Navegando por Assunto "Processamento de imagens - Técnicas digitais"

Agora exibindo 1 - 2 de 2

Acesso aberto (Open Access)
Análise de imagens de sensores remotos orbitais para mapeamento de ambientes costeiros tropicais e de índices de sensibilidade ambiental ao derramamento de óleo no Golfão Maranhense
(Universidade Federal do Pará, 2006-04-18) TEIXEIRA, Sheila Gatinho; SOUZA FILHO, Pedro Walfir Martins e; http://lattes.cnpq.br/3282736820907252
The oil spills in Brazil are more and frequents, causing several impacts on environments and biological communities. Envronmental sensitivy index maps of oil spils are indispensbles components of contingency and emergency answer plans for this tpe of accident. These maps present a system of classification based on geomorphologic characteristics of the áreas, which are defined by the following factors: wave and tidal energy exposure relative degree, shoreline slope and substrate type, and also, the easiness to clean and remove the oil impacted áreas. In this contect, the “Golfão Maranhense” region, located on Northern “Maranhão” State, was chosen in order to map and analyze the environmental sensitivity indexes (ESis) of oil spills on coastal environments, for in this área, we find the second largest port in draught of the world, the Porto f “Itaqui”. Moreover, this region is the route of six hundred oil tankers per year, which are potential agents that cause the oil spill. The methodological approach for creating the maps of environmental sensitivity index included the integrated analysis of coastal environments based on digital image processing from remote optical sensores, in this case, Landsat-4TM, CBERG-2 CCD and SPOT-2 HRV, SAR (Synthetic Aperture Radar) images from RADARSAT-1 Wide 1, SRTM (Shuttle Radar Topography Mission) elevation data, geographic information system and Field surveys related to geomorphology, topography and sedimentology. Using the aforementioned methods, the coastal environments recognized in “Golfão Maranhense” were grouped according to their envronmental sensitivity index: 1 – Solid men-made structures (ESI 1B); 2- Cliffs (ESI 1C); 3 – Fine grained sand beaches and móbile dunes (ESI 3ª); 4- Tidal sandflats (ESI 7); 5- Mixed intertidal Banks, tidal mudflats and ebb-tidal delta (ESI 9ª); 6- Supratidal sandflat (ESI 9C); 7- Saltmarshes (ESI 10ª); 8- Fresh marhes and intermittent lakes (ESI 10B) and 9- Mangrove (ESI 10C). This approach is efficient to recognize and analyze coastal environments and, therefore, it pernitted the sensivity index attribution for the oil spill on yhese environments, in a georefenced data base, which allows making faster e more efficiently decisions in case oil spills come to happen.
Acesso aberto (Open Access)
Estimativa da produção de uma lavoura através de imagens digitais capturadas por veículo aéreo não tripulado (VANT)
(Universidade Federal do Pará, 2018-10-08) SEREJO, Gerson Lima; GOMES, Ana Claudia da Silva; http://lattes.cnpq.br/9898138854277399; SANTOS, Viviane Almeida dos; http://lattes.cnpq.br/1489376127395764
The use of Unmanned Aerial Vehicles (UAV) is becoming an important accessible tool for small to medium sized agribusiness. Its application supports the execution of complex and laborious activities, as well as promotes new studies and challenges for the field to assist the farmer's decision making. The County of Tucuruí, in the State of Pará Brazil, is part of a region that concentrates a great amount of rural properties characterized by being of family agriculture. The objective of this work is to present an exploratory study for applying steps of Digital Image Processing (DIP) and Computational Vision (CV) in images captured by a UAV to obtain the quantification of cassava seedlings and, consequently, harvest of this crop in a farm of the county. The scientific contribution of this study corresponds to the results obtained from the application of 4 vegetation indices: ExG, ExR, (ExG-ExR) and MaxG. The MaxG index presented the best result, counting 91% of the seedlings, in the best case, with an accuracy of 70%. The ExR index was more appropriate for counting the seedlings in initial stages of germination. The index (ExG-ExR) allowed the estimation with unsupervised thresholding, which improves the development of CV systems for this purpose. The ExG index surprised us with the lowest performance for the studied context, counting 58% of the seedlings, in the worst case, with accuracy of 73%. As practical contributions to the farmer, this study made it possible to raise awareness of the importance of forecasting the harvest to better plan the negotiation of production, later plantings and the search for resources to increase the mechanized area of the crop. Further indepth research needs to be conducted to confirm these findings.