Percorrer por data de Publicação, começado por "2014-08"
A mostrar 1 - 4 de 4
Resultados por página
Opções de ordenação
- A stochastic approach to optimize Maritime pine (Pinus pinaster Ait.) stand management scheduling under fire risk. An application in PortugalPublication . Ferreira, L.; Constantino, M.; Borges, J. G.The paper discusses research aiming at the development of a management scheduling model for even-aged stands that may take into consideration fuel treatments to address the risk of wildfires. A Stochastic dynamic programming (SDP) approach is proposed to determine the policy (e.g. the fuel treatment and thinning schedules and the rotation age) that produces the maximum expected discounted net revenue. Fuel treatment activities encompass shrub cleanings. Emphasis was on combining a deterministic stand-level growth and yield model with wildfire occurrence and damage models to design a SDP network. SDP stages are defined by age and state variables include both the stand basal area and the number of years since the last fuel treatment. Fire occurrence and damage scenarios are addressed at each stage. Results from an application to Maritime pine (Pinus pinaster Ait.) stand management scheduling in Leiria National Forest, Portugal, are presented. Results suggest that the modeling strategy may help assess the impact of wildfire risk on the optimal stand management schedule. They confirm that the maximum expected discounted net revenues decreases. Further, albeit some timber may be salvaged after the wildfire, rotation age also decreases when the risk of fire is considered. Finally, they provide interesting insights about the role of thinning and fuel treatment policies in mitigating risk.
- Internet of Things Aware WS-BPEL Business Processes - Context Variables and Expected ExceptionsPublication . Domingos, Dulce; Martins, Francisco; Cândido, Carlos; Martinho, RicardoBusiness processes can use Internet of Things (IoT) information to monitor context data in real-time and to respond to changes in their values in a timely fashion. For this matter, business process definition and execution languages should foresee an easy way for process modelers to define which values to monitor, and which automatic behaviors to adopt when these values change. In this paper, we propose the use of context variables to monitor sensor values, as well as a when-then language construct to detect and handle changes in these values within business processes.We define aWeb Services Business Process Execution Language (WS-BPEL) extension to convey these constructs, and implement then using a “BPEL language transformation” approach. With these contributions, process modelers can define IoT-aware business processes avoiding the increase of process complexity and keeping their focus on modeling the processes’ main logic. In addition, the language transformation approach assures the portability of processes using our constructs amongst WS-BPEL execution engines.
- Human milk fat substitutes: Advances and constraints of enzyme‐catalyzed productionPublication . Ferreira-Dias, Suzana; Tecelão, CarlaThe production of triacylglycerols (TAGs) with similar structures to the human milk fat TAGs is only possible using sn-1,3 regioselective lipases as biocatalysts. The advances on the processes currently performed for the production of human milk fat substitutes (HMFS), namely the biocatalysts used, the best operation conditions, reactor types and operation modes, as well as the major drawbacks of these enzymatic systems, are addressed in this article.
- Camera reading for blind peoplePublication . Neto, Roberto Ferreira; Fonseca, Nuno Miguel da Costa SantosA ausência de visão torna a vida de um invisual bastante difícil. No entanto, a utilização da tecnologia pode melhorar ligeiramente pequenos aspectos do dia-a-dia. Nesse contexto, o trabalho que se apresenta tem como objetivo a descrição do processo de desenvolvimento de uma aplicação para cegos. O projeto chama-se Camera Blind For Blind People, e tem como propósito final o desenvolvimento de uma aplicação que permita a um utilizador cego utilizar um dispositivo móvel para obter a leitura de um texto que esteja escrito numa folha de papel, num sinal, numa parede ou noutro suporte. Pretende-se com este projeto conceber o protótipo de uma aplicação para iOS, construída a partir da utilização conjunta e integrada de frameworks de reconhecimento óptico de caracteres (OCR) e de frameworks de sintetização de voz (TTS), que possibilite ao utilizador, recorrendo à câmara de um dispositivo móvel, captar uma imagem e obter a leitura do texto que exista nessa mesma imagem. O processo de reconhecimento do texto através do OCR será optimizado através da aplicação de filtros nas imagens captadas, realizando desta forma um pré-processamento que permita obter melhorias nos resultados do OCR, para que o texto reconhecido se aproxime, o mais possível, do texto original existente no suporte fotografado. Os filtros utilizados são os disponibilizados pelo sistema operativo iOS, através da sua framework de manipulação de imagem, Core Image. A aplicação será afinada e optimizada de forma a oferecer uma melhor usabilidade aos seus utilizadores, nomeadamente nas situações em que a leitura OCR apresente resultados insatisfatórios que tornem imperceptíveis a leitura dos textos reconhecidos.