Ana Lúcia Bezerra Candeias

Fuzzy Kohonen clustering networks for interval data

The Fuzzy Kohonen Clustering Network combines the idea of fuzzy membership values for learning rates. It is a kind of self-organizing fuzzy neural network that can show great superiority in processing the ambiguity and the uncertainty of data sets or images. Symbolic data analysis provides suitable tools for managing aggregated data described by intervals. This paper introduces Fuzzy Kohonen Clustering Networks for partitioning interval data. The first network is based on a fixed Euclidean distance for interval and the second one considers weighted distances that change at each iteration, but are different from one cluster to another. Experiments with real and synthetic interval data sets demonstrate the usefulness of these networks.

Image segmentation of ovitraps for automatic counting of Aedes Aegypti eggs

The Aedes Aegypti mosquito is the vector of the most difficult public health problems in tropical and semi-tropical world: the epidemic proliferation of dengue, a viral disease that can cause human beings death specially in its most dangerous form, dengue haemorrhagic fever. One of the most useful methods for mosquito detection and surveillance is the ovitraps: special traps to collect eggs of the mosquito. It is very important to count the number of Aedes Aegypti eggs present in ovitraps. This counting is usually performed in a manual, visual and non-automatic form. This work approaches the development of automatic methods to count the number of eggs in ovitraps images using image processing, particularly color segmentation and mathematical morphology-based non-linear filters.

Texture classification based on co-occurrence matrix and self-organizing map

This article presents a hybrid approach for texture-based image classification using the gray-level co-occurrence matrices (GLCM) and self-organizing map (SOM) methods. The GLCM is a matrix of how often different combinations of pixel brightness values (grey levels) occur in an image. The GLCM matrices extracted from an image database are processed to create the training data set for a SOM neural network. The SOM model organizes and extracts prototypes from processed GLCM matrices. This paper proposes a novel strategy to index match scores by searching through prototypes. A benchmark data set is used to demonstrate the usefulness of the proposed methodology. The evaluation of performance is based on accuracy in the framework of a Monte Carlo experience. This approach is compared with several classifiers in Li et al [1]. The experimental results on the Brodatz texture image database demonstrate that the proposed method is encouraging with an average successful rate of 97%.

Biophysical parameters in the detection of changes in soil cover and use in watersheds.

O estudo de parâmetros biofisicos como o Indice de Vegetacao da Diferenca Normalizada (NDVI), albedo e temperatura da superficie (LST), aplicado a ecossistemas, tem sido relevante para o entendimento de mudancas relacionadas a degradacao do meio ambiente. Algumas alteracoes que provocam desequilibrio de interacoes ecologicas em ecossistemas, como o desmatamento, a mineracao, a agricultura inadequada e o superpastejo, entre outros, estao interrelacionadas. Para avaliacao de mudancas temporais relacionadas a degradacao do ecossistema caatinga obtiveram-se as imagens da diferenca de tres parâmetros: NDVI, albedo e temperatura da superficie, para os anos de 1985 e 2001 (estacao seca), utilizando-se imagens TM. Este estudo foi aplicado a bacia do rio Brigida onde ha uma exploracao intensa dos recursos naturais, em que os resultados mostram aumento na temperatura da superficie, diminuicao do NDVI e pouca variacao no albedo da superficie evidenciando, assim, que entre os anos de 1985 e 2001 houve avanco na degradacao dos recursos naturais, nesta bacia.

IFKCN: Applying fuzzy Kohonen clustering network to interval data

The recording of interval data has become a common practice in real world applications and nowadays this kind of data is often used to describe objects. In this paper, we introduce a new fuzzy Kohonen clustering network for symbolic interval data (IFKCN). The network combine the idea of fuzzy membership values for learning rates and the algorithm is able to show superiority in processing the ambiguity and the uncertainty present in data sets. Experiments with benchmark interval data sets and an artificial interval data set for evaluating the usefulness of the proposed method were carried out.

Espacialização da umidade do solo por meio da temperatura da superfície e índice de vegetação

The study of soil moisture is important in determining the resilience of ecosystems and their recovery, as well as in the modeling of water-vegetation-atmosphere relationship. Remote sensing is an important tool for the acquisition, mapping and monitoring soil moisture through the surface temperature and vegetation indices. For the soil moisture content assessment, several methods have been proposed, however its application is still limited. In this work the soil moisture index (SMI) was applied and modeled with the objectives: to establish and delineate areas with different levels of desertification through SMI mapping and to map the dynamic of border, as well as to verify possible relationships betweem SMI and soil water parameters. In the application of this model as input data was used: NDVI (normalized difference vegetation index) and LST (land surface temperature). It was observed that SMI accessed by the average of the SMI derived by NDVI and LST can be used in the determination of soil surface degradation and in the production of maps showing different levels of this degradation. It was also verified, that there was no direct relationship between SMI and physical parameters of soil moisture content. Remote sensing showed to be an important tool in the evaluation of soil moisture indices in degraded areas and to delineate the border effect in this desertification nucleus.

