National Institute for Space Research
Biblioteca Digital da Memória Científica do INPENot a member yet
6175 research outputs found
Sort by
Spatial Variability Analysis of CBERS CCD Images in Forest Regions
No full textThis paper analyses the spatial variability of remote sensing images obtained by the Charge Coupled Device (CCD) sensor presented in the China Brazil Earth Remote Satellite (CBERS) at spatial regions of forests. Semivariograms are mostly used to model the spatial variability of environmental data as elevation, temperature, health risks, geology, hydrology and mining information, etc... In this work the semivariograms were used to model the variability of the spectral information represented in remote sensing images. A limited number of sample points of images were considered, instead of full images, due to the size constraints of the input data to calculate the semivariances. A very large amount of input points requires a very high processing time to evaluate the semivariograms. Thus, this work aims to explore the possibility of representing the spatial variability of an entire image using a limited amount of samples draw from it. Semivariogram analysis were done with random sample sets of different sizes. Two spatial regions with different patterns were considered, one with typical forested area and other with partially deforested area. It was found that is feasible to get representative semivariograms of the whole image from small sample sets. In addition it was obtained different semivariograms for the two regions showing that stationary hypothesis cannot be assumed for forest images with different patterns.Pages: 2180-218
Fotografias oblíquas digitais de pequeno formato na elaboração de plantas de quadra
No full textThe use of maps produced from oblique photographs have been taken in recent years, with significant importance, both for cartographic visualization systems for Internet, such as Geographic Information Systems and virtual reality. Satellite systems for high resolution photogrammetric systems and special arrangements with multiple cameras are the main sources of acquiring oblique images, but imaging systems low cost has also been used. This paper aims to present the development of a mosaic of photographs oblique of Itaquitinga-PE. The work was performed as part of a project whose goal was to test alternatives for providing small cities of cartographic databases, which serve to support the management of the city and the implementation of Multipurpose Cadastre. In this article, we discuss the procedures for planning, acquisition and processing of images. The results of blocks and rectified map produced are presented and discussed.Pages: 1974-198
Algoritmo automático de detecção de áreas queimadas em imagens MODIS - aplicação na região do Jalapão, TO
No full textThis paper presents a procedure to detect burned areas based on 1km Moderate Resolution Imaging Spectroradiometer (MODIS) imagery onboard the TERRA and AQUA satellites. The method uses channels 2 (near-infrared NIR, centered at 0,856 µm) and 20 (middle-infrared MIR, centered at 3.96 µm), together with information of active fires detected by INPEs monitoring system. Monthly composites of the V,W burned index with fixed and relative thresholds derived from the pixels in the neighborhood of burned areas are applied. An overview is provided for the results obtained in 2005 for a Cerrado (savanna) region over Parque Nacional do Jalapão, in the east of Tocantins, regarding the spatial and temporal distribution of burned areas. The accuracy assessment was performed using burned area maps derived from medium-resolution (30 m) Landsat imagery and a comparison between our algorithm and the NASA/MODIS burned area product MCD45 was also evaluated. The INPE-MODIS quantification of the burned area extent is in agreement within 10% with that of scars derived from Landsat images; for instance the total amount of burned area in the region of study detected by our algorithm was about 2208 km2, while the scars detected by Landsat images totalized 2082 km2. The NASA MODIS mCD45 product was off by more than 40%. As expected, mapped burned areas in general showed a good agreement in terms of spatial distribution with actives fires.Pages: 6901-690
Caracterização geomorfométrica da folha de Ribeirão Preto - SP para predição de mapas digitais de solo
No full textThe increasing number of features that have been developed for tools like geographic information systems (GIS) and digital elevation models (DEM's) are suggesting new ways to deal with soil science. There currently is a global clamour for soil data and information for environmental monitoring and modelling. The main issue is that large countries like Brazil do not have a reasonable and detailed amount of soil data and information. The reason for this lack of soil spatial data infrastructure worldwide is simply that conventional soil survey methods are slow and expensive (McBratney et al., 2003). In this paper, we group four relief variables (slope, profile curvature, planar curvature and watershed distance) with soil data, from the Ribeirao Preto area. The morphometric parameters were extracted from SRTM data, refined by the The Brazilian Agricultural Research Corporation's (EMBRAPA). We have also updated the soils nomenclatures to the new Brazilian System of Soil Classification (SiBCS)once the soil map used was produced in 1983. Some generalizations were done with groups of soils that presented strong similarities with few variations. The generated data is a valuable tool when producing soil maps using prediction methods and will be used in a posterior work for the prediction of soil maps in areas with similar characteristics.Pages: 4731-473
Veículos aéreos não tripulados: panorama atual e perspectivas para o monitoramento de atividades ilícitas na Amazônia
No full textIn a country of continental dimensions like Brazil, the identification and monitoring of various kinds of illicit activities constitute a growing challenge for governmental institutions in their different spheres of activity. Some of these challenges are related, for example, to combating drug trafficking and environmental crimes in the Amazon borders, as well as the trafficking of weapons in the triple border regions. In response to these challenges, unmanned aerial vehicles (UAV) come to Brazil as the newest trend in remote sensing. Beside the reduction in operating costs compared to manned aircraft, the use of UAV's brings the possibility of monitoring illegal activities in real time or in adverse conditions, what makes of this instrument an excellent alternative to the area of security and defense and opens new perspectives for the monitoring of environmental crimes. Considering this scenario, this paper aims to present an overview on the unmanned aerial vehicles in Brazil and more specifically, highlight the prospects of using unmanned aircraft to combat illegal activities in the Amazon. The improvements obtained with this data can be further enhanced by means of automated tools of image processing implemented in strategic Geospatial Systems currently available. The range of possible applications of such data represents a landmark in the areas of public security, environment and so many others that require the usage of high resolution data in real time for civil and military emergency situations.Pages: 9324-933
