Using GNG to improve 3D feature extraction—Application to 6DoF egomotion

Several recent works deal with 3D data in mobile robotic problems, e.g. mapping or egomotion. Data comes from any kind of sensor such as stereo vision systems, time of flight cameras or 3D lasers, providing a huge amount of unorganized 3D data. In this paper, we describe an efficient method to build complete 3D models from a Growing Neural Gas (GNG). The GNG is applied to the 3D raw data and it reduces both the subjacent error and the number of points, keeping the topology of the 3D data. The GNG output is then used in a 3D feature extraction method. We have performed a deep study in which we quantitatively show that the use of GNG improves the 3D feature extraction method. We also show that our method can be applied to any kind of 3D data. The 3D features obtained are used as input in an Iterative Closest Point (ICP)-like method to compute the 6DoF movement performed by a mobile robot. A comparison with standard ICP is performed, showing that the use of GNG improves the results. Final results of 3D mapping from the egomotion calculated are also shown.

3D Feature Extraction and Modelling for SLAM

Paper submitted to the 39th International Symposium on Robotics ISR 2008, Seoul, South Korea, October 15-17, 2008.

Discussing the physical meaning of the absorption feature at 2.1 keV in 4U 1538–52

High resolution X-ray spectroscopy is a powerful tool for studying the nature of the matter surrounding the neutron star in X-ray binaries and its interaction between the stellar wind and the compact object. In particular, absorption features in their spectra could reveal the presence of atmospheres of the neutron star or their magnetic field strength. Here we present an investigation of the absorption feature at 2.1 keV in the X-ray spectrum of the high mass X-ray binary 4U 1538–52 based on our previous analysis of the XMM-Newton data. We study various possible origins and discuss the different physical scenarios in order to explain this feature. A likely interpretation is that the feature is associated with atomic transitions in an O/Ne neutron star atmosphere or of hydrogen and helium like Fe or Si ions formed in the stellar wind of the donor.

Feature selection by multi-objective optimisation: Application to network anomaly detection by hierarchical self-organising maps

Feature selection is an important and active issue in clustering and classification problems. By choosing an adequate feature subset, a dataset dimensionality reduction is allowed, thus contributing to decreasing the classification computational complexity, and to improving the classifier performance by avoiding redundant or irrelevant features. Although feature selection can be formally defined as an optimisation problem with only one objective, that is, the classification accuracy obtained by using the selected feature subset, in recent years, some multi-objective approaches to this problem have been proposed. These either select features that not only improve the classification accuracy, but also the generalisation capability in case of supervised classifiers, or counterbalance the bias toward lower or higher numbers of features that present some methods used to validate the clustering/classification in case of unsupervised classifiers. The main contribution of this paper is a multi-objective approach for feature selection and its application to an unsupervised clustering procedure based on Growing Hierarchical Self-Organising Maps (GHSOMs) that includes a new method for unit labelling and efficient determination of the winning unit. In the network anomaly detection problem here considered, this multi-objective approach makes it possible not only to differentiate between normal and anomalous traffic but also among different anomalies. The efficiency of our proposals has been evaluated by using the well-known DARPA/NSL-KDD datasets that contain extracted features and labelled attacks from around 2 million connections. The selected feature sets computed in our experiments provide detection rates up to 99.8% with normal traffic and up to 99.6% with anomalous traffic, as well as accuracy values up to 99.12%.

The outburst decay of the low magnetic field magnetar SWIFT J1822.3−1606: phase-resolved analysis and evidence for a variable cyclotron feature

We study the timing and spectral properties of the low-magnetic field, transient magnetar SWIFT J1822.3−1606 as it approached quiescence. We coherently phase-connect the observations over a time-span of ∼500 d since the discovery of SWIFT J1822.3−1606 following the Swift-Burst Alert Telescope (BAT) trigger on 2011 July 14, and carried out a detailed pulse phase spectroscopy along the outburst decay. We follow the spectral evolution of different pulse phase intervals and find a phase and energy-variable spectral feature, which we interpret as proton cyclotron resonant scattering of soft photon from currents circulating in a strong (≳1014 G) small-scale component of the magnetic field near the neutron star surface, superimposed to the much weaker (∼3 × 1013 G) magnetic field. We discuss also the implications of the pulse-resolved spectral analysis for the emission regions on the surface of the cooling magnetar.

