Face recognition by cortical multi-scale line and edge representations

Empirical studies concerning face recognition suggest that faces may be stored in memory by a few canonical representations. Models of visual perception are based on image representations in cortical area V1 and beyond, which contain many cell layers for feature extraction. Simple, complex and end-stopped cells provide input for line, edge and keypoint detection. Detected events provide a rich, multi-scale object representation, and this representation can be stored in memory in order to identify objects. In this paper, the above context is applied to face recognition. The multi-scale line/edge representation is explored in conjunction with keypoint-based saliency maps for Focus-of-Attention. Recognition rates of up to 96% were achieved by combining frontal and 3/4 views, and recognition was quite robust against partial occlusions.

Quantificação das capturas acessórias e rejeições de arrasto de crustáceos na costa do Algarve

Dissertação mest., Biologia Marinha, Universidade do Algarve, 2009

Viabilização da reutilização de efluente tratado, usos, aplicações e respectivos impactos na gestão dos mesmos

Dissertação mest., Engenharia Biológica, Universidade do Algarve, 2008

Perfil de ácidos gordos de larvas recém eclodidas produzidas em cativeiro de camarões marinhos ornamentais candidatos a aquacultura

Dissertação de Mestrado, Biologia Marinha, especialização em Pescas e Aquacultura, Faculdade de Ciências do Mar e do Ambiente, Universidade do Algarve, 2009

Feed-backs between genetic structure and perturbation-driven decline in seagrass (Posidonia oceanica) meadows

We explored the relationships between perturbation-driven population decline and genetic/genotypic structure in the clonal seagrass Posidonia oceanica, subject to intensive meadow regression around four Mediterranean fish-farms, using seven specific microsatellites. Two meadows were randomly sampled (40 shoots) within 1,600 m2 at each site: the “impacted” station, 5–200 m from fish cages, and the “control” station, around 1,000 m downstream further away (considered a proxy of the pre-impact genetic structure at the site). Clonal richness (R), Simpson genotypic diversity (D*) and clonal sub-range (CR) were highly variable among sites. Nevertheless, the maximum distance at which clonal dispersal was detected, indicated by CR, was higher at impacted stations than at the respective control station (paired t-test: P < 0.05, N = 4). The mean number of alleles (Â) and the presence of rare alleles ( r) decreased at impacted stations (paired t-test: P < 0.05, and P < 0.02, respectively, N = 4). At a given perturbation level (quantified by the organic and nutrient loads), shoot mortality at the impacted stations significantly decreased with CR at control stations (R 2 = 0.86, P < 0.05). Seagrass mortality also increased with  (R 2 = 0.81, P < 0.10), R (R 2 = 0.96, P < 0.05) and D* (R 2 = 0.99, P < 0.01) at the control stations, probably because of the negative correlation between those parameters and CR. Therefore, the effects of clonal size structure on meadow resistance could play an important role on meadow survival. Large genotypes of P. oceanica meadows thus seem to resist better to fish farm-derived impacts than little ones. Clonal integration, foraging advantage or other size-related fitness traits could account for this effect.

A relevância do justo valor: entidades com títulos cotados nas bolsas de Lisboa e de Madrid

Dissertação de Mestrado, Contabilidade, Faculdade de Economia, Universidade do Algarve, 2015