Estimation of the lateral correlation structure of subsurface water content from surface-based ground-penetrating radar reflection images

Over the past decade, significant interest has been expressed in relating the spatial statistics of surface-based reflection ground-penetrating radar (GPR) data to those of the imaged subsurface volume. A primary motivation for this work is that changes in the radar wave velocity, which largely control the character of the observed data, are expected to be related to corresponding changes in subsurface water content. Although previous work has indeed indicated that the spatial statistics of GPR images are linked to those of the water content distribution of the probed region, a viable method for quantitatively analyzing the GPR data and solving the corresponding inverse problem has not yet been presented. Here we address this issue by first deriving a relationship between the 2-D autocorrelation of a water content distribution and that of the corresponding GPR reflection image. We then show how a Bayesian inversion strategy based on Markov chain Monte Carlo sampling can be used to estimate the posterior distribution of subsurface correlation model parameters that are consistent with the GPR data. Our results indicate that if the underlying assumptions are valid and we possess adequate prior knowledge regarding the water content distribution, in particular its vertical variability, this methodology allows not only for the reliable recovery of lateral correlation model parameters but also for estimates of parameter uncertainties. In the case where prior knowledge regarding the vertical variability of water content is not available, the results show that the methodology still reliably recovers the aspect ratio of the heterogeneity.

Estimation of the correlation structure of crustal velocity heterogeneity from seismic reflection data

Numerous sources of evidence point to the fact that heterogeneity within the Earth's deep crystalline crust is complex and hence may be best described through stochastic rather than deterministic approaches. As seismic reflection imaging arguably offers the best means of sampling deep crustal rocks in situ, much interest has been expressed in using such data to characterize the stochastic nature of crustal heterogeneity. Previous work on this problem has shown that the spatial statistics of seismic reflection data are indeed related to those of the underlying heterogeneous seismic velocity distribution. As of yet, however, the nature of this relationship has remained elusive due to the fact that most of the work was either strictly empirical or based on incorrect methodological approaches. Here, we introduce a conceptual model, based on the assumption of weak scattering, that allows us to quantitatively link the second-order statistics of a 2-D seismic velocity distribution with those of the corresponding processed and depth-migrated seismic reflection image. We then perform a sensitivity study in order to investigate what information regarding the stochastic model parameters describing crustal velocity heterogeneity might potentially be recovered from the statistics of a seismic reflection image using this model. Finally, we present a Monte Carlo inversion strategy to estimate these parameters and we show examples of its application at two different source frequencies and using two different sets of prior information. Our results indicate that the inverse problem is inherently non-unique and that many different combinations of the vertical and lateral correlation lengths describing the velocity heterogeneity can yield seismic images with the same 2-D autocorrelation structure. The ratio of all of these possible combinations of vertical and lateral correlation lengths, however, remains roughly constant which indicates that, without additional prior information, the aspect ratio is the only parameter describing the stochastic seismic velocity structure that can be reliably recovered.

Classification of urban multi-angular image sequences by aligning their manifolds

When dealing with multi-angular image sequences, problems of reflectance changes due either to illumination and acquisition geometry, or to interactions with the atmosphere, naturally arise. These phenomena interplay with the scene and lead to a modification of the measured radiance: for example, according to the angle of acquisition, tall objects may be seen from top or from the side and different light scatterings may affect the surfaces. This results in shifts in the acquired radiance, that make the problem of multi-angular classification harder and might lead to catastrophic results, since surfaces with the same reflectance return significantly different signals. In this paper, rather than performing atmospheric or bi-directional reflection distribution function (BRDF) correction, a non-linear manifold learning approach is used to align data structures. This method maximizes the similarity between the different acquisitions by deforming their manifold, thus enhancing the transferability of classification models among the images of the sequence.

