Catálogo de publicaciones - libros

We consider approaches to modeling spatial uncertainty with vague minimum bounding rectangles. Then two indexing approaches using grid files and R-trees are developed. Finally a number of spatial queries are illustrated based on these indexing schemes.

The popularity of spatial databases increases since the amount of the spatial data that need to be handled has increased by the use of digital maps, images from satellites, video cameras, medical equipment, sensor networks, etc. Spatial data are difficult to examine and extract interesting knowledge; hence, applications that assist decision-making about spatial data like weather forecasting, traffic supervision, mobile communication, etc. have been introduced. In this thesis, more natural and precise knowledge from spatial data is generated by construction of fuzzy spatial data cube and extraction of fuzzy association rules from it in order to improve decisionmaking about spatial data. This involves an extensive research about spatial knowledge discovery and how fuzzy logic can be used to develop it. It is stated that incorporating fuzzy logic to spatial data cube construction necessitates a new method for aggregation of fuzzy spatial data. We illustrate how this method also enhances the meaning of fuzzy spatial generalization rules and fuzzy association rules with a case study about weather pattern searching. This study contributes to spatial knowledge discovery by generating more understandable and interesting knowledge from spatial data by extending spatial generalization with fuzzy memberships, extending the spatial aggregation in spatial data cube construction by utilizing weighted measures, and generating fuzzy association rules from the constructed fuzzy spatial data cube.

A high-level interactive approaches for objects extraction (roads, rivers, forests, buildings, etc.) from remote sensing images and colorscanned maps are proposed based on connected components accumulation and live-wire techniques. In contrast to known interactive segmentation techniques, the proposed approach allows extraction of object centerline as well as boundary. The obtained experimental results are very promising and show that the proposed approach allows us solving object extraction task with high speed and quality.

The use of classifications for land cover mapping from satellite imagery is shown. Environmental and land cover maps represent the probable environmental statement of various types of land use and development of landscape. Remotely sensed data could be particularly efficient for environmental and land use mapping in order to outline main classification types.

In this paper, the problem of sustainable development is highlighted on the global level and for particular country (Ukraine). Strong and weak approaches to sustainable development and specific management tools, i.e., economic, institutional, social (education), and technological (GIS) are given to address environmental security issues. The integrated characteristic of environmental and human well-being (health) is proposed as the principal indicator of sustainable development.

Current status of implementation of Earth remote sensing technologies in Ukraine, structure of the Ukrainian Segment of GEOSSGMES, some examples of works for solving of nature management problems are offered in article.

The paper describes the ProDec system serving to develop and run decision procedures based upon IF…THEN rules. The system encompasses the editing, testing, and running functions. The key feature consists in the use of fuzzy values of the conditioning parameters. These fuzzy values are defined in the development and editing stage and can be easily referred to in the testing and use stages through natural language expressions. In this manner the inherent uncertainty associated with virtually all decision situations is captured and communication with the procedure is facilitated. The use of ProDec for river catchment emergency management purposes is illustrated.