National Institute for Space Research
Biblioteca Digital da Memória Científica do INPENot a member yet
6175 research outputs found
Sort by
Análise do processamento e ajustamento de dados GPS para linhas de base curtas e longas
No full textAmong the methods used for GPS surveys, the fast static relative positioning detaches because of its high productivity. However, according to INCRA Technical Standard, its necessary that the maximum distance between the reference station and the station whose coordinates need to be determined be of 20km. In order to verify the errors from the baseline length, 14 stations of the local GNSS lan, located in Federal University of Uberlândia, Santa Mônica campus, were surveyed with GPS receivers of single frequency (L1). Then, it was held the processing and adjustment of the datas in the software Topcon Tools, version 8.2. This step was carried out three times, just changing the reference station. Initially, it was decided to choose stations that formed short baselines with the rovers: MGUB and UBER. After, to analyze the degradation of the results due to the increase of the vectors between the stations, was chosen MGMC as the base reference. All these stations belong to the brazilian network of continuos monitoring and its datas are provided by IBGE. From this experiment, it was possible to verify that the precision of the results vary considerably by varying the lengths of the baseline. So, its important that the station took as reference be near to the points whose positions are desired, thereby increasing the accuracy of the results.Pages: 4684-469
Geomorfometria e geofísica aplicadas à discriminação de unidades sedimentares na Bacia Paraíba, nordeste do Brasil
No full textIntegrating remote sensing and geophysical techniques enables discrimination of geologic units in the Paraíba Basin. Therefore, this work aims to recognize the possible relationships between geomorphometric and geophysical gamma-spectrometric data related to the sedimentary units of Barreiras Formation (Miocene) from the Post-Barreiras Sediments (late Quaternary) in the central part of Paraíba state, Brazil. The dataset used in this work includes: a) dissection geomorphometric index (IDissec) extracted from digital elevation model (DEM) - Shuttle Radar Topography Mission (SRTM); b) aerogamma-spectrometric product of radioisotope Thorium (Th) concentration; c) absolute dating of rocks, derived from optically stimulated luminescence technique (OSL); and d) geological map of the central portion of the Paraíba Basin. Sample plots were selected from the database, corresponding to areas previously recognized during field work. The results indicate an association between areas with low concentrations of Th and low dissection index values. Thus, we conclude that the integration of geophysical data (Th) and geomorphometric index (IDissec) has great feasibility for mapping sedimentary units in the Paraíba Basin, being a potential tool to be extended to the entire northeastern Brazilian coast, due to the large equivalence of the sedimentary units.Pages: 3534-354
Avaliação de áreas para instalação de aterro sanitário no município de Barreiras, Bahia
No full textThe increasing amount of waste generated by urban areas is becoming worrying, especially as regards the conservation of the environment and public health of the population, which requires some care needed for the construction of sanitary landfills. For installation of this enterprise are needed technical studies, environmental and socioeconomic used to evaluate a number of criteria for the selection of suitable areas deploying the sanitary landfill in order to minimize impacts. This study aims to demonstrate some weaknesses in the sanitary landfill Barriers city, Bahia, Brazil. The vicinity to the highway (BR-242) of general circulation was the main irregularity found in the current sanitary landfill, which may facilitate disease transmission through locomotors vehicles that travel near the area. We performed a proposed new sanitary landfill, based on the ratio of distances between some important variables in this process. The identification of this new proposal, it was observed that there is no restriction regarding the proximity to roads, slope flat, away from rivers, the urban and protected areas. The Geographic Information System demonstrated to be a practical, economical and reliable, which can contribute to the selection of the areas appropriate the installation to sanitary landfill because it integrates a number of physical, biotic and social analysis in computational media, which minimize the environmental impacts caused by the project and avoid unnecessary costs.Pages: 842-84
