Domestic Environment Monitoring with Opportunistic Sensor networks (DEMOS)

Javier Baliosian, Eduardo Grampín, Jorge Visca, Leonado Vidal, Martín Giachino. Universidad de la República, Uruguay.

Lisandro Zambenedetti Granville, Luciano Paschoal Gaspary. Universidade Federal do Rio Grande so Sul, Brazil.

Several Latin American countries such as Uruguay and Brazil are implementing the well-known One Laptop Per Child Program by which, every child in primary school obtains in property a laptop with wireless capabilities. They carry their laptops from home to school and back every day and, as seen in the experience, they also carry their laptops to parks, community centers etc. The objective of this project is to take advantage of that computers ubiquity, mobility and wireless capabilities to obtain, transport, aggregate and communicate environmental information and risks from a network of small sensors. Those sensors should be deployed inside houses, at schools or outside, at parks or other public places which are visited by the most environmentally vulnerable children during their daily life. The environmental data collected by those sensors will be transmitted, using opportunistic networking techniques, to the children’s laptops as their move in their daily life. Later at the school, using the same techniques, the data will be transmitted to an environment monitoring station using the Internet. This monitoring station may be operated by governmental or non-governmental organizations or even an on-line facility for the open control by the same community that is object of the monitoring.

The results of this project will be a combination of software, processes and tools. With more detail, the results will include: (i) an environment monitoring network architecture, (ii) algorithms and protocols for the recollection and aggregation of environmental data (iii) software tools for visualization and analysis of the collected data and (iv) an experimental deployment of the solution.

This project will be developed and experimented in Uruguay and Brazil. However, as several developing countries are deploying or considering OLPC-like programs for their most vulnerable children, this project can be easily replicated on those countries. Additionally, the project idea is also applicable to other communication platforms that are becoming ubiquitous in the Latin-American countries such as cellular phones with Bluetooth or other wireless capabilities.

TimeSaver Decentralized Virtual Worlds for the Provision and Integration of Public Services in Latin America

Rosa Alarcón. Pontificia Universidad Católica de Chile, Chile.

Flavio Soares Correa da Silva, Universidade de Sao Paulo, Brazil.

Semantic-based interoperability, mobility and virtual worlds have captured the focus of attention as promising technologies for e-Government. Current research considers mostly services provided within country borders while cross-border services remain a challenge as technological infrastructures and cultures strongly differ. We propose the development of a distributed software architecture, called TimeSaver, to build a cluster of special purpose, decentralized virtual worlds for the provisioning of public services (e.g. document issuing) across Latin American countries. We shall focus mainly on the semantic-interoperability and integration of services, through the appropriate use of knowledge representation techniques, as well as natural metaphors for modeling interaction and information flow in 3D virtual worlds.

LOW COST COMPUTER BASED SYSTEM FOR QUALITY EVALUATION AND PRESERVATION OF GRAINS STORED IN POLYMER BAGS

Claudia Pons , Paul Puleston, Juan Pons, Gabriel Baum. Universidad Nacional de la Plata. Argentina.

Carlos Luna, Daniel Calegari. Universidad de la República. Uruguay.

Nora Szasz. Universidad ORT. Uruguay.

Josefina Marinissen. Instituto Nacional de Tecnología Agropecuaria. Argentina

We observe a great expansion in Argentina, Uruguay and Chile of an ad-hoc low-cost storage technique named “harvest bags” that consists in keeping the grains into hermetic polyethylene bags which are stored in the same field of crop (on-farm). The term “harvest bags” is colloquial and other terms are often used, e.g. “silobags”. In Argentina the most common term is “silobolsas”. It is likely that several million tons of grain will be stored in harvest bags onfarm and by private storage and stockfeed enterprises within the next years, in Latin America countries and also in other parts of the world. Research confirms that although harvest bags have some limitations, they offer growers a relatively cheap and reliable grain storage solution. Therefore, it is important to work towards the improvement of the harvest bag technology; in particular, the incorporation of computer technology would be a valuable asset. We believe that a low-cost intelligent software system for monitoring and adapting the internal conditions of the grain stored into silo bags can be of great help to improve the quality standards of this type of storage system. Besides, we plan to apply advanced software engineering methodologies for the development of such software system, which will bring us additional benefits.

