[Asia Economy Reporter Oh Su-yeon] A service capable of monitoring and detecting dangerous situations even in dark environments or CCTV blind spots has been launched. It detects accidents such as falls without privacy concerns and sends alarms to secure the golden time.
On the 7th, LG Uplus held a demonstration event for 'U+ Smart Radar' at its headquarters located in Yongsan-gu, Seoul. They demonstrated four scenarios: real-time location tracking, real-time posture recognition, dangerous situations, and falls.
When a demonstrator entered the space where the U+ Smart Radar device was installed, a human-shaped pictogram appeared on the administrator's screen. When sitting on a chair or lying on the floor, the pictogram changed to the same posture. This was done by recognizing and analyzing movements using artificial intelligence (AI) algorithms.
When the demonstrator entered a section set as a danger zone, the U+ Smart Radar device emitted an alarm sound, and notification messages were sent to the administrator's PC and mobile phone. When simulating falling off a bed or collapsing, it immediately recognized the situation and alerted the administrator in the same way. Additionally, it provides a three-minute movement history before and after the incident to prevent unnecessary manpower waste such as police dispatch if it is not a serious accident.
The biggest concern when managing safety using existing CCTV is privacy protection. When two people entered the demonstration space, the U+ Smart Radar recognized them as Person 1 and Person 2 without any distinctive features appearing on the pictogram. Since it does not record video but expresses human movements as simple pictograms, it protects personal information and eliminates privacy concerns even in sensitive spaces like public restrooms. An LG Uplus official in charge of the demonstration explained, "It recognizes situations through scanning without video recording."
Utilizing Autonomous Vehicle Radar Sensors... Applied to Subways and Welfare Facilities
U+ Smart Radar applies technology that uses a 77 GHz radar sensor used in autonomous vehicles to monitor accident risk situations more precisely in real time. It detects finely with 98% accuracy even in blind spots and dark environments where CCTV cannot see. It can detect various states such as falls, entries, and movement paths for up to five people gathered in one place.
When a pre-set dangerous situation such as a fall, collapse, or area entry/exit occurs, the radar sensor on the wall detects it, removes noise from the data, and visualizes it. The AI algorithm analyzes the image status and abnormal signs and sends the data to the cloud server.
With a small size of 13 cm width, 13 cm height, and 3.5 cm thickness, it is easy to install and can be applied in various spaces. When installed on a wall about 2 meters high from the floor, it detects movements within 7 meters horizontally and vertically, for up to five people.
Seung-hoon Jeon, Executive Director of Smart Infrastructure Business at LG Uplus, said, "Radar was not used for industrial site detection purposes, but as technology advanced and AI developed, it has evolved to the level of recognizing human states. Radar has the advantage that personal identity is not revealed through point clouds. This is the first commercial case in Korea and the highest level of technology."
Based on this technology, it is highly useful in various facilities. One device can monitor situations occurring in wards, classrooms, etc., making it useful for simultaneously monitoring vulnerable groups. For example, if fall movements are detected in care facilities or homes where elderly people live alone, it watches for posture recovery and sends alarms if abnormal signs occur. It instantly detects emergencies in the homes of disabled or elderly people living alone and provides feedback on occupancy status and location information to prevent safety accidents.
© The Asia Business Daily(www.asiae.co.kr). All rights reserved.
