Development of a human performance model as a baseline for automatic change detection software capabilities in mine warfare

摘要

This study focused on the development of a human performance model as a baseline performance capability for automatic change detection software for use in mine warfare. Through a series of survey images, operator performance was observed under a variety of sonar image conditions, including increasing clutter levels and changes in image altitude and orientation. While a rough model was developed utilizing only the physical attributes of the images, to obtain a close fit between the model and actual observations, the variability of personal proficiency was included in the final model. The inclusion of this parameter greatly improved model accuracy and highlights the need to better understand differences between operator performances in mine warfare planning.