Constraining 3-D electrical resistance tomography with GPR reflection data forimproved aquifer characterization

Surface-based ground penetrating radar (GPR) and electrical resistance tomography (ERT) are common tools for aquifer characterization, because both methods provide data that are sensitive to hydrogeologically relevant quantities. To retrieve bulk subsurface properties at high resolution, we suggest incorporating structural information derived from GPR reflection data when inverting surface ERT data. This reduces resolution limitations, which might hinder quantitative interpretations. Surface-based GPR reflection and ERT data have been recorded on an exposed gravel bar within a restored section of a previously channelized river in northeastern Switzerland to characterize an underlying gravel aquifer. The GPR reflection data acquired over an area of 240×40 m map the aquifer's thickness and two internal sub-horizontal regions with different depositional patterns. The interface between these two regions and the boundary of the aquifer with then underlying clay are incorporated in an unstructured ERT mesh. Subsequent inversions are performed without applying smoothness constraints across these boundaries. Inversion models obtained by using these structural constraints contain subtle resistivity variations within the aquifer that are hardly visible in standard inversion models as a result of strong vertical smearing in the latter. In the upper aquifer region, with high GPR coherency and horizontal layering, the resistivity is moderately high (N300 Ωm). We suggest that this region consists of sediments that were rearranged during more than a century of channelized flow. In the lower low coherency region, the GPR image reveals fluvial features (e.g., foresets) and generally more heterogeneous deposits. In this region, the resistivity is lower (~200 Ωm), which we attribute to increased amounts of fines in some of the well-sorted fluvial deposits. We also find elongated conductive anomalies that correspond to the location of river embankments that were removed in 2002.

Development of a 3D very high-resolution seismic reflection system for lacustrine settings - a case study over a thrust fault zone in Lake Geneva

