Computer-aided High-efficiency Detection of Fracture Initiation Sites in Steel

Finding the fracture initiation sites is critical for understanding mechanism and thus improving products’ quality, whereas some hard-to-detected features are easily missed in human-effort-based characterizations with human eye involve lots of labor force. In this study, computer-aided detection proposed to augment expertise efficiently find sites, reduce cost. With a deep-learning you only look once object detector, steel were successfully detected study. Furthermore, based on trained model, an easy-to-use application detecting has been further developed, exhibiting great potential high-efficiency sites.