Could technology have averted the stampede at Itaewon, where 151 people died during a Halloween celebration parade? As South Korea mourns the young lives lost in the Itaewon disaster and the government is looking for answers to prevent such accidents in the future, AI companies with computer vision technology are actively developing solutions.
KIOT, an artificial intelligence ( AI ) startup, announced that it has developed an AI-based crowd-counting solution for roads. KIOT has created an artificial neural network platform that can learn the crowd data through AI to understand the traffic volume and the number of people per unit area. The related technology has been patented under the ‘object recognition and counting method using deep learning AI technology.’
AI crowd counter, a hardware ( HW ) measuring device, is equipped with edge computing technology that installs terminals on-site and distributes data processing. It is faster than the centralized processing method and the large information processing scale. It is paired with a congestion control system that is a software ( SW ) platform. A device also notifies citizens of the congestion level through electronic signs linked to the system. When people gather, it induces dispersion with phrases such as ‘congestion.’
The Ministry of Land, Infrastructure and Transport has been conducting a demonstration project this year. Initially, the product was developed as a measuring device to disperse the crowd in the subway. However, due to the prolonged COVID-19 period, the demand for manpower measurement in the vicinity of public roads and buildings increased. The solution was diversified as the need for a distributed guidance system for outdoor collective manpower grew. In particular, the company explains that it can be used efficiently on narrow sidewalks in Itaewon, where many casualties occurred.
KIOT is a startup established in 2016. It makes AI crowd counters, industrial safety alarms, and more. It also supplies solutions to companies such as Korea Zinc and local governments such as Dalseong-gun, Daegu, and Dobong-gu, Seoul. It is a company headquartered in Daegu. Jaejoon Lee, CEO of KIOT, said, “Platforms that predict the bottleneck of manpower have always been pushed out of the priorities of government smart city projects. It is a proactive safety measure that can be avoided.”