<p>Decommissioning of the Fukushima Daiichi nuclear power station (NPS) is an unprecedented effort in which the entire plant was contaminated with radioactive material from the accident and radiation dose levels were not low. A multitude of workers of Tokyo Electric Power Company (TEPCO), nuclear reactor manufacturers, construction companies, and their contractors are currently involved in the decommissioning project and are therefore exposed to various health risks. Based on the contamination status, TEPCO indicates it should ensure that workers wear proper personal protective equipment (PPE) (e.g., anorak bodysuits, hard hats, full-face masks) in the most radioactive areas of the plant. Nonetheless, even though workers are trained to stay away from potential dangers, there are still many types of risks that can occur within only a few minutes of carelessness. Besides, TEPCO contracted various tasks to more than 20 companies (prime contractors), each of which, in turn, outsourced some of its tasks to multiple layers of subcontractors. This complex structure may hinder consistent enforcement of occupational health rules. Therefore, an automated monitoring approach is essential to be carried out to help facilitate the safety monitoring task of decommissioning workers to ensure that PPEs are used appropriately.</p><p>This work presents a graph-based scene understanding approach that automatically identifies whether workers at a decommissioning site are using PPE properly. Firstly, the image information representation module adopts lightweight deep learning models to detect objects and individuals, and geometric relationship analysis is deployed to combine deep learning-based object detection and individual detection models for interpretable interpretation and output. Subsequently, regulatory rules documented with natural language sentences are transformed into a calculable and structured graph structure. Finally, taking advantage of scene graph structure, an automated reasoning approach is proposed to integrate the processed regulatory and image information, which deploys graph structure analysis for scene understanding and proper PPE use identification. The experimental results demonstrate that the high precision, high recall and fast speed of the model can effectively detect decommissioning workers' failure use of PPE and can facilitate improved safety inspection and supervision.</p>

A Graph-Based Scene Understanding Approach for Ensuring Proper Use of Personal Protective Equipment at the Decommissioning Site of Fukushima Daiichi Nuclear Power Station

Paper Proceedings: https://asmedigitalcollection.asme.org/ICONE/ICONE28/volume/85246

2021 28th International Conference on Nuclear Engineering (ICONE28)

Supporting the need among researchers worldwide for additional online and multimedia content, ASME now offers the ASME Conference Video Collection. This platform will collect and provide long-term access to conference presentation videos, encompassing all technology areas within ASME's technical conference and events program. About 30,000-40,000 engineers across all mechanical engineering and related disciplines participate in ASME conferences each year. This includes authors of approximately 6,000 to 8,000 technical papers published by ASME, the majority of whom also submit video presentations for our virtual conferences. The video collection will be available as an add-on to existing product packages or as a separate subscription.