AI-based visual tech to be applied to catch fly tippers in Korea

2018.11.20 16:26:49 | 2018.11.20 16:27:22

이미지 확대
A visual-based artificial intelligence (AI) technology developed by South Korea’s Electronics and Telecommunications Research Institute (ETRI) will be deployed on CCTV cameras to better catch out fly tippers

ETRI announced on Tuesday that it will use its visual AI technology, “Deep View,” to strictly control litterbugs in city areas. The technology helps precisely track down movements of people placing down or throwing objects as well as those illegally throwing away rubbish. Applying the technology in the future will help proactively detect and prevent various crimes and incidents from happening in city areas.

So far, there have been much difficulty in the academic circle in recognizing actions appearing in CCTVs as studies on action comprehension utilized widely-used data such as sports or YouTube videos. As part of efforts to resolve such difficulty, ETRI focused developed on action comprehension technology that can operate on actual environment by pulling together necessary functions, requirements, and data with local governments.

ETRI focused research on common actions and behaviors, particularly on detecting illegal fly-tipping activity in city areas. The research team used deep learning-based recognition technology to detect joint location of passersby as well as objects they are holding and develop modeling between the passersby and the object to detect illegal dumping activity.

By Won Ho-sup and Lee Eun-joo

[ⓒ Pulse by Maeil Business Newspaper & mk.co.kr, All rights reserved]