Análise das componentes do balanço de energia à superfície para a cidade de Manaus-AM a partir de imagens do sensor Landsat 5-TM
No full textThis research aimed to estimate and analyze the conduct of the surface radiation balance (Rn) in the city of Manaus-AM from information provided by Landsat 5-TM. The use of remote sensing in this context aims to fill a gap in Brazil, where the available data of solar radiation is very limited and mostly refers to climatological averages. For this, we used the procedures of the method Algorithm for Land Surface Balance - SEBAL. Was obtained values of the components of Rn-based reflective radiance from the channels (1, 2, 3, 4, 5, and 7) and the thermal channel (6) 5-Landsat TM. For comparative purposes, the main types of land cover were analyzed in the study area: urban area, water bodies, vegetation and bare soil. The emissivity values did not differ significantly between the targets, the values of albedo and surface temperature were directly proportional to the targets while for values Rn and long-wave emission was observed an inverse relationship in which the surfaces with higher emissivity values showed lower Rn. So clear was the contrast between the natural surfaces (vegetation and water bodies) that showed the highest values of Rn, while the urban area and areas of exposed soil showed lower values due to low capacity and high thermal conductivity heat material associated therewith.Pages: 7974-798
Avaliação da Integridade das Áreas de Preservação Permanente (APP) entre os anos de 1984 e 2008 no Cerrado do Oeste baiano
No full textThis paper aims to evaluate the integrity of the Permanent Preservation Areas in the São Desidério municipality, located in the region of Western Bahia, in the years 1984 and 2008. Images from sensors PRISM/ALOS and Landsat-TM were used. The images were co-registered and classified by visual interpretation. Permanent Preservation Areas were delimited and the land-use and land-cover were mapped inside these areas. The classes were quantified and analyzed over time. The Permanent Preservation Areas occupied an area of 96.779 hectares. The antropic area class inside the Permanent Preservation Areas increased significantly from 4,27% (4.131 ha) in 1984 to 8,46% (8.204 ha) in 2008, developing at a rate of 170 hectares per year.Pages: 3436-344
Utilização de um Modelo Estatístico Robusto para a Avaliação da Acurácia de Modelos Digitais do Terreno Gerados Através de Dados LIDAR
No full textDigital Terrain Models (DTMs) are today generated from LIDAR (Light Detection and Ranging) remote sensing data. The laser scanner technology is very efficient and accurate. However, blunders may occur. Such DTMs are very dense and relatively accurate in open terrain. Built-up and wooded areas, however, need automated filtering and classification to generate terrain (bare earth) data when Digital Terrain Models (DTM) have to be produced. Automated processing of the raw data is not always successful. Many outliers and systematic errors may therefore be present in the data sets. The amount of data is huge, but the accuracy assessment has to be made with few check points only as it is very labour intensive to obtain them. In this article are presented the results of the experiment finished. This experiment assessment the robust statistical methods proposal by Höhle and Höhle (2009). The use of this robust statistical method is illustrated by application at two situations. It is concluded that measures such as median, normalized median absolute deviation, and sample quantiles should be used in the accuracy assessment of such DTMs. The studied methodology adapts to the specialities of LIDAR data, where blunders and non-normal distribution are often present, especially in non-open terrain.Pages: 6143-615
Identificação de buritizais através do sensor orbital ALOS PALSAR e estimativa de estoque do recurso
No full textThis study used the ALOS PALSAR sensor (bands L-HH and L-HV) to estimate the area of Mauritia palm stands within the savanna environment of the Araça Indigenous Area, in Roraima. A fuzzy classifier was trained with four land-cover classes observed in a high spatial resolution QuickBird image that covered parts of the PALSAR scene outside the Indigenous Area. For each pixel in the entire PALSAR scene, the classifier provided a probability for each cover type, expressed as four grey-scale images. Based on independent validation sites in the QuickBird image, all pixels with >80% probability for the class Mauritia were taken to be stands of pure Mauritia palm, giving a total area of 144 hectares within the Indigenous Area. Based on previous studies of Mauritia fruit production, we estimated 189 trees ha-1 and 7,4 ton ha-1 y-1, giving a total potential production from pure Mauritia stands in the Araçá indigenous lands of 1066 ton y-1. Utilizing merchant study analyses in the states of Roraima and Pará, we estimated that revenue of wine (juice) and Mauritia oil production for the entire Araçá indigenous lands could be 266.400 liters of wine per year, which could be worth an estimated R$302.400. This same production could produce 47.520 liters of oil. This income potential could help preserve the Mauritia palm swamps in the indigenous communities of the savanna region.Pages: 8336-834