E-CLOUDSS : BUILDING E-GOVERNEMENT CLOUDS USING DISTRIBUTED SEMANTIC SERVICES

Genoveva Vargas-Solar, José Luis Zechinelli Martini. Universidad de las Américas, Puebla. Mexico.

Martin Musicante, Paulo Pires. Universidade Federal do Rio Grande do Norte. Brazil.

Regina Motz, Alberto Pardo. Universidad de la República. Uruguay.

The objective of E-CLOUDSS is to propose an infrastructure for mashing up reliable semantic services for building egovernment clouds. Mashups represent a new wave for building Web applications. A mashup is an application that presents content available from different sources by reusing the contents provided by third parties (e.g Web pages, Web services). A service is an autonomous software that offers some functionality through a network. Services coordination can be used for mashing up services and thereby specifying application logic by (i) capturing the interactions and dependencies among services; and (ii) by specifying reliability properties that the system must ensure in order to ensure QoS requirements such as connection, security, availability, persistency, etc. Therefore the project addresses the specification of reliable semantic web services. E-CLOUDSS addresses the management (definition and enforcing at execution time) of non functional properties associated to services’ coordination for building reliable mash-ups. Effective ways to perform virtual executions is one of the main subjects of study of E-CLOUDSS. Depending on the knowledge and optionality degree of the concrete WS, different compilation scenarios arise which derive in turn into different execution mechanisms of the orchestration. E-CLOUDSS proposes to study the combination of the transformations using program fusion and generic programming.

E-CLOUDSS contributes to the mashing up of services for building e-government clouds and contributes to the development of solutions for promoting the economic and technologic development of the Latin American governments administrations. Technology transfer is one of the goals of the project. E-CLOUDSS will develop general fundamental research topics that are in the centre of advanced research in systems construction and architectures. ECLOUDSS will contribute to show the strong points of LA fundamental research on Computer Science and Information Technologies. Concerning social impact, e-government and e-economy must be developed and promoted in LA. Global markets and production chains are a reality in any region of the world and particularly in LA. Thus, having tools for implementing systems for supporting information and data exchange, along with the automatization of processes, will certainly contribute to the economic development of the region. Separating the semantic and the functional aspects for mashing up services, makes the approach easy to reproduce for any country, by modeling the particularities of the context. The consortium of E-CLOUDSS will also help to have a generic solution adaptable for the three participating countries.

A REAL TIME SYSTEM BASED ON COMPUTER VISION TECHNIQUEs TO SUPERVISE AND ALLOCATE CASH REGISTERS AT GROCERY STORES

Álvaro Soto, Domingo Mery. Pontificia Universidad Católica de Chile. Chile.

Enrique Sucar. Instituto Nacional de Astrofísica Óptica y Electrónica. Mexico.

The field of Computer Vision is in a prelude to deliver a new generation of highly versatile machine vision applications that will widely impact diverse areas of our society. The service industry, one of the most relevant areas in the economy of countries such as Chile and Mexico, is one of the clear candidates to benefit from new applications of computer vision technology. In this area, several tasks, such as monitoring waiting times in queues, supervising the interaction of users with products, or detecting patterns of navigation in stores, require some level of visual perception. Until now, there is still no off-the-shelf technology to assist or automate these tasks, and the intensive or prohibitive levels of manual labor to perform them, limit their execution to occasional cases, mainly to obtain data to calibrate off-line models. In this project we propose the development of a system to solve one of the relevant problems of retail companies: the supervision and allocation of cash registers in grocery stores. The intended application will use computer vision technology to continually measure relevant parameters related to the level of service at checkout lines, such as waiting times, attention times, and number of incidents that cause delays. These measurements will feed a decision system that based on past and current measurements will periodically suggest an optimal number of check-out points to achieve a desired level of service in terms of waiting times. In

technical terms, this proposal is rooted in the combination of computer vision, machine learning and decision theory. We believe that the learning scheme is the right path to cope with the high ambiguity and complexity of the visual world.

Furthermore, we foresee that by adding a layer of decision making, we can depart from the traditional passive measurement scenario of computer vision applications to a case where we include suitable actions that directly satisfy a specific need of the target domain. We believe that this is just the tip of an iceberg of a wide variety of applications that will mark a new era of computer vision technology.