Subkilometer crater discovery with boosting and transfer learning

Abstract

Counting craters in remotely sensed images is the only tool that provides relative dating of remote planetary surfaces. Surveying craters requires counting a large amount of small subkilometer craters, which calls for highly efficient automatic crater detection. In this article, we present an integrated framework on autodetection of subkilometer craters with boosting and transfer learning. The framework contains three key components. First, we utilize mathematical morphology to efficiently identify crater candidates, the regions of an image that can potentially contain craters. Only those regions occupying relatively small portions of the original image are the subjects of further processing. Second, we extract and select image texture features, in combination with supervised boosting ensemble learning algorithms, to accurately classify crater candidates into craters and noncraters. Third, we integrate transfer learning into boosting, to enhance detection performance in the regions where surface morphology differs from what is characterized by the training set. Our framework is evaluated on a large test image of 37, 500 × 56, 250 m 2 on Mars, which exhibits a heavily cratered Martian terrain characterized by nonuniform surface morphology. Empirical studies demonstrate that the proposed crater detection framework can achieve an F1 score above 0.85, a significant improvement over the other crater detection algorithms.

Online Feature Selection with Streaming Features

“…There is a very extensive, near-global coverage of the Martian surface with high resolution planetary images. A portion of the High Resolution Stereo Camera (HRSC) nadir panchromatic image h0905 is selected, taken by the Mars Express spacecraft, to serve as the test set [11]. The selected image has a resolution of 12.5 meters/pixel and a size of 3,000 by 4,500 pixels (37,500×56,250m 2 ).…”

“…The LARS algorithm is an embedded feature selection method recently introduced to handle classification or regression problems by using optimization with specified loss and penalty functions. The Naïve boosting algorithm was proposed by Ding et al [11]; it integrates the boosting algorithm and greedy feature selection algorithms for crater detection. In Tables 9 and 10, we report the prediction accuracies of all methods on the three regions.…”

“…More specially, Mars crater detection from high resolution planetary images is an important task in planetary research because it provides an effective solution for measuring the relative ages of planetary surfaces. While texture features have proven to be effective in crater detection [11], tens of thousands of texture-based features, in different scales and different resolutions, can potentially be generated for high resolution planetary images. It is infeasible to pregenerate texture features from planetary images that have a near global coverage of the Martian surface.…”

Detecting Impact Craters in Planetary Images Using Machine Learning

“…Unlike the case of topographic data, no clear, physics-based features are available, because image is not a physical reality of the surface but rather its projection into 2-D space under given illumination. We encode images of crater candidates in terms of texture features (Ding, et al, 2011). First, each candidate (irregular fragment of an image) is embedded in a square image block, centered on the location of the candidate and having a dimension twice the diameter of the candidate.…”

“…Such texture features are broadly utilized for object detection and have proven to work well for crater detection (Martins, Pina, Marques, & Silveira., 2008). Overall, we represent each crater candidate by 1089 texture features; see (Ding, et al, 2011) for details. A large number of features restricts our choices of a learning algorithm; for example, decision trees or support vector machines would be ineffective in such context.…”