Automation using deep learning methods is a useful alternative to manual methods of petrographic segmentation, but often requires user familiarity with coding and/or algorithms. We examine the Dragonfly^TM program's deep learning tools for application by users with a variety of skill levels as a method for petrographic image segmentation. An image processing methodology, bimodal image stacking, was created for low-input-data, high-efficacy training of models which can then be applied to varied samples. Using backscatter electron images we show that the resulting model segmentations agree with manual segmentation total and modal crystallinity values within 5%, and calculated plagioclase crystal size distribution (CSD) values within 2σ, despite limitations in discriminating mafic phases. Model creation and training takes <24 hours, 1–3 hours of which are supervised, and the resultant model can then be applied to new uncharacterized samples in <15 minutes per image. This allows for non-experts to create and utilize deep learning models to segment images of variable brightness and texture, at low user-time cost and resulting in size and shape data which are within uncertainty of manual segmentation. While some limitations are noted (for example, sieve-textured phases may need manual correction, and different minerals with similar BSE intensity may not be resolved as separate phases), this methodology can be utilized for general application of models to wide ranges of volcanic crystalline and bubble textures, and to create a library of models for rapid petrological analysis during volcanic eruptions.