OPF-MRF: Optimum-path forest and Markov random fields for contextual-based image classification

Some machine learning methods do not exploit contextual information in the process of discovering, describing and recognizing patterns. However, spatial/temporal neighboring samples are likely to have same behavior. Here, we propose an approach which unifies a supervised learning algorithm - namely Optimum-Path Forest - together with a Markov Random Field in order to build a prior model holding a spatial smoothness assumption, which takes into account the contextual information for classification purposes. We show its robustness for brain tissue classification over some images of the well-known dataset IBSR. © 2013 Springer-Verlag.

A wrapper approach for feature selection based on Bat Algorithm and Optimum-Path Forest

Besides optimizing classifier predictive performance and addressing the curse of the dimensionality problem, feature selection techniques support a classification model as simple as possible. In this paper, we present a wrapper feature selection approach based on Bat Algorithm (BA) and Optimum-Path Forest (OPF), in which we model the problem of feature selection as an binary-based optimization technique, guided by BA using the OPF accuracy over a validating set as the fitness function to be maximized. Moreover, we present a methodology to better estimate the quality of the reduced feature set. Experiments conducted over six public datasets demonstrated that the proposed approach provides statistically significant more compact sets and, in some cases, it can indeed improve the classification effectiveness. © 2013 Elsevier Ltd. All rights reserved.

Insights of meso and micro-scale processes for the Caxiuanã forest region from high resolution simulation

Dados meteorológicos e simulações numéricas de alta resolução foram usados para estimar campos espaciais na região leste da Amazônia onde se situam a Floresta e a Baía de Caxiuanã, no Estado do Pará. O estudo foi feito para o período de Novembro de 2006, quando foi realizado o experimento de campo COBRA-PARÁ. Análises de imagens do sensor MODIS mostram a ocorrência de vários fenômenos locais como avenidas de nuvens, sistemas convectivos precipitantes, e importante influência das interfaces entre a floresta e as superfícies aquáticas. Simulações numéricas para o dia 7 de novembro de 2006 mostraram que o modelo representou bem as principais variáveis meteorológicas. Os resultados mostram que a Baía de Caxiuanã provoca importante impacto nos campos meteorológicos adjacentes, principalmente, através da advecção pelos ventos de nordeste que induzem a temperaturas do dossel mais frias a oeste da baía. Simulações de alta resolução (LES) produziram padrões espaciais de temperatura e umidade alinhados com os ventos durante o período diurno e mudanças noturnas causadas principalmente pela presença da baía e chuvas convectivas. Correlações espaciais entre os ventos de níveis médios e os fluxos verticais de calor latente mostraram que existe uma mudança de correlações negativas para as primeiras horas do dia passando para correlações positivas para o período da tarde e início da noite.

Cryptic Genetic Diversity Is Paramount in Small-Bodied Amphibians of the Genus Euparkerella (Anura: Craugastoridae) Endemic to the Brazilian Atlantic Forest

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Perceptions of environmental change and use of traditional knowledge to plan riparian forest restoration with relocated communities in Alcantara, Eastern Amazon

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Feedbacks between vegetation and disturbance processes promote long-term persistence of forest-grassland mosaics in south Brazil

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

SIMULATION OF USE AND MANAGEMENT EFFECTS ON THE SOIL ORGANIC MATTER POOLS OF THE ATLANTIC FOREST BIOME, BRAZIL

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Soil erodibility assessment in a pasture and forest remnant using the Indrbitzen device

Brazil has the largest cattle herd in the world with approximately 200 million head. An important feature of the Brazilian cattle industry is that most of its herd is raised on pasture, which constitutes one of the most economical and practical ways to produce and provide food for cattle. However, this production model is mishandled and can lead to soil degradation. Maintaining soil quality is essential for the conservation of natural ecosystems and the areas of production, thus soil quality improves the conditions for biogeochemical cycles. In this context, the objective of this study was to develop a device for testing the Inderbitzen way of assessing soil erodibility in two situations of usage and occupation. Therefore, one area was used as a sample collection occupied by grazing and the other as a forest fragment; both located in the city of Sorocaba in Sao Paulo State, Brazil. Thus, we concluded that the proposed device – the Inderbitzen – proved capable of assessing soil erodibility of the pasture and remnant forest. Accordingly, there was a tendency for a smaller loss of forest soils in the remnant when compared to the degraded pasture. The greatest resistance of the soil erosion in the forest remnant may be associated with the amount of organic matter released by the forest litter in all its diversity, influencing the quality of the structure of aggregates. Keywords: erosion, forest remnant, degraded pasture, Inderbitzen test.

