Recently, Wangjiang Street in Shangcheng District officially launched the AI intelligent visual recognition system for climbing over the river embankment, which timely judges the people who have climbed over the guardrail and accidentally fallen into the water, and issued a background warning. What does AI use to make judgments? What are the advantages over manual inspections? Let's find out.
Wujiang Road Zhijiang intersection along the river and down the river
Approaching the intersection of Wujiang Road and Zhijiang Road, you can see a locked lower river point on the railing, the traces of rewelding and the crease of the break are obvious, and the pointed end of the railing has been broken and replaced by layers of wrapped cotton cloth.
Did the people who went down the river climb down from here to swim? "Yes, 10 people climb over the embankment every day, and they often come again tomorrow after the persuasion," said a street volunteer. The parasols and small stools on the steps down the river are simple 'changing rooms' that they have made. It is very dangerous to climb over the river embankment to swim and watch the tide without permission, and it is even more dangerous now that it is the flood season. ”
Looking along Wangjiang Road to the embankment of the three bridges, about 1.5 kilometers long, this is the jurisdiction of Wangjiang Street, along the river, "tide danger, do not go down the embankment" warning signs are clearly visible, the staff said, over the embankment AI intelligent visual recognition system relies on the roadside 4.5 meters high spherical camera, through the linkage screening, accurate identification and other algorithm technology iteration, with "computing power" for "manpower", greatly improving the efficiency of monitoring and early warning.
At the beginning of June, the Wangjiang Street Integrated Information Command Room received a "shout" from the AI algorithm. A man struggles in the middle of the river, and the AI system captures the situation for the first time. The shift leader responded quickly and arrived at the scene within 5 minutes with the water police station and medical emergency forces. Thanks to the timely warning, the people who fell into the water were quickly rescued and their lives were saved.
It is understood that AI carries out intelligent monitoring through video perception, big data models and other technologies, providing a pair of tireless "eyes" for the safety of the river embankment, whether it is an adventurer who climbs over the guardrail or an unfortunate person who accidentally falls into the water, AI can quickly make a judgment and issue an early warning to the background, solving the dilemma of "the river embankment point is long and long, and it is difficult to find and deal with it by manpower".
"In the past, we organized daily inspections by IMIS team members and civilized persuasion by volunteers, and it took an average of 15 to 20 minutes to complete a circle. Now, with the help of AI, we can find it in time, dissuade and deal with it through video shouting, manpower and other methods, and effectively ensure the safety of the masses. The relevant person in charge of the Wangjiang Street Regional Development and Governance Center said.
Street comprehensive management team members and volunteers patrol along the Qiantang River
The "eyes" are more flexible, the "mind" is clearer, and behind the change of passive disposal to active early warning, it is a useful exploration of the Shangcheng District Social Governance Center to create a "video consultation and joint command" scenario. "We have opened up the individual equipment and video surveillance issued by different departments to realize the resonance of the equipment at the same frequency, which not only solves the problems of multiple incidents, frequent contradictions, and untimely response scheduling, but also promotes the handling of incidents through multi-party video negotiation and remote accurate guidance, and realizes the three-level online linkage of districts, streets, and communities. The relevant person in charge of the Emergency Command Section of the Shangcheng District Social Governance Center said.
At present, Wangjiang and Ziyang streets have begun to pilot, and Sijiqing, Pengbu, Jiubao and other streets along the river will also be put into use. In the future, Shangcheng District will accelerate the integration of IoT devices, strengthen the monitoring of key indicators such as rainfall and water level, use big data and intelligent algorithms to predict flood risks, issue early warnings in a timely manner, and help relevant departments and the public take measures in advance to effectively reduce disaster losses and achieve early warning and early prevention.