Avaliação de Áreas com Restrição Legal de Uso do Solo em Arranjos Espaciais Distintos de Ocupação das Terras no Nordeste do Estado do Pará
No full textO artigo analisa o uso e ocupação do solo de dois diferentes tipos de propriedades (comuns e privadas), localizados no Nordeste do Estado do Pará, Brasil, que constitui a zona mais antiga de ocupação na região amazônica. Devido a essas peculiaridades, tem, conseqüentemente, mais perturbado áreas, que demandam estudos para mitigar os impactos ambientais. A fim de realizar a análise espacial, foi utilizada ferramenta GIS, ArcGIS e softwares SPRING. Além disso, a imagem de satélite SPOT 5 foi utilizado para analisar o uso e ocupação do solo, assim como a confecção de mapas de restrição legal de uso da terra. Para esta análise, a nova resolução sobre Áreas de Preservação Permanente (APP, área jurídica brasileira que inclui proteção da integridade ambiental das fronteiras fluviais) e Reserva Legal (ARL), ambos presentes no novo Código Florestal (CF). Verificou-se que os resultados de Áreas de Preservação Permanente e Reservas Legais ficaram aquém do previsto e, portanto, que, apesar de ser de diferentes tipos de propriedade, nenhum deles atingiu o montante estabelecido pelas áreas CF para anotação de restrição de uso. No entanto, são necessárias medidas específicas para cada tipo de imóvel, respeitando suas particularidades, com ações voltadas para a preservação e, portanto, respeitar as regras do CF em vigor.Pages: 7305-731
Geração de imagens sintéticas para utilização em estudos de detecção de mudanças
No full textRemotely-sensed images of the same area acquired on different dates tend to be affected by radiometric differences and registration problems. These influences are considered as noise in the process and it is a critical problem when the difference image is used for change detection purposes. Most change detection methods are based on image differencing and this kind of image presents more noisy than the original ones because the variance is statistically computed by the sum of each image individual variance (minus two times the covariance. This fact causes in the difference image the appearance of at least two populations: i) the noise-like and ii) the real changes. It is not easy to separate the noise from the signal because it is impossible to determine where the noise ends and the signal begins. The knowledge about the noise is essential for the generation of synthetic images and there is a lack of studies about the acceptable level to be added on images to simulate real situations. This paper reports on the use of synthetic images for change detection accuracy assessment and attempts to suggest values of SNR (signal-to-noise ratio). The experimental results obtained from experiments performed on TM-Landsat-showed that SNR values lower than 8 as more realistic for the bands in which there is maximum spectral response.Pages: 1696-170
Uma abordagem baseada em objeto na análise de imagens com a utilização de segmentações independentes
No full textGeographic Object-Based Image Analysis (GEOBIA) makes it possible to exploit a number of new features in the remote sensing image classification process. Such possibility is brought by the introduction of a segmentation step in the analysis process. High resolution images from remote sensors have multiscale nature, and include many details and information about the objects that compose them, it is important to segmentation these images effectively with all types of information and characteristics of these objects. The new features refer to aggregated spectral pixel values, textural, morphological and topological features computed for the different image segments. The usual segmentation approach in GEOBIA works relies on a hierarchy of segmentations, each level related to a number of object classes that have similar sizes, i.e., are detectable in a similar scale. We, therefore, propose an approach founded on the assumption that if segmentations are not specialized for each object class, then many of the new segment features cannot be properly exploited in the classification process. The usual practice do not consider independent segmentations for each class of interest in the problem of interpretation, grouping objects of same scale in a single segmentation, and this may miss important details about the objects it contains. The proposed approach relies on a specific rule to solve eventual spatial conflicts among different segmentations. Preliminary experimental results show that the proposed approach performed better that the usual one.Pages: 8230-823
Uso do geoprocessamento no cálculo do tempo de concentração da água da chuva em ambientes urbanos
No full textThis research sought to develop a methodology to assist in studies of urban drainage, using geoprocessing and remote sensing techniques, considering geographical and geomorphological elements, especially in basins with limited hydrological data of rainfall and runoff. In order to develop this work, it was used data provided by the City of Goiânia which were stored in a geographic database. The use of this information within a Geographic Information System (GIS) environment with the help of computational routines in Visual Basic Language (VBA) integrated with a GIS allowed the characterization of the area of study, allowing the estimation of the speed and time per pixel, the creation of time-area histograms and the estimation time of concentration for the scenario of land use and occupation. Analyses of results showed the high rate of soil waterproofing and urbanization, the velocity and time of concentration. Data generated using the proposed methodology indicated time of concentration in the study area 4.7 hours. The methodology proposed in this paper is efficient for estimate the time of concentration in basins with limited hydrologic data. The results can be used in hydrologic modeling with the intention of assisting in the prevention of floods.Pages: 4390-439
Estimativa da velocidade de veículos aplicando o detector de Harris
No full textThe number of automobiles has increased in recent years in Brazil, because of the growing economy, making it easier for the population to buy more than one vehicle. The increase of the fleet caused some problems such as air pollution and saturation of urban roads. Urban planning did not preview, therefore alternatives to control the number of vehicles are being proposed. This paper describes an application of computer vision to determine the speed of vehicles based on a series of images taken by a low cost camera. The method is based on the identification of homologous points in a series of images, which allows to estimate the movement and the speed of the objects.Pages: 2100-210