상세정보
학술대회
A Risk Estimation Approach based on Deep Learning in Shipbuilding Industry
- 발행일
- 201910
- 출처
- International Conference on Information and Communication Technology Convergence (ICTC) 2019, pp.1438-1441
- 협약과제
- 19ZS1300, 주력 산업 고도화를 위한 지능형 상황인지 기반 기술 개발, 김도현
- 초록
- Shipbuilding industry is one of the most hazardous industries. To reduce accidents, various safety policies and practices have been established and recommended to obey. Despite workers being made aware of risk associated with not following these practices, many workers do not follow these practices for various reasons such as inconvenience of wearing a personal safety equipment, increase of cost, insufficient working time, and so on. In addition, there are many cases that workers carry out a task or pass through without knowing risk of the task or risk of the area to be worked. It is difficult for each individual worker to know various surrounding circumstances, and there are also limitations in that safety supervisors play a role of safety management for all worksites. Therefore, we propose an automated risk estimation approach to support identifying hazardous zones and estimating risk by verifying whether safety measures are performed for the identified hazardous zone based on deep learning method.
- KSP 제안 키워드
- Carry out, Deep learning method, Personal safety, Risk estimation, Safety measures, Shipbuilding industry, Working time, deep learning(DL), safety management