Barriers, rather than refugia, underlie the origin of diversity in toads endemic to the Brazilian Atlantic Forest

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

The Priestley-Taylor parameter α for the Amazon forest

Priestley and Taylor provided a practical formulation of the partitioning of net radiation between heat flux and evaporation contained within a parameter α. Their model (PTM) needs verification under a range of environmental conditions. Micrometeorological data sets collected over the Amazon forest at the Ducke Reserve site (2°57′S; 59°57′W) gave an opportunity to evaluate α. Evidence presented here and by others shows that there is pronounced diurnal variation in α, with minimum values around midday and maximum values in the morning and evening hours. During unstable and stable conditions in the daylight hours, the Bowen ratio (B) varied from 0.10 to 0.57 and -0.71 to -0.08, respectively, whereas α varied from 0.67 to 1.16 and 1.28 to 3.12, respectively. A mean value of α = 1.16±0.56 was obtained from daytime hourly values for two days. The daily data sets from three expeditions gave a mean of α = 1.03±0.13. This work confirms that α is a function of atmospheric stability over the Amazon forest. Thus the PTM should be applied with caution over time-intervals of one day or less because of the sensitivity to variation in α. The calculated values of α are in general agreement with those reported in literature. © 1991.

Real-time Sampling of Particulate Matter Smaller than 2.5 µm from Amazon Forest Biomass Combustion

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Tree height integrated into pantropical forest biomass estimates

Aboveground tropical tree biomass and carbon storage estimates commonly ignore tree height (H). We estimate the effect of incorporating H on tropics-wide forest biomass estimates in 327 plots across four continents using 42 656 H and diameter measurements and harvested trees from 20 sites to answer the following questions: 1. What is the best H-model form and geographic unit to include in biomass models to minimise site-level uncertainty in estimates of destructive biomass? 2. To what extent does including H estimates derived in (1) reduce uncertainty in biomass estimates across all 327 plots? 3. What effect does accounting for H have on plot- and continental-scale forest biomass estimates? The mean relative error in biomass estimates of destructively harvested trees when including H (mean 0.06), was half that when excluding H (mean 0.13). Power- and Weibull-H models provided the greatest reduction in uncertainty, with regional Weibull-H models preferred because they reduce uncertainty in smaller-diameter classes (< 40 cm D) that store about one-third of biomass per hectare in most forests. Propagating the relationships from destructively harvested tree biomass to each of the 327 plots from across the tropics shows that including H reduces errors from 41.8 Mg ha(-1) (range 6.6 to 112.4) to 8.0 Mg ha(-1) (-2.5 to 23.0).

Two new species of marsupial tree-frogs genus Gastrotheca Fitzinger, 1843 (Anura, Hemiphractidae) from the Brazilian Atlantic Forest

Two new species of Gastrotheca are described from northeastern Minas Gerais and southern Bahia, in the Atlantic Forest of Brazil. Data on morphology, calls, mitochondrial, and nuclear DNA are provided. Allied to G. fissipes and G. megacephala, the new taxa provide evidence for a higher diversity of species of Gastrotheca than previously thought at the Atlantic Forest. The data also suggest that G. pulchra, another Atlantic Forest taxon, is more closely related to non-Atlantic Forest species than to the remaining analyzed Brazilian Gastrotheca species. This implies that the Gastrotheca at the Brazilian coastal forests have at least two independent origins.

Bioaccumulation pattern of lanthanides in pteridophytes and magnoliophytes species from Atlantic Forest

The availability of chemical elements for plants is mainly dependent on the nature of the soil and characteristics of each species. The transfer factors of lanthanides from the soil to the tree leaves of the Atlantic Forest, Brazil, were calculated for one fern species (Alsophila sternbergii-Pteridophyta division) and four magnoliophytes species (Bathysa australis, Euterpe edulis, Garcinia gardneriana and Guapira opposita-Magnoliophyta division) obtained in two areas of Serra do Mar State Park and collected in two different seasons. Samples were analyzed by instrumental neutron activation analysis (INAA). The soil-to-plant transfer factor (TF = C(plant):C(soil)) in magnoliophytes species was correlated to the mass fraction of lanthanides in the soil, described by a exponential model (TF = a.C (soil) (-b) ). Despite the tree fern Alsophila sternbergii presented a hyperaccumulation of lanthanides, this species did not have a significant relationship between TF and mass fraction in soil. Results indicated that plants of Magnoliophyta division selected the input of lanthanides from the soil, while the same was not observed in Alsophila sternbergii.

Spatial distribution of arboviral mosquito vectors (Diptera, Culicidae) in Vale do Ribeira in the South-eastern Brazilian Atlantic Forest

Mosquitoes are vectors of arboviruses that can cause encephalitis and hemorrhagic fevers in humans. Aedes serratus (Theobald), Aedes scapularis (Rondani) and Psorophora ferox (Von Humboldt) are potential vectors of arboviruses and are abundant in Vale do Ribeira, located in the Atlantic Forest in the southeast of the State of Sao Paulo, Brazil. The objective of this study was to predict the spatial distribution of these mosquitoes and estimate the risk of human exposure to mosquito bites. Results of the analyses show that humans are highly exposed to bites in the municipalities of Cananeia, Iguape and Ilha Comprida. In these localities the incidence of Rocio encephalitis was 2% in the 1970s. Furthermore, Ae. serratus, a recently implicated vector of yellow fever virus in the State of Rio Grande do Sul, should be a target for the entomological surveillance in the southeastern Atlantic Forest. Considering the continental dimensions of Brazil and the inherent difficulties in sampling its vast area, the habitat suitability method used in the study can be an important tool for predicting the distribution of vectors of pathogens.