Foliar diseases of maize are a constant threat, and changing climates are leading to expanding distribution of the pathogens that lead to these diseases. Disease phenotyping is an extremely important part of their study, as effects of pathogen growth need to be accurately assessed due to their variable nature. However, many foliar diseases can be difficult to phenotype, including the rusts caused by fungi of the order Pucciniales. These pathogens form small, numerous pustules on their hosts that are typically assessed visually with the aid of a scoring system. One such pathogen, Puccinia sorghi, is the causal agent of common rust of maize, and it can form hundreds of pustules per leaf. As such, counting each pustule is extremely time-consuming and traditional scoring systems are unreliable and subjective. To have a faster, more reliable, reproducible, and quantitative method, we have developed an image-based phenotyping platform for P. sorghi that isolates pustules based on color. Using only a flatbed scanner and a laptop, individual leaves are scanned and characteristics such as total leaf coverage and pustule number, size, and distribution are extracted from the images of inoculated leaves. The images are analyzed in an entirely autonomous manner and the same parameters are applied to each leaf, resulting in unbiased phenotyping between images. Computer vision allows us to identify small but significant changes between experimental conditions, leading to more accurate common rust phenotyping needed for functional genomics studies of maize-P. sorghi interactions.