Automated Crater Detection, A New Tool for Mars Cartography and Chronology

Abstract

An automated crater detection algorithm is presented which exploits image data. The algorithm is briefly described and its application demonstrated on a variety of different Martian geomorphological areas and sensors (Viking Orbiter Camera, Mars Orbiter Camera (MOC), Mars Orbiter Laser Altimeter (MOLA), and High Resolution Stereo Camera (HRSC)). We show assessment results through both an intercomparison of automated crater locations with those from the manually-derived Mars Crater Consortium (MCC) catalogue and the manually-derived craters. The detection algorithm attains an accuracy of 70 to 90 percent and a quality factor of 60 to 80 percent depending on target sensor type and geomorphology. We also present crater detection results derived from HRSC images onboard the ESA Mars Express on a comparison between manually-determined Size-Frequency Distributions (SFDs) and those derived fully automatically. The approach described appears to offer great potential for chronological research, geomatic and geological analysis and for other purposes of extra-terrestrial planetary surface mapping.

Detecting Impact Craters in Planetary Images Using Machine Learning

“…The first class of methods rely exclusively on pattern recognition techniques to identify crater rims having circular or elliptical features in an image (for example, (Barata, Alves, Saraiva, & Pina, 2004) (Cheng, Johnson, Matthies, & Olson, 2002) (Honda, Iijima, & Konishi, 2002) (Kim, Muller, S., J., & Neukum, 2005) (Leroy, Medioni, & Matthies, 2001) (Salamaniccar & Loncaric, 2010) . The general idea of such methods is to first preprocess an image to enhance the edges of the crater rims, and then to detect the craters using variants of the Hough Transform (Hough V, 1962), genetic algorithms (Honda, Iijima, & Konishi, 2002), or the radial consistency algorithm (Earl, Chicarro, Koeberl, Marchetti, & Milnes, 2005) that identifies regions of rotational symmetry.…”

Adaptive Selective Learning for automatic identification of sub-kilometer craters

“…The unsupervised methods rely on image processing techniques to identify crater rims in an image as circular or elliptical features [Leroy et al, 2001;Honda et al, 2003;Cheng et al, 2003;Barata et al, 2004;Kim et al, 2005]. The original image is preprocessed to enhance the edges of the rims, and the actual detection is achieved by means of the Hough Transform (HT) [Hough, 1962], genetic algorithms [Honda et al, 2003], or the radial consistency algorithm [Earl et al, 2005] that identifies regions of rotational symmetry.…”

“…Most recently, advances in face detection research are incorporated into crater detection techniques. In [Kim et al 2005], the combination of edge detection, template matching, and neural network-based false positive recognition scheme is used for detecting craters on Mars. In a boosting algorithm, originally developed by [Viola and Jones, 2004] in the context of face detection, is adopted for identification of craters on Mars.…”

Automatic extraction of lunar impact craters from Chang’E-1 satellite photographs

“…Their extraction work used topographic and spectral information from lunar photos and images like Apollo data, Clementine UV-Vis is multi-spectral images and SELENE data. There was also much work [6][7][8] …”

“…Their extraction work used topographic and spectral information from lunar photos and images like Apollo data, Clementine UV-Vis is multi-spectral images and SELENE data. There was also much work [6][7][8] on automatic extraction for Martian impact craters. Different detection methods were also used, including template matching [9][10][11], texture analysis [12], genetic algorithms [13] and object-oriented approach [14], etc.…”