical objects continuously, hence has great application value in the intelligent construction of water conservancy.

Key words: artificial intelligence, digital twin, water gauge recognition, floating object recognition, surface water body detection