Un système efficace de sismique tridimensionnelle (3-D) haute-résolution adapté à des cibles lacustres de petite échelle a été développé. Dans le Lac Léman, près de la ville de Lausanne, en Suisse, des investigations récentes en deux dimension (2-D) ont mis en évidence une zone de faille complexe qui a été choisie pour tester notre système. Les structures observées incluent une couche mince (<40 m) de sédiments quaternaires sub-horizontaux, discordants sur des couches tertiaires de molasse pentées vers le sud-est. On observe aussi la zone de faille de « La Paudèze » qui sépare les unités de la Molasse du Plateau de la Molasse Subalpine. Deux campagnes 3-D complètes, d?environ d?un kilomètre carré, ont été réalisées sur ce site de test. La campagne pilote (campagne I), effectuée en 1999 pendant 8 jours, a couvert 80 profils en utilisant une seule flûte. Pendant la campagne II (9 jours en 2001), le nouveau système trois-flûtes, bien paramétrés pour notre objectif, a permis l?acquisition de données de très haute qualité sur 180 lignes CMP. Les améliorations principales incluent un système de navigation et de déclenchement de tirs grâce à un nouveau logiciel. Celui-ci comprend un contrôle qualité de la navigation du bateau en temps réel utilisant un GPS différentiel (dGPS) à bord et une station de référence près du bord du lac. De cette façon, les tirs peuvent être déclenchés tous les 5 mètres avec une erreur maximale non-cumulative de 25 centimètres. Tandis que pour la campagne I la position des récepteurs de la flûte 48-traces a dû être déduite à partir des positions du bateau, pour la campagne II elle ont pu être calculées précisément (erreur <20 cm) grâce aux trois antennes dGPS supplémentaires placées sur des flotteurs attachés à l?extrémité de chaque flûte 24-traces. Il est maintenant possible de déterminer la dérive éventuelle de l?extrémité des flûtes (75 m) causée par des courants latéraux ou de petites variations de trajet du bateau. De plus, la construction de deux bras télescopiques maintenant les trois flûtes à une distance de 7.5 m les uns des autres, qui est la même distance que celle entre les lignes naviguées de la campagne II. En combinaison avec un espacement de récepteurs de 2.5 m, la dimension de chaque «bin» de données 3-D de la campagne II est de 1.25 m en ligne et 3.75 m latéralement. L?espacement plus grand en direction « in-line » par rapport à la direction «cross-line» est justifié par l?orientation structurale de la zone de faille perpendiculaire à la direction «in-line». L?incertitude sur la navigation et le positionnement pendant la campagne I et le «binning» imprécis qui en résulte, se retrouve dans les données sous forme d?une certaine discontinuité des réflecteurs. L?utilisation d?un canon à air à doublechambre (qui permet d?atténuer l?effet bulle) a pu réduire l?aliasing observé dans les sections migrées en 3-D. Celui-ci était dû à la combinaison du contenu relativement haute fréquence (<2000 Hz) du canon à eau (utilisé à 140 bars et à 0.3 m de profondeur) et d?un pas d?échantillonnage latéral insuffisant. Le Mini G.I 15/15 a été utilisé à 80 bars et à 1 m de profondeur, est mieux adapté à la complexité de la cible, une zone faillée ayant des réflecteurs pentés jusqu?à 30°. Bien que ses fréquences ne dépassent pas les 650 Hz, cette source combine une pénétration du signal non-aliasé jusqu?à 300 m dans le sol (par rapport au 145 m pour le canon à eau) pour une résolution verticale maximale de 1.1 m. Tandis que la campagne I a été acquise par groupes de plusieurs lignes de directions alternées, l?optimisation du temps d?acquisition du nouveau système à trois flûtes permet l?acquisition en géométrie parallèle, ce qui est préférable lorsqu?on utilise une configuration asymétrique (une source et un dispositif de récepteurs). Si on ne procède pas ainsi, les stacks sont différents selon la direction. Toutefois, la configuration de flûtes, plus courtes que pour la compagne I, a réduit la couverture nominale, la ramenant de 12 à 6. Une séquence classique de traitement 3-D a été adaptée à l?échantillonnage à haute fréquence et elle a été complétée par deux programmes qui transforment le format non-conventionnel de nos données de navigation en un format standard de l?industrie. Dans l?ordre, le traitement comprend l?incorporation de la géométrie, suivi de l?édition des traces, de l?harmonisation des «bins» (pour compenser l?inhomogénéité de la couverture due à la dérive du bateau et de la flûte), de la correction de la divergence sphérique, du filtrage passe-bande, de l?analyse de vitesse, de la correction DMO en 3-D, du stack et enfin de la migration 3-D en temps. D?analyses de vitesse détaillées ont été effectuées sur les données de couverture 12, une ligne sur deux et tous les 50 CMP, soit un nombre total de 600 spectres de semblance. Selon cette analyse, les vitesses d?intervalles varient de 1450-1650 m/s dans les sédiments non-consolidés et de 1650-3000 m/s dans les sédiments consolidés. Le fait que l?on puisse interpréter plusieurs horizons et surfaces de faille dans le cube, montre le potentiel de cette technique pour une interprétation tectonique et géologique à petite échelle en trois dimensions. On distingue cinq faciès sismiques principaux et leurs géométries 3-D détaillées sur des sections verticales et horizontales: les sédiments lacustres (Holocène), les sédiments glacio-lacustres (Pléistocène), la Molasse du Plateau, la Molasse Subalpine de la zone de faille (chevauchement) et la Molasse Subalpine au sud de cette zone. Les couches de la Molasse du Plateau et de la Molasse Subalpine ont respectivement un pendage de ~8° et ~20°. La zone de faille comprend de nombreuses structures très déformées de pendage d?environ 30°. Des tests préliminaires avec un algorithme de migration 3-D en profondeur avant sommation et à amplitudes préservées démontrent que la qualité excellente des données de la campagne II permet l?application de telles techniques à des campagnes haute-résolution. La méthode de sismique marine 3-D était utilisée jusqu?à présent quasi-exclusivement par l?industrie pétrolière. Son adaptation à une échelle plus petite géographiquement mais aussi financièrement a ouvert la voie d?appliquer cette technique à des objectifs d?environnement et du génie civil.<br/><br/>An efficient high-resolution three-dimensional (3-D) seismic reflection system for small-scale targets in lacustrine settings was developed. In Lake Geneva, near the city of Lausanne, Switzerland, past high-resolution two-dimensional (2-D) investigations revealed a complex fault zone (the Paudèze thrust zone), which was subsequently chosen for testing our system. Observed structures include a thin (<40 m) layer of subhorizontal Quaternary sediments that unconformably overlie southeast-dipping Tertiary Molasse beds and the Paudèze thrust zone, which separates Plateau and Subalpine Molasse units. Two complete 3-D surveys have been conducted over this same test site, covering an area of about 1 km2. In 1999, a pilot survey (Survey I), comprising 80 profiles, was carried out in 8 days with a single-streamer configuration. In 2001, a second survey (Survey II) used a newly developed three-streamer system with optimized design parameters, which provided an exceptionally high-quality data set of 180 common midpoint (CMP) lines in 9 days. The main improvements include a navigation and shot-triggering system with in-house navigation software that automatically fires the gun in combination with real-time control on navigation quality using differential GPS (dGPS) onboard and a reference base near the lake shore. Shots were triggered at 5-m intervals with a maximum non-cumulative error of 25 cm. Whereas the single 48-channel streamer system of Survey I requires extrapolation of receiver positions from the