Constrained self-organizing feature map to preserve feature extraction topology

In many classification problems, it is necessary to consider the specific location of an n-dimensional space from which features have been calculated. For example, considering the location of features extracted from specific areas of a two-dimensional space, as an image, could improve the understanding of a scene for a video surveillance system. In the same way, the same features extracted from different locations could mean different actions for a 3D HCI system. In this paper, we present a self-organizing feature map able to preserve the topology of locations of an n-dimensional space in which the vector of features have been extracted. The main contribution is to implicitly preserving the topology of the original space because considering the locations of the extracted features and their topology could ease the solution to certain problems. Specifically, the paper proposes the n-dimensional constrained self-organizing map preserving the input topology (nD-SOM-PINT). Features in adjacent areas of the n-dimensional space, used to extract the feature vectors, are explicitly in adjacent areas of the nD-SOM-PINT constraining the neural network structure and learning. As a study case, the neural network has been instantiate to represent and classify features as trajectories extracted from a sequence of images into a high level of semantic understanding. Experiments have been thoroughly carried out using the CAVIAR datasets (Corridor, Frontal and Inria) taken into account the global behaviour of an individual in order to validate the ability to preserve the topology of the two-dimensional space to obtain high-performance classification for trajectory classification in contrast of non-considering the location of features. Moreover, a brief example has been included to focus on validate the nD-SOM-PINT proposal in other domain than the individual trajectory. Results confirm the high accuracy of the nD-SOM-PINT outperforming previous methods aimed to classify the same datasets.

Barnstable, Massachusetts 15 Minute Digital Raster Graphic

This layer is a digital raster graphic of the historic 15-minute USGS topographic quadrangle map of Barnstable, Massachusetts. The edition date is 1893 and the map was reprinted in 1907. A digital raster graphic (DRG) is a scanned image of a U.S. Geological Survey (USGS) standard series topographic map, including all map collar information. The image inside the map neatline is geo-referenced to the surface of the earth and fit to the Universal Transverse Mercator projection. The horizontal positional accuracy and datum of the DRG matches the accuracy and datum of the source map. The names of quadrangles which border this one appear on the map collar in their respective positions (N,S,E,W) in relation to this map.

Barre, Massachusetts 15 Minute Digital Raster Graphic

This layer is a digital raster graphic of the historic 15-minute USGS topographic quadrangle map of Barre, Massachusetts. The suvery (ground condition) date is 1887, the edition date is March, 1894 and the map was reprinted in 1942. A digital raster graphic (DRG) is a scanned image of a U.S. Geological Survey (USGS) standard series topographic map, including all map collar information. The image inside the map neatline is geo-referenced to the surface of the earth and fit to the Universal Transverse Mercator projection. The horizontal positional accuracy and datum of the DRG matches the accuracy and datum of the source map. The names of quadrangles which border this one appear on the map collar in their respective positions (N,S,E,W) in relation to this map.

Becket, Massachusetts 15 Minute Digital Raster Graphic

This layer is a digital raster graphic of the historic 15-minute USGS topographic quadrangle map of Becket, Massachusetts. The survey (ground condition) date is 1886. A digital raster graphic (DRG) is a scanned image of a U.S. Geological Survey (USGS) standard series topographic map, including all map collar information. The image inside the map neatline is geo-referenced to the surface of the earth and fit to the Universal Transverse Mercator projection. The horizontal positional accuracy and datum of the DRG matches the accuracy and datum of the source map. The names of quadrangles which border this one appear on the map collar in their respective positions (N,S,E,W) in relation to this map.

