Catálogo de publicaciones - libros

A novel method for the analysis of spectra and detection of absorption features in hyperspectral signatures is proposed, based on the ability of wavelet transformations to enhance absorption features. Field spectra of wheat grown on different levels of available nitrogen were collected, and compared to the foliar nitrogen content. The spectra were assessed both as absolute reflectances and recalculated into derivative spectra, and their respective wavelet transformed signals. Wavelet transformed signals, transformed using the Daubechies 5 motherwavelet at scaling level 32, performed consistently better than reflectance or derivative spectra when tested in a bootstrapped phased regression against nitrogen.

This paper explores wavelets as a measure of linear complexity. An introduction to wavelets is given before applying them to spatial data and testing them as a complexity measure on a vector representation of the Australian coastline. Wavelets are shown to be successful at measuring linear complexity. The technique used breaks a single line into different classes of complexity, and has the advantages that it is objective, automated and fast.

In many natural resource inventories class descriptions have atrophied to little more than simple labels or ciphers; the data producer expects the data user to share a common understanding of the way the world works and how it should be characterized (that is the producer implicitly assumes that their epistemology, ontology and semantics are universal). Activities like the UK e-science programme and the EU INSPIRE initiative mean that it is increasingly difficult for the producer to anticipate who the users of the data are going to be. It is increasingly less likely that producer and user share a common understanding and the interaction between them necessary to clarify any inconsistencies has been reduced. There are still some cases where the data producer provides more than a class label making it possible for a user unfamiliar with the semantics and ontology of the producer to process the text and assess the relationship between classes and between classifications. In this paper we apply computer characterization to the textual descriptions of two land cover maps, LCMGB (land cover map of Great Britain produced in 1990) and LCM2000 (land cover map 2000). Statistical analysis of the text is used to parameterize a look-up table and to evaluate the consistency of the two classification schemes. The results show that automatic processing of the text generates similar relations between classes as that produced by human experts. It also showed that the automatically generated relationships were as useful as the expert derived relationships in identifying change.

Spatial information systems like GIS assist a user in making a spatial decision by presenting information that supports the decision process. Planning a holiday via the Internet can be a daunting process. Decision-making is based on finding the relevant data sets in short time among an overwhelming amount of data sources. In many cases the user has to figure out how the presented information can fit the decision he has to make on his own. To address this problem a tourist would need tools to retrieve the spatial information according to his current preferences. The relevant data is determined as data elements describing facts corresponding to the tourist’s preferences. The present work suggests a way how a user dealing with a tourist information system can indicate his preferences through the user interface. According to the suggested alternatives the user can change his preferences and generate a new set of alternatives. A feedback loop provides a tool that allows the user to explore the available alternatives. The work was motivated by classical dialog based booking processes between human operators, carried out on the telephone. We introduce the conceptual model of a user interface, which considers the agents preferences and describes the overall interaction process.

Granularity in time and space has a fundamental role in our perception and understanding of various phenomena. Currently applied analysis methods are based on a single level of granularity that is user-driven, leaving the user with the difficult task of determining the level of spatiotemporal abstraction at which processing will take place. Without a priori knowledge about the nature of the phenomenon at hand this is often a difficult task that may have a substantial impact on the processing results. In light of this, this paper introduces a spatiotemporal data analysis and knowledge discovery framework, which is based on two primary components: the spatiotemporal helix and scale-space analysis. While the spatiotemporal helix offers the ability to model and summarize spatiotemporal data, the scale space analysis offers the ability to simultaneously process the data at multiple scales, thus allowing processing without a priori knowledge. In particular, this paper discusses how scale space representation and the derived deep structure can be used for the detection of events (and processes) in spatiotemporal data, and demonstrates the robustness of our framework in the presence of noise.

The ‘industry-strength’ data models are complex to use and tend to obscure the fundamental issues. Going back to the original proposal of Chen for Entities and Relationships, I describe here a reduced data model with Objects and Relations. It is mathematically well founded in the category of relations and has been implemented to demonstrate that it is viable. An example how this is used to structure data and load data is shown.

Query expansion describes the automated process of supplementing a user’s search with additional terms or geographic locations to make it more appropriate for the user’s needs. Such process relies on the system’s knowledge about the relation between geographic terms and places. Geodata repositories host spatial data, which can be queried over their metadata, such as keywords. One way to organize the system’s knowledge structure for keyword-based query expansion is to use a similarity network. In a complete similarity network the total number of similarity values between keyterms increases with the square of included keywords. Thus, the task of determining all these values becomes time consuming very quickly. One efficient method is to start with a sparse similarity network, and automatically estimate missing similarity values from other values with an algorithm. Hence, this paper introduces and evaluates four such algorithms.

The Tangible Augmented Street Map (TASM) is a novel interface to geographic objects, such as tiled maps of labeled city streets. TASM uses tangible Augmented Reality, the superimposition of digital graphics on top of real world objects in order to enhance the user’s experience. The tangible object (i.e. a cube) replicates the role of an input device. Hence the cube can be rotated to display maps that are adjacent to the current tile in geographic space. The cube is capable of theoretically infinite movement, embedded in a coordinate system with topology enabled. TASM has been tested for usability using heuristic evaluation, where selected experts use the cube, establishing non-correspondence with recognized usability principles. While general and vague, the heuristics helped prioritize immediate geographic and system-based tasks needed to improve the usability of TASM, also pointing the way towards a group of geographically oriented heuristics. This addresses a key geovisualization challenge — the creation of domain-specific and technology-related theory.

The quality of maps, geo-visualization and usage of multimedia presentation techniques for spatial communication is an important issue for map creation, distribution and acceptance of these information systems (IS) by a public community. The purpose of this paper is to present an evaluation method based on stochastic reasoning for supporting map designers. We investigate the applicability of Bayesian Belief networks and present a prototypical implementation. We will give an outlook to future research questions.

Topological predicates between spatial objects have for a long time been a focus of intensive research in a number of diverse disciplines. In the context of spatial databases and geographical information systems, they support the construction of suitable query languages for spatial data retrieval and analysis. Whereas to a large extent conceptual aspects of topological predicates have been emphasized, the development of efficient evaluation techniques for them has been rather neglected. Recently, the design of topological predicates for different combinations of spatial data types has led to a large increase of their numbers and accentuated the need for their efficient implementation. The goal of this paper is to develop efficient implementation techniques for them within the framework of the spatial algebra SPAL2D.