Electric Power Fuse Identification with Deep Learning

As part of arc flash studies, survey pictures electrical installations need to be manually analyzed. A challenging task is identify fuse types, which can determined from physical characteristics such as shape, color and size. To automate this process using deep learning techniques, a new dataset past projects data the web was created. Multiple experiments were performed train final model, reaching an average precision (AP $_{50}$ ) 91.06 % on holdout set, confirms its potential for identification types in photos. By identifying only, take clear label text eliminated, allowing taken away danger, thereby improving safety workers. All resources needed repeat are openly accessible, including code datasets.