Corrosion costs an estimated 3 to 4% of GDP for most nations each year, leading to significant loss of assets. Research regarding automatic corrosion detection is ongoing, with recent progress leveraging advances in deep learning. Research is hindered by the lack of a publicly available dataset; thus, models use locally produced datasets suitable for the immediate conditions, but unable to produce generalized models for corrosion detection. These algorithms will output a considerable number of false positives and false negatives when challenged in the field. In this paper, we present SpotRust, a deep learning-based corrosion detector that performs pixel-level segmentation of corrosion. Moreover, three Bayesian variants of SpotRust are presented that provide uncertainty estimates depicting the confidence levels at each pixel, to better inform decision makers. Experiments are performed on a newly collected dataset consisting of 280 images, we validate our algorithms are accurate and reliable. Code is provided at https://github.com/StuvX/SpotRust.