Pedro Pina

Impact Crater Recognition on Mars Based on a Probability Volume Created by Template Matching

This paper presents a methodology that brings together a number of techniques in the fields of image processing and pattern recognition with the purpose of achieving the automated detection of impact craters on images of planetary surfaces. The modular approach adopted for its development includes a phase of candidate selection, followed by template matching, in which the probability associated to each detection is established, and finally, by the analysis of the probability volume, in which the identification of craters on the image is achieved. It is tested on a set of images from four different regions of the surface of the planet Mars, all obtained by the same sensor in the last decade. The recognition rates for craters with radii that are larger than five pixels are very good, both globally and for each of the individual areas. The performance of the algorithm in the face of the variation of some of its parameters is analyzed and discussed in detail. We believe that this is a tool that is suitable for a general application in any area of a planet or satellite captured in an image, whatever the geomorphological setting, the optical sensor, and the conditions of illumination are.

Crater Detection by a Boosting Approach

An approach to automatically detect impact craters on planetary surfaces is presented in this letter. It is built up from a boosting algorithm proposed by Viola and Jones (2004) whose simplicity combined with an original learning strategy leads to a fast and robust process with consistent results. The approach is validated with image data sets from Mars surface captured by the Mars Orbiter Camera onboard Mars Global Surveyor probe.

Automatic Recognition of Impact Craters on the Surface of Mars

This paper presents a methodology to automatically recognise impact craters on the surface of Mars. It consists of three main phases: in the first one the images are segmented through a PCA of statistical texture measures, fol- lowed by the enhancement of the selected contours; in a second phase craters are recognised through a template matching approach; in a third phase the rims of the plotted craters are locally fitted through the watershed transform.

Development of a Methodology for Automated Crater Detection on Planetary Images

This paper presents a methodology for the automated detection of impact craters on images of planetary surfaces. This modular approach includes a phase of candidate selection, followed by template matching and finally the analysis of a probability volume that allows for the identification of craters on the image. It is applied to a set of images of the surface of the planet Mars, with results that are very promising, in face of future improvements in the methodology.

Automated Detection of Martian Dune Fields

An approach for the automated detection of dune fields on remotely sensed images of the surface of Mars is presented in this letter. It is based on the extraction of local information from images (i.e., gradient features), which, in turn, is tested with boosting and support vector machine classifiers. A detection rate of about 95% is obtained for fivefold cross validation on a set of 78 panchromatic images captured by the Mars Orbiter Camera of the Mars Global Surveyor probe on different locations of the planet.

Characterization of cells in cork

Various topological and metric properties of the cells in the phelogen of the cork oak have been measured in tangential sections of cork by image analysis methods. These include the fractions of cells with i sides (i-cells), the fractions of adjacencies between i- and k-cells and various distributions of cell areas in relation to topology.

Average topological properties of successive neighbours of cells in random networks

Abstract Random planar (trivalent) networks are studied with the purpose of finding correlations between the number i of sides of a cell (i cell) and average properties of its successive neighbours. The properties investigated are the average number n i k of neighbours of order k of i cells and the average number m i k of sides of these neighbours. The neighbours of a given order k are classified into various types according to their ‘proximity’ to k+1 neighbours. Approximate relations are anticipated, such as a linear variation in n i k with both i and k, and a generalization of the Aboav-Weaire relation to more distant neighbours, that is a linear relation between n i k m i k and i. These relations were ‘experimentally’ assessed by analysing two types of random trivalent network (Voronoi and Poisson) using image analysis methods.

Identification of martian polygonal patterns using the dynamics of watershed contours

This paper presents a methodology to automatically identify polygonal patterns on the surface of Mars. These structures, which are typical of periglacial regions, result from climate oscillations and present a wide variation in size, shape and topology and occur in different types of terrains with rather different constituents and spectral reflectances. The proposed approach is mainly based on the analysis of the dynamics of watershed contours and is successfully applied to a set of different types of patterned terrains of Mars shown by MGS/MOC images.

A morphological approach for feature space partitioning

A mathematical morphology-based methodology to construct decision region borders that geometrically model the training sets of points is presented in this letter. It is shown that the incorporation of the geometric features of the training sets leads to higher classification rates. Our approach is illustrated with two features of seven land-cover classes [forest (3), soil (2), vegetation, and water] constructed from remotely sensed images of a region in the center of Portugal.

A multi-layer approach for the analysis of neighbourhood relations of polygons in remotely acquired images

In this letter we present a novel approach to extract topological features of polygonal networks, based on a multi-layer strategy; the motivation for this new development was the need to analyse the small-scale polygonal patterns observed on remotely sensed, high spatial resolution images of the surface of Mars. The major improvement of the algorithm consists of the distribution of polygons by layers in such a way that adjacent polygons cannot coexist on any given layer; this is followed by a global analysis of each layer to extract topological features. This novel approach can be indistinctively applied to any kind of tri and tetravalent network (presenting respectively three and four polygons at each vertex); its computational performance is extremely favourable when compared with previous approaches to this problem. The experimental dataset used to evaluate the algorithm consisted of 47 segmented polygonal networks seen on the surface of Mars and presenting very distinct visual appearances.