Abstract
The ability to take advantage of the information society is becoming a must for everyone. Computer literacy is becoming more and more important for individuals as well as corporations. Therefore no efforts must be spared to integrate everyone in this new and exciting development, If for non-disabled persons this can be a challenge, for certain types of disabilities the effort of adaptation is huge. In this paper we focus on blind users. One of the current alternatives for blind users is based on specific hardware that tends to be limited in features and expansibility or very expensive. The other alternative is PC based software. This latest option, although very powerful, has also several disadvantages, namely the lack of practical mobility. In this paper we present a new approach the PalmPCs. PalmPCs are a cheap alternative compared to the hardware mentioned above, they are truly mobile, and they are widely available today. We argue that software can be written for these devices that satisfies the needs of the majority of the blind users. A small application for WindowsCE is presented in which the interface was specifically designed for blind users.

Abstract
A new paradigm far the description of image patter ns is presented. it is proposed that images are described by fuzzy IF... THEN rules instead of pixel values. This new approach may benefit from recognised fuzzy systems superior incorporation of measurement uncertainties, greater resources for managing complexity and better ability to deal with natural language [1]. The concept of relevance has been proposed as a measure of the relative importance of sets of rules [2] [3]. Based on this concept a new methodology was developed: SLIM (Separation of Linguistic Information Methodology) [2] [3]. An algorithm implementing SLIM is presented in this paper, derived from the fuzzy Gmeans clustering algorithm, here applied to organise the fuzzy IF... THEN rules that describe the image. The Lena and the Abington Grass images have been successfully used to illustrate the identification process. The proposed SLIM algorithm has been successfully applied to illustrate a segmentation operation in the "fuzzy rules domain", using the Abington Cross image.

Abstract
There is a need for reliable lighting design applications because available tools are limited and inappropriate for interactive or creative use. Architects and lighting designers need those applications to define, predict, test and validate lighting solutions for their problems. We present a new approach to the lighting design problem based on a methodology that includes the gelometry of the scene, the properties of materials and the design goals. It is possible to obtain luminaire characteristics or other kind of results that maximise the attainment of the design goals, which may include different types of constraints or objectives (lighting, geometrical or others). The main goal, in our approach, is to improve the lighting design cycle. In this work we discuss the use of optimisation in lighting design, describe the implementation of the methodology, present real-world based examples and analyse in detail some of the complex technical problems associated and speculate on how to overcome them.

Abstract
This paper presents a description of the image generation sub-system developed to allow the presentation of realistic visual feedback in interactive visual simulation with large scene databases. The developed image generator applies all the standard state-of-art image generation algorithms aimed to real-time interactive simulation. Some of these important algorithms are also explained in this document. I addition, some innovative optimization techniques like the hierarchical back face rejection of objects, the visibility preprocessing and the automatic optimization of levels-of-detail are being developed and detailed in this paper. These techniques will allow a better use of any image generation system and improve significantly the visualization of huge scene databases even in high-end graphics architectures.

Abstract
In this paper we introduce a classification for spatio-temporal systems based on the properties of the represented objects. Building on this classification, we assert that complex objects can be derived from simpler ones, following an evolutionary approach which starts with the study of simple objects and ends by enriching them with new features. This paper focuses on the definition of a data model for the representation of moving points. The model is based on the decomposition of the trajectory of moving points into sections. The movement within each section of a trajectory is described by a variability function. Because for most systems is not possible to store the exact knowledge about the movement of an object, the answers to queries may be imprecise. We propose two additional approaches to deal with impression-the superset and the subset semantics-based on a maximum value for the variability function, and a smooth technique to integrate them in the model. Finally, we analyze certain functional aspects of the implementation of the data model in a Relational Database Management System (RDBMS) and outline directions for future research.

Abstract