Mapping ”Brain Coral” Regions on Mars using Deep Learning

One of the key drivers of the Mars Exploration Program is the search for evidence of past or present life. In this context, the most relevant martian environments to search for extant life are those associated with liquid water, and locations that experience periodic thawing of near-surface ice. In this work, we use convolutional neural networks to detect surface regions containing ``Brain Coral'' terrain, a landform on Mars whose similarity in morphology and scale to sorted stone circles on Earth suggests that it may have formed as a consequence of freeze/thaw cycles. We use high-resolution images ($\sim$100-1000 megapixels) acquired by the Mars Reconnaissance Orbiter to search for these landforms at resolutions close to a few tens of centimeters per pixel (25--50 cm). Over 52,000 images ($\sim$28 TB) were searched ($\sim$5\% of the Martian surface) where we find detections in over 150 images. To expedite the processing we leverage a classifier network (prior to segmentation) in the Fourier domain that can take advantage of JPEG compression by leveraging blocks of coefficients from a discrete cosine transform in lieu of decoding the entire image at the full spatial resolution. The hybrid pipeline approach maintains $\sim$96$\%$ accuracy while cutting down on $\sim$80$\%$ of the total processing time compared to running the segmentation network at the full resolution on every image. The timely processing of big data sets helps inform mission operations, and geologic surveys to prioritize candidate landing sites, avoid hazardous areas or map the spatial extent of specific terrain.