boat position, for Survey II they could be accurately calculated (error <20 cm) with the aid of three additional dGPS antennas mounted on rafts attached to the end of each of the 24- channel streamers. Towed at a distance of 75 m behind the vessel, they allow the determination of feathering due to cross-line currents or small course variations. Furthermore, two retractable booms hold the three streamers at a distance of 7.5 m from each other, which is the same distance as the sail line interval for Survey I. With a receiver spacing of 2.5 m, the bin dimension of the 3-D data of Survey II is 1.25 m in in-line direction and 3.75 m in cross-line direction. The greater cross-line versus in-line spacing is justified by the known structural trend of the fault zone perpendicular to the in-line direction. The data from Survey I showed some reflection discontinuity as a result of insufficiently accurate navigation and positioning and subsequent binning errors. Observed aliasing in the 3-D migration was due to insufficient lateral sampling combined with the relatively high frequency (<2000 Hz) content of the water gun source (operated at 140 bars and 0.3 m depth). These results motivated the use of a double-chamber bubble-canceling air gun for Survey II. A 15 / 15 Mini G.I air gun operated at 80 bars and 1 m depth, proved to be better adapted for imaging the complexly faulted target area, which has reflectors dipping up to 30°. Although its frequencies do not exceed 650 Hz, this air gun combines a penetration of non-aliased signal to depths of 300 m below the water bottom (versus 145 m for the water gun) with a maximum vertical resolution of 1.1 m. While Survey I was shot in patches of alternating directions, the optimized surveying time of the new threestreamer system allowed acquisition in parallel geometry, which is preferable when using an asymmetric configuration (single source and receiver array). Otherwise, resulting stacks are different for the opposite directions. However, the shorter streamer configuration of Survey II reduced the nominal fold from 12 to 6. A 3-D conventional processing flow was adapted to the high sampling rates and was complemented by two computer programs that format the unconventional navigation data to industry standards. Processing included trace editing, geometry assignment, bin harmonization (to compensate for uneven fold due to boat/streamer drift), spherical divergence correction, bandpass filtering, velocity analysis, 3-D DMO correction, stack and 3-D time migration. A detailed semblance velocity analysis was performed on the 12-fold data set for every second in-line and every 50th CMP, i.e. on a total of 600 spectra. According to this velocity analysis, interval velocities range from 1450-1650 m/s for the unconsolidated sediments and from 1650-3000 m/s for the consolidated sediments. Delineation of several horizons and fault surfaces reveal the potential for small-scale geologic and tectonic interpretation in three dimensions. Five major seismic facies and their detailed 3-D geometries can be distinguished in vertical and horizontal sections: lacustrine sediments (Holocene) , glaciolacustrine sediments (Pleistocene), Plateau Molasse, Subalpine Molasse and its thrust fault zone. Dips of beds within Plateau and Subalpine Molasse are ~8° and ~20°, respectively. Within the fault zone, many highly deformed structures with dips around 30° are visible. Preliminary tests with 3-D preserved-amplitude prestack depth migration demonstrate that the excellent data quality of Survey II allows application of such sophisticated techniques even to high-resolution seismic surveys. In general, the adaptation of the 3-D marine seismic reflection method, which to date has almost exclusively been used by the oil exploration industry, to a smaller geographical as well as financial scale has helped pave the way for applying this technique to environmental and engineering purposes.<br/><br/>La sismique réflexion est une méthode d?investigation du sous-sol avec un très grand pouvoir de résolution. Elle consiste à envoyer des vibrations dans le sol et à recueillir les ondes qui se réfléchissent sur les discontinuités géologiques à différentes profondeurs et remontent ensuite à la surface où elles sont enregistrées. Les signaux ainsi recueillis donnent non seulement des informations sur la nature des couches en présence et leur géométrie, mais ils permettent aussi de faire une interprétation géologique du sous-sol. Par exemple, dans le cas de roches sédimentaires, les profils de sismique réflexion permettent de déterminer leur mode de dépôt, leurs éventuelles déformations ou cassures et donc leur histoire tectonique. La sismique réflexion est la méthode principale de l?exploration pétrolière. Pendant longtemps on a réalisé des profils de sismique réflexion le long de profils qui fournissent une image du sous-sol en deux dimensions. Les images ainsi obtenues ne sont que partiellement exactes, puisqu?elles ne tiennent pas compte de l?aspect tridimensionnel des structures géologiques. Depuis quelques dizaines d?années, la sismique en trois dimensions (3-D) a apporté un souffle nouveau à l?étude du sous-sol. Si elle est aujourd?hui parfaitement maîtrisée pour l?imagerie des grandes structures géologiques tant dans le domaine terrestre que le domaine océanique, son adaptation à l?échelle lacustre ou fluviale n?a encore fait l?objet que de rares études. Ce travail de thèse a consisté à développer un système d?acquisition sismique similaire à celui utilisé pour la prospection pétrolière en mer, mais adapté aux lacs. Il est donc de dimension moindre, de mise en oeuvre plus légère et surtout d?une résolution des images finales beaucoup plus élevée. Alors que l?industrie pétrolière se limite souvent à une résolution de l?ordre de la dizaine de mètres, l?instrument qui a été mis au point dans le cadre de ce travail permet de voir des détails de l?ordre du mètre. Le nouveau système repose sur la possibilité d?enregistrer simultanément les réflexions sismiques sur trois câbles sismiques (ou flûtes) de 24 traces chacun. Pour obtenir des données 3-D, il est essentiel de positionner les instruments sur l?eau (source et récepteurs des ondes sismiques) avec une grande précision. Un logiciel a été spécialement développé pour le contrôle de la navigation et le déclenchement des tirs de la source sismique en utilisant des récepteurs GPS différentiel (dGPS) sur le bateau et à l?extrémité de chaque flûte. Ceci permet de positionner les instruments avec une précision de l?ordre de 20 cm. Pour tester notre système, nous avons choisi une zone sur le Lac Léman, près de la ville de Lausanne, où passe la faille de « La Paudèze » qui sépare les unités de la Molasse du Plateau et de la Molasse Subalpine. Deux campagnes de mesures de sismique 3-D y ont été réalisées sur une zone d?environ 1 km2. Les enregistrements sismiques ont ensuite été traités pour les transformer en images interprétables. Nous avons appliqué une séquence de traitement 3-D spécialement adaptée à nos données, notamment en ce qui concerne le positionnement. Après traitement, les données font apparaître différents faciès sismiques principaux correspondant notamment aux sédiments lacustres (Holocène), aux sédiments glacio-lacustres (Pléistocène), à la Molasse du Plateau, à la Molasse Subalpine de la zone de faille et la Molasse Subalpine au sud de cette zone. La géométrie 3-D détaillée des failles est visible sur les sections sismiques verticales et horizontales. L?excellente qualité des données et l?interprétation de plusieurs horizons et surfaces de faille montrent le potentiel de cette technique pour les investigations à petite échelle en trois dimensions ce qui ouvre des voies à son application dans les domaines de l?environnement et du génie civil.