Um método de estimativa de irradiação solar direta normal a partir de imagens de satélite geoestacionário: resultados preliminares
No full textMapping and monitoring of the direct normal irradiance (DNI) is most relevant for proper installation of heliothermic power plants using concentrating solar systems. This work presents a satellite-based method for estimating DNI and integrated direct normal solar irradiation (QN) which uses a minimal set of local weather information. Basic parameters for the computation of QN by satellite are: i) direct normal irradiance under clear sky conditions DNIcc (inferred by a physical parameterization), and ii) cloud cover (estimated from images of GOES Imager VIS channel). Satellite estimations for DNI and QN during July 2007 were compared with measurements of solarimetric stations of SONDA network in Northeast Brazil (Petrolina, Natal and Sao Luís). Preliminary results suggest that accuracy of the proposed method is similar to that of models reported in the literature. DNI daily cycles in conditions of clear sky and partially cloudy were represented satisfactorily, with daily average errors ranging from -11,9 to 5,7 W m-2. Assessed QN shows close linear relationship with measured values, with R2 values as high as 0,97. MBE (Mean Bias Error) and RMSE (Root Mean Square Error) were 8,9 and 15,9%, respectively. Proposed algorithm is a promising tool for providing information of QN (with spatial resolution of 4x4 km), in particular over regions with unavailable in situ measurements.Pages: 6471-647
Análise de filtragem do ruído Speckle em imagens do Radar de Abertura Sintética do SIPAM
No full textThe technology of synthetic aperture radar (SAR) has become widely used in Remote Sensing (RS) in Brazil, especially after the acquisition and use of RS aircraft for the Brazilian Air Force. One of the difficulties of analysis of SAR images is the presence of speckle noise, whose grainy effect complicates the analysis and interpretation of these images, both visual and automatic. The alternative is to try to reduce it by using adaptive filters. Therefore, this study aims to develop a process indicative of quality filters within the parameters of Mean Square Error (MSE), signal to noise ratio (SNR) and evaluation of preservation of edges (Beta). The results suggest that the filter has Lee most suitable parameters for removal of speckle in X band and HH sensor SAR R-99.Pages: 8390-839
Comparação entre os Classificadores Máxima Verossimilhança, SVM e Rede Neural MLP para Uso e Cobertura da Terra em Parcela da FLONA Tapajós e Arredores
No full textThe available tools to classify an image has grown in the last years, with the development of more efficient algorithms and computer technology. This paper aims to evaluate the performance of two supervised classifiers using different parameters, Support Vector Machine and MLP based Neural Net, when compared with Maximum likelihood, on an image from Landsat5 satellites sensor Tematic Mapper. The study area is located in a parcel and surroundings of Floresta Nacional do Tapajós, in which there is forest in different stages, new and old regeneration and agriculture. The respective Kappa coefficient was validated using an hypothesis test with 5% of significance. It was defined six classes of land use and land cover associated with Primary Forest and Primary Forest in Exploration, Degraded Forest, Old Regeneration and Intermediate Regeneration, Initial Regeneration, Prepared Soil for Agriculture and Soybean with 100 days from sowing, Pasture and Soybean with 40 days from sowing. Although SVM has showed good results, it was statistically similar to maximum likelihood. Neural Network has showed statistically inferior or equal results, bus demanded more time, process capacity and is more difficult, due the necessity to choose more parameters. It was concluded that future investigations are needed to achieve the optimum classification using the chosen algorithms.Pages: 2377-238