Automating and scaling personalized safety training using eye-tracking data

Abstract Research has shown that a large proportion of hazards remain unrecognized, which expose construction workers to unanticipated safety risks. Recent studies have also found that a strong correlation exists between viewing patterns of workers, captured using eye-tracking devices, and their hazard recognition performance. Therefore, it is important to analyze the viewing patterns of workers to gain a better understanding of their hazard recognition performance. From the training standpoint, scan paths and attention maps, generated using eye-tracking technology, can be used effectively to provide personalized and focused feedback to workers. Such feedback is used to communicate the search process deficiency to workers in order to trigger self-reflection and subsequently improve their hazard recognition performance. This paper proposes a computer vision-based method that tracks workers on a construction site and automatically locates their fixation points, collected using a wearable eye-tracker, on a 3D point cloud. This data is then used to analyze their viewing behavior and compute their attention distribution. The presented case studies validate the proposed method.