Explore how to integrate satellite and drone imagery into data science workflows in this 25-minute talk. Learn about the potential of overhead imagery as unstructured data for machine learning applications. Discover RasterFrames, an extension of Spark SQL with a Python API, designed for processing spatial image data. Understand the fundamentals of making sense of imagery data and how RasterFrames' custom DataSource leverages trends in image publishing. See how RasterFrames integrates with Spark SQL, enabling users to incorporate imagery and location-aware datasets into existing data pipelines. Gain insights into RasterFrames' compatibility with Spark ML, TensorFlow, Keras, and PyTorch. Examine a practical data science case study using overhead imagery in PySpark to solidify your understanding of these concepts.