Metal corrosion is important in the fields of biomedicine as well as construction and transportation etc. While most corrosion occurs inhomogeneously, there is so far no satisfactory parameter to characterize corrosion inhomogeneity. Herein, we employ the Poisson raindrop question to model the corrosion process and derive an equation to relate corrosion coverage and corrosion mass. We also suggest corrosion mass at 50% coverage, termed as half-coverage mass Mcorro50%, as an inhomogeneity parameter to quantify corrosion inhomogeneity. The equation is confirmed and the half-coverage mass Mcorro50%, is justified in our experiments of iron corrosion in five aqueous media, normal saline (NS), phosphate buffered saline (PBS), Hank’s solution (HS), deionized water (DI), and artificial seawater (AS), where the former three ones are biomimetic and very important in studies of biomedical materials. The half-coverage mass Mcorro50%, is proved to be more comprehensive and mathematically convergent than the traditional pitting factor. Iron corrosion is detected using visual observation, scanning electron microscopy with a build-in energy dispersive spectrometer, inductive coupled plasma emission spectrometry, and electrochemical measurements. Both rates and inhomogeneity extents of iron corrosion are compared among the five aqueous media. The factors underlying the medium effects on corrosion rate and inhomogeneity are discussed and interpreted. Corrosion rates of iron in the five media differ about 7 folds, and half-coverage mass values differ about 300,000 folds. The fastest corrosion and the most significant inhomogeneity occur both in biomimetic media, but not the same one. The new equation and the new quantity are stimulating for dealing with a dynamic and stochastic process with inhomogeneity including but not limited to metal corrosion. The findings are meaningful for research and development of biomaterials.