V. Di Lecce

Fuzzy Clustering With Partial Supervision in Organization and Classification of Digital Images

In a Web-oriented society, organization, retrieval, and classification of digital images have become one of the major endeavors. In this paper, we study the mechanisms of fuzzy clustering and fuzzy clustering with partial supervision in the analysis and classification of images. It is demonstrated that the main features of fuzzy clustering become essential in revealing the structure in a collection of images and supporting their classification. The discussed operational framework of fuzzy clustering is realized by means of fuzzy c-means (FCM). When dealing with the mode of partial supervision, we augment an original objective function guiding the clustering process by an additional component expressing a level of coincidence between the membership degrees produced by the FCM and class allocation supplied by the user(s). The study also contrasts the use of the technology of fuzzy sets in image clustering with other approaches studied in this area. A suite of experiments deals with two collections of images, namely, Columbia object image library (COIL-20) and a database composed of 2000 outdoor images.

Selection of reference signatures for automatic signature verification

This paper presents an effective procedure to select the reference specimens for a signature verification system. Specifically, from the analysis of local stability in handwritten signatures, a suitable measure is proposed to determine the capability of different sets of signatures in supporting effective verification. The measure uses a correlation-based criterium which detects and recovers non-linear time distortions in different specimens. In the experimental test, the selected set of signatures has been used for reference in a system for dynamic signature verification based on a multi-expert verification strategy. The experimental results points out the capability of the new technique in selecting effective reference signatures.

Model characterization in measurements of environmental pollutants via data correlation of sensor outputs

The aim of this paper is to develop a simple air pollutant model for the analysis and the characterization of environmental data, acquired by means of multisensor monitoring system and elaborated by suitable software agents. Modeling is basically important in order to validate measured data and to forecast the time-varying behavior of contaminating substances. The proposed model is initially based on the individuation of possible correlations existing between some pollutants. Afterward, to increase the accuracy of estimated values, the influence of meteorological quantities is taken into account to improve significantly the so-obtained model. Finally, some information about reliability degree of estimate is provided.

A basic ontology for multi agent system communication in an environmental monitoring system

Air quality monitoring system is characterized by a large number of information sources used by experts capable of understanding the effects of single pollutants. By using an adequate ontological approach, it is possible to define a system having the ability of doing data mining and giving information to unskilled users too. To do this, we propose in this paper a multiagent system (MAS), layered in five levels, suitable to supply answer to a query characterized by a high semantic level. This is possible using progressive interpreting/multiplying techniques of a complex query in simple queries according with well-known compilers and OS theories. We develop a multiagent system that assists users in generating a uniform description for each information source, using descriptive domain ontology. Users and agents can query the extracted data using a standard querying interface. The ultimate goal is to provide useful information to users, supporting distributed workflow management environments.

Neural technologies for increasing the GPS position accuracy

Aim of this paper is to present a method to improve the accuracy of a GPS receiver. It is well known that there are many factors affecting the accuracy of a GPS receiver. In this work, the authors point out that many of these factors, considered in a given geographic area, have a certain periodicity. An important example of this kind of factors is the sky satellite position relative to receiver. The proposed method uses a neural network to correct the position computed by the receiver. The neural network is trained to learn the errors introduced into the measuring system by the cyclic phenomenon in the various hours of the day.

GPS-aided lightweight architecture to support multi-sensor data synchronization

GPS sentences carry the UTC time information. In the case of extended sensor grids or world wide sensor networks the UTC information appears to be very attractive but one risk is to have degraded timing accuracy imputable to much noise on the signal transmission path. Well known problems are time delays due to ionosphere, troposphere or receiver hardware specifics. The software phase instead is often skipped although it can represent actually a noisy element for the correct synchronization. This paper presents a GPS-based lightweight low-cost system architecture that can support multi-sensor data synchronization through an accurate timestamp of the incoming data streams. The proposed architecture handles both hardware and software problems in order to achieve correct post-processing data synchronization by means of software timestamping. Particular attention is paid to the combined hardware/software solution that minimises the overall delay time before the software timestamp event. The proposed architecture is also suitable to provide accurate evaluation of software algorithms impact over multi-sensor measurements.

Image basic features indexing techniques for video skimming

In this paper a comparison of the most widespread automatic indexing techniques, suitable in skimmed video generation, and their performances is presented. To evaluate the performances, using the low-level frame features, the signatures are computed, the shots are identified using neural network clustering techniques, in each shot the mean distance between contiguous frames is computed and the shot is resampled according to a related distance value to produce a skimmed video sequence. The most relevant feature proves to be the angular spectrum. Using this feature the mean value of the skimming factor is 2.6 in the used test set.

Web agents in an environmental monitoring system

agents are autonomous, reactive and pro-active problem solvers. They co-operate to achieve the overall goal of a system. Numerous systems are based on mutual support and/or competition of agents, which are developed to satisfy the queries of a human user. Systems of multi-agent acquirement (agencies) extracting clear information for the user have been developed during the monitoring of air quality. An evolution of the system is necessary to expand the basis of knowledge of the agency and estimate that can perform, as well as possible, the service offered by MAS. In this paper we will illustrate a proposal, which has been planned to improve a MAS, previously developed and employed for the monitoring of air quality. The solution proposed by the authors prefigures the introduction of an agent up to provide main information about the reliability of the forecasting models in use. This resolution has been planned and implemented during the test phase, on a real system.

Taxonomies and Ontologies in Web Semantic Applications: The New Emerging Semantic Lexicon-Based Model

This paper addresses the new emerging approach of Semantic Lexicon-based systems for modeling semantic Web applications. The paper endeavors to shed lights on some misconceptions in the literature about the use of taxonomy and ontology as interchangeable terms, defining a midpoint between the two extremes. Semantic Lexicon is considered as the right match between the semantic layer represented by a given ontology and the lexical layer represented by a given taxonomy. Agent-based implementations that employ WordNet as Semantic Lexicon are currently being tested within this framework.

An Intelligent System for Decentralized Load Management

This work proposes a model of an intelligent short term demand side management system based on a MAS. The system is designed to avoid peaks of power request greater than a given threshold and to give maximum comfort to user. The proposed system is composed of a distributed network of processing nodes (PN). Each PN hosts one agent and it is able to manage a single socket tap allowing or disallowing it to supply power. Each agent reacts to a new critical condition entering in competition with the others to gain the access at a shared limited resource. As the results shown the proposed agency can be the consumers key to take advantage of a DSM program automatically