Urban image classification with semisupervised multiscale cluster kernels

This paper presents a semisupervised support vector machine (SVM) that integrates the information of both labeled and unlabeled pixels efficiently. Method's performance is illustrated in the relevant problem of very high resolution image classification of urban areas. The SVM is trained with the linear combination of two kernels: a base kernel working only with labeled examples is deformed by a likelihood kernel encoding similarities between labeled and unlabeled examples. Results obtained on very high resolution (VHR) multispectral and hyperspectral images show the relevance of the method in the context of urban image classification. Also, its simplicity and the few parameters involved make the method versatile and workable by unexperienced users.

Neuroscience, quantum indeterminism and the Cartesian soul.

Quantum indeterminism is frequently invoked as a solution to the problem of how a disembodied soul might interact with the brain (as Descartes proposed), and is sometimes invoked in theories of libertarian free will even when they do not involve dualistic assumptions. Taking as example the Eccles-Beck model of interaction between self (or soul) and brain at the level of synaptic exocytosis, I here evaluate the plausibility of these approaches. I conclude that Heisenbergian uncertainty is too small to affect synaptic function, and that amplification by chaos or by other means does not provide a solution to this problem. Furthermore, even if Heisenbergian effects did modify brain functioning, the changes would be swamped by those due to thermal noise. Cells and neural circuits have powerful noise-resistance mechanisms, that are adequate protection against thermal noise and must therefore be more than sufficient to buffer against Heisenbergian effects. Other forms of quantum indeterminism must be considered, because these can be much greater than Heisenbergian uncertainty, but these have not so far been shown to play a role in the brain.

