The service can assess who is responsible for the accident in a short period of time. (image: GIST)
GWANGJU, Dec. 26 (Korea Bizwire) — South Korean researchers have developed a system that can analyze black box videos with artificial intelligence (AI) to determine who is responsible for traffic accidents.
The AI accident assessment system developed by a research team from the Gwangju Institute of Science and Technology (GIST) does not involve humans directly, but the AI learns past data and makes final judgments about the accident itself.
In order to minimize human labor, data used for AI network learning can be utilized as video sources.
In particular, there is no need for separate processing such as labeling, which takes the most time and money to improve quality or develop AI networks.
Therefore, the service can assess who is responsible for the accident in a short period of time.
In addition to “car-to-car” accidents in the future, the research team plans to expand the functions of the AI network to allow analysis of various accidents, such as “car-to-person” and “car-to-bicycle.”
“In addition to black boxes, we expect that the technology will be used for legal services through the recent increase in video recordings such as CCTV analysis, which will be the basis for analyzing images of AI laws,” said Lee Yong-koo, the professor at GIST who led the research.
M. H. Lee (mhlee@koreabizwire.com)
