The Martian Boulder Automatic Recognition System, MBARS

Abstract

Boulder‐sized clasts are common on the surface of Mars, and many are sufficiently large to be resolved by the high resolution imaging science experiment (HiRISE) camera aboard the Mars reconnaissance orbiter. The size, number, and location of boulders on the surface and their spatial distribution can reveal the processes that have operated on the surface, including boulder erosion, burial, impact excavation, and other mechanisms of boulder transport and generation. However, quantitative analysis of statistically significant boulder populations, which could inform these processes, entails prohibitively laborious manual segmentation, granulometry, and morphometry measurements over large areas. Here, we develop, describe, and validate an automated tool to locate and measure boulders on the Martian surface: the Martian Boulder Automatic Recognition System (MBARS). Our open‐source Python‐based toolkit automatically measures boulder diameter and height in HiRISE images enabling rapid and accurate assessments of boulder populations. We compare our algorithm with existing boulder‐counting methods, manual analyses, and objects of known size to verify accuracy and precision. Additionally, we test how MBARS quantitatively characterizes boulders around an impact crater in the Martian northern lowlands. We compare this to previous work on rock excavation during impact cratering using manually counted boulders around lunar craters.

Automatic Characterization of Boulders on Planetary Surfaces From High‐Resolution Satellite Images

“…The comparison is made against manually outlined boulders mapped on a small 160 × 160 m portion of HiRISE image TRA_000828_2495, the test area over which Hood et al. (2022) compared their approach to the NM algorithm (Figure 11).…”

“…The prediction shapefile for the NM method was obtained using both a conservative and a more liberal brightness thresholds, which was selected by Hood et al. (2022). Upon visual inspection, predictions made using a conservative threshold were found to yield better results and is thus used here for our comparison.…”

“…Finally, we compare the performance of BoulderNet with that of two other existing rule-based boulder detection algorithms-the Nagle-McNaughton (NM) GIS-based boulder detection algorithm (Nagle-McNaughton et al, 2020) and the Martian Boulder Automatic Recognition System (MBARS; Hood et al, 2022). The comparison is made against manually outlined boulders mapped on a small 160 × 160 m portion of HiRISE image TRA_000828_2495, the test area over which Hood et al (2022) compared their approach to the NM algorithm (Figure 11).…”

“…A first family of models relies on rule-based approaches, often detecting the shadows cast by boulders rather than the boulders themselves (Golombek et al, 2008;Hood et al, 2022;B. Wu et al, 2018).…”

“…Wu et al, 2018). In some of these algorithms, the outlines of shadows are extracted using different segmentation techniques before being fitted to an ellipse, from which the height and diameter of an equivalent cylindrical (Golombek et al, 2012) or spheroidal (Hood et al, 2022) boulder are reconstructed using various geometric assumptions. Similarly, the detection routine of B.…”