Improvement of 1-look SAR image segmentations with mathematical morphology

Synthetic Aperture Radar (SAR) images are an important source of information. This kind of imaging is little affected by adverse atmospheric conditions, such as ram, clouds, fog, etc., since it operates at frequencies other than the visible. Also, since the sensor is active and carries its own source of illumination, it can operate by night. The problem that arises with the use of this technology is a signal- dependent noise, called speckle. This kind of noise is common to all imaging devices that use coherent illumination, such as laser, microwaves, etc. One of the most useful techniques for image analysis is the segmentation. Using statistical modelling, two multiclass segmentation techniques for 1-look and linear detection SAR images are derived: the maximum likelihood and the Iterated Conditional Modes (ICM), both assuming multiplicative Rayleigh models for the data. Although the ICM segmentation yields significatively better results than the maximum likelihood segmentation, the 1-look linear detection case is noisy enough to deserve some improvement. Mathematical Morphology, a non linear approach to signal processing, is then used as a refinement technique in order to extract information.

Comportamento espacial da clorofila-a no reservatório de Itaparica, rio São Francisco

In the analysis of trophic state of the water body is fundamental to know chlorophyll-a concentration. Thus, this work has as main aim to determinate and to assess the behavior of chlorophyll-a in the Itaparica reservoir, Sao Francisco river. This way, we used Landsat-TM imagery, in which it was used bands from 1 to 5 and 7. The algorithm used was written in LEGAL/SPRING 5.2. From the chlorophyll-a result was held slicing the water body in six concentration classes. As observed by histogram, the minimum value of Chl-a was < 1 mg/L and the highest was 249.5 mg/L. The classes that had the biggest area were Classe 01 (0-5 mg/L) with 27.4%, followed by Classe 02 (5-10 mg/L) with 24.6% of the total area of the study area. Through graphical analysis of points located along the reservoir it was possible to verify that chlorophyll concentration augmented from fluvial to lacustrine region and from the contact of streams with reservoir. In the next studies there is a need to validate the values with field data in order to verify the mapping accuracy in this reservoir, taking into account the day and also the transit time of the sensor.

A Gis-Based Decision Support Tool for Renewable Energy Management and Planning in Semi-Arid Rural Environments of Northeast of Brazil —General Description and Methodology

This work describes the development of a system of management and planning elaborated on a GIS platform (Geographic Information System) and is directed to decision makers, be they administrators, planners or consultants in renewable energy. It was conceived to deal with the management and planning of photovoltaic solar systems, biomass and aeolic energy in rural regions of the Northeast of Brazil. The prototype of the GIS tool covers an area of 183,500 km2 (in a second phase it will be extended to 1,561,178 km2) and it is made up of two main blocks: management of installed renewable energy systems and planning of insertion of new renewable energy systems. The system was mainly developed for PV systems as a possible tool of support for the management and planning of the Program of Energetic Development of the States and Municipalities (PRODEEM), a program of insertion of large scale photovoltaic solar energy (thousands of systems), in the rural environment, directed by the Ministry of Mines and Energy of Brazil. The tool for the management of the photovoltaic systems permits the execution of the technical analyses that are involved in the different combinations of the following layers of information: PV systems installed for the purpose of application (energetic, water pumping and others), identification of systems components, identification of component failures, maintenance and training posts, map of monthly mean solar irradiation, infrastructure access, electric transmission lines and distance from a given municipality to the nearest maintenance post. All the maps above can be combined in such a way that information at state or municipal level can be obtained. Additionally, in all the circumstances above, resulting thematic maps as well as issue reports can be printed. The tool for planning the insertion of new photovoltaic systems permits the execution of the technical analyses that are involved in the different combinations of the following layers of additional information: daily solar irradiation on the collector plane for the month of minimum solar irradiation, indices of non-electrified rural areas, indices of human development and localization of subterranean water wells. For each municipality there is the possibility of simplified dimensioning of the following systems: autonomous systems and systems interconnected to the electric network and for pumping water for a given demand and for a given manometric height. Finally the architecture of the GIS system for management and planning of renewable systems is open and permits a credential researcher to update data from diverse data banks and to program operations between thematic maps which were not previously foreseen. Due to limitation of the space only the general description and the methodology of development of the project will be described. The description of the software and its use for management and planning of the insertion of renewables sources of energy will be presented in other works.

The legacy of large dams and their effects on the water-land nexus

Man-made river dams and reservoirs have increasingly been constructed to modify free-flowing rivers to benefit society through hydropower generation, irrigation, and other water supplies, navigation, and flood prevention. However, this ongoing global boom (Zarfl et al. 2015) also triggers harmful outcomes to local, directly affected stakeholder groups, and the environment. Particularly, profound social impacts of involuntary resettlement need alleviatingmeasures and room for remembrance. Restoring vital characteristics of aquatic ecosystems after artificial reservoir establishment, to any possible degree, may contribute to higher welfare and sustainability. Large reservoirs cause both particularly large positive and negative effects on society, the economy, and the environment. In Brazil, reservoirs were initially constructed for the primary purpose of hydroelectricity generation, to prevent flooding, and to provide irrigation capacities in the dryer parts of the country. However, an increasing number of users and usages have increased the pressure on stored and flowing water. Their requirements differ in terms of river discharge, water quality, and reservoir levels, most often reducing the options of the water users downstream. The resolution of conflicts over water allocation and management has been legally supported by the Brazilian Water Act since 1997, which introduced the paradigm of multiple and equally important water uses. Several of these water uses are closely linked to land-use practices, particularly irrigated agriculture. By law, domestic supply for the river basin’s residents sets a priority for water consumption during pronounced water scarcity. In terms of accessibility and water quality, water abstraction for domestic supply is being affected by water levels. Domestic water supply is also directly linked to water pollution through dilution of untreated wastewater in the river or reservoir. The intertwined and often divergent necessities and externalities of water and land use call for a coordinated management and governance approach to mitigate conflicts. Yet many of