ROCKSTER (Rock Segmentation Through Edge Regrouping) is a rock detection algorithm that analyzes 2D geologic scenes and identifies rocks and other targets of interest. A multicore ROCKSTER enables long-range autonomous rover traverse science to be performed efficiently and to make use of multicore or parallel computing capabilities.
ROCKSTER initially locates partial boundary contours of rocks using a procedure similar to the well-known Canny edge detector. In particular, an intensity gradient is calculated over the image; ridges in the intensity gradient are linked together using non-maximum suppression, hysteresis thresholding, and edge-following, yielding a set of raw contours.
This initial set of contours must then be further processed to provide a usable segmentation of rocks from the background. ROCKSTER performs this step by splitting the initial contours into low-curvature fragments. Potential T-junctions that were missed by the edge detector are identified and used to split fragments further into even smaller pieces. A gap-filling mechanism is then applied to add new contour fragments between existing fragment end-points. The final step is to regroup the edge fragments into coherent contours, which is accomplished through background flooding. Conceptually, water is poured into the image from the sides, but the water is not allowed to cross over any edge fragments; thus, regions that are totally enclosed by edge fragments remain “dry” while other areas become “wet.” Extracting contours around the dry areas yields the final rock segmentation.
Multicore ROCKSTER provides data parallelization of ROCKSTER by first dividing the image to analyze into tiles or strips. ROCKSTER is then run on each image tile using a separate processor core. Once all cores finish, rock detections are aggregated.
This work was done by Benjamin J. Bornstein, Paul L. Springer, Bradley J. Clement, Tara A. Estlin, and William K. Reinholtz of Caltech for NASA’s Jet Propulsion Laboratory.
This Brief includes a Technical Support Package (TSP).
(reference NPO47880) is currently available for download from the TSP library.
Don't have an account? Sign up here.