Belchertown, Massachusetts 15 Minute Digital Raster Graphic

This layer is a digital raster graphic of the historic 15-minute USGS topographic map of the Belchertown, Massachusetts quadrangle. The suvey (ground condition) dates are 1885 and 1887; the edition date is November, 1893. A digital raster graphic (DRG) is a scanned image of a U.S. Geological Survey (USGS) standard series topographic map, including all map collar information. The image inside the map neatline is geo-referenced to the surface of the earth and fit to the Universal Transverse Mercator projection. The horizontal positional accuracy and datum of the DRG matches the accuracy and datum of the source map. The names of quadrangles which border this one appear on the map collar in their respective positions (N,S,E,W) in relation to this map.

Berlin (N.Y.) Massachusetts 15 Minute Digital Raster Graphic

This layer is a digital raster graphic of the historic 15-minute USGS topographic quadrangle map entitled Berlin, (N.Y.) which also shows towns and features in Massachusetts. The survey dates (ground condition) for this map are 1885-88, and the edition date is 1890. A digital raster graphic (DRG) is a scanned image of a U.S. Geological Survey (USGS) standard series topographic map, including all map collar information. The image inside the map neatline is geo-referenced to the surface of the earth and fit to the Universal Transverse Mercator projection. The horizontal positional accuracy and datum of the DRG matches the accuracy and datum of the source map. The names of quadrangles which border this one appear on the map collar in their respective positions (N,S,E,W) in relation to this map.

Blackstone, Massachusetts 15 Minute Digital Raster Graphic

This layer is a digital raster graphic of the historic 15-minute USGS topographic map of the Blackstone, Massachusetts quadrangle. The survey date (ground condition) of this map is 1886 and the edition date is October, 1893. A digital raster graphic (DRG) is a scanned image of a U.S. Geological Survey (USGS) standard series topographic map, including all map collar information. The image inside the map neatline is geo-referenced to the surface of the earth and fit to the Universal Transverse Mercator projection. The horizontal positional accuracy and datum of the DRG matches the accuracy and datum of the source map. The names of quadrangles which border this one appear on the map collar in their respective positions (N,S,E,W) in relation to this map.

Boston (1900), Massachusetts 15 Minute Digital Raster Graphic

This layer is a digital raster graphic (DRG) of the historic 15-minute USGS topographic map of the Boston, Massachusetts quadrangle. The survey date (ground condition) of this map ranges from 1898 to 1900, the edition date is July, 1903 and it was reprinted in 1918. A digital raster graphic (DRG) is a scanned image of a U.S. Geological Survey (USGS) standard series topographic map, including all map collar information. The image inside the map neatline is geo-referenced to the surface of the earth and fit to the Universal Transverse Mercator projection. The horizontal positional accuracy and datum of the DRG matches the accuracy and datum of the source map. The names of quadrangles which border this one appear on the map collar in their respective positions (N,S,E,W) in relation to this map.

Boston North (1943), Massachusetts 15 Minute Digital Raster Graphic

This layer is a digital raster graphic (DRG) of the historic 15-minute USGS topographic map of the Boston North, Massachusetts quadrangle. The survey date (ground condition) of this map is 1943, the edition date is 1946. A digital raster graphic (DRG) is a scanned image of a U.S. Geological Survey (USGS) standard series topographic map, including all map collar information. The image inside the map neatline is geo-referenced to the surface of the earth and fit to the Universal Transverse Mercator projection. The horizontal positional accuracy and datum of the DRG matches the accuracy and datum of the source map. The names of quadrangles which border this one appear on the map collar in their respective positions (N,S,E,W) in relation to this map.

Boston South (1950), Massachusetts 15 Minute Digital Raster Graphic

This layer is a digital raster graphic (DRG) of the historic 15-minute USGS topographic map of the Boston South, Massachusetts quadrangle. The survey date (ground condition) of this map is 1943, it was revised in 1949 and reprinted with corrections in 1950. A digital raster graphic (DRG) is a scanned image of a U.S. Geological Survey (USGS) standard series topographic map, including all map collar information. The image inside the map neatline is geo-referenced to the surface of the earth and fit to the Universal Transverse Mercator projection. The horizontal positional accuracy and datum of the DRG matches the accuracy and datum of the source map. The names of quadrangles which border this one appear on the map collar in their respective positions (N,S,E,W) in relation to this map.