Characteristic reflection patterns in the southeast Canadian cordillera, northern Appalachians and Swiss Alps

There are some striking similarities and some differences between the seismic reflection sections recorded across the fold and thrust belts of the southeast Canadian Cordillera, Quebec-Maine Appalachians and Swiss Alps. In the fold and thrust belts of all three mountain ranges, seismic reflection surveys have yielded high-quality images of. (1) nappes (thin thrust sheets) stacked on top of ancient continental margins; (2) ramp anticlines in the hanging walls of faults that have ramp-flat or listric geometries; (3) back thrusts and back folds that developed during the terminal phases of orogeny; and (4) tectonic wedges and regional decollements. A principal result of the Cordilleran and Appalachian deep crustal studies has been the recognition of master decollements along which continental margin strata have been transported long distances, whereas a principal result of the Swiss Alpine deep crustal program has been the identification of the Adriatic indenter, a crustal-scale wedge that caused delamination of the European lithosphere. Significant crustal roots are observed beneath the fold and thrust belts of the Alps, southeast Canadian Cordillera and parts of the southern Appalachians, but such structures beneath the northern Appalachians have probably been removed by post-orogenic collapse and/or crustal attenuation associated with the Mesozoic opening of the Atlantic Ocean.

Generating controlled image sets in cognitive neuroscience research.

The investigation of perceptual and cognitive functions with non-invasive brain imaging methods critically depends on the careful selection of stimuli for use in experiments. For example, it must be verified that any observed effects follow from the parameter of interest (e.g. semantic category) rather than other low-level physical features (e.g. luminance, or spectral properties). Otherwise, interpretation of results is confounded. Often, researchers circumvent this issue by including additional control conditions or tasks, both of which are flawed and also prolong experiments. Here, we present some new approaches for controlling classes of stimuli intended for use in cognitive neuroscience, however these methods can be readily extrapolated to other applications and stimulus modalities. Our approach is comprised of two levels. The first level aims at equalizing individual stimuli in terms of their mean luminance. Each data point in the stimulus is adjusted to a standardized value based on a standard value across the stimulus battery. The second level analyzes two populations of stimuli along their spectral properties (i.e. spatial frequency) using a dissimilarity metric that equals the root mean square of the distance between two populations of objects as a function of spatial frequency along x- and y-dimensions of the image. Randomized permutations are used to obtain a minimal value between the populations to minimize, in a completely data-driven manner, the spectral differences between image sets. While another paper in this issue applies these methods in the case of acoustic stimuli (Aeschlimann et al., Brain Topogr 2008), we illustrate this approach here in detail for complex visual stimuli.

A Survey of Active Learning Algorithms for Supervised Remote Sensing Image Classification

Defining an efficient training set is one of the most delicate phases for the success of remote sensing image classification routines. The complexity of the problem, the limited temporal and financial resources, as well as the high intraclass variance can make an algorithm fail if it is trained with a suboptimal dataset. Active learning aims at building efficient training sets by iteratively improving the model performance through sampling. A user-defined heuristic ranks the unlabeled pixels according to a function of the uncertainty of their class membership and then the user is asked to provide labels for the most uncertain pixels. This paper reviews and tests the main families of active learning algorithms: committee, large margin, and posterior probability-based. For each of them, the most recent advances in the remote sensing community are discussed and some heuristics are detailed and tested. Several challenging remote sensing scenarios are considered, including very high spatial resolution and hyperspectral image classification. Finally, guidelines for choosing the good architecture are provided for new and/or unexperienced user.

MSight: an image analysis software for liquid chromatography-mass spectrometry.

Images obtained from high-throughput mass spectrometry (MS) contain information that remains hidden when looking at a single spectrum at a time. Image processing of liquid chromatography-MS datasets can be extremely useful for quality control, experimental monitoring and knowledge extraction. The importance of imaging in differential analysis of proteomic experiments has already been established through two-dimensional gels and can now be foreseen with MS images. We present MSight, a new software designed to construct and manipulate MS images, as well as to facilitate their analysis and comparison.