Avaliação de técnicas de classificação digital de imagens landsat em diferentes padrões de cobertura da terra em Rondônia

Remotely sensed imagery has been widely used for land use/cover classification thanks to the periodic data acquisition and the widespread use of digital image processing systems offering a wide range of classification algorithms. The aim of this work was to evaluate some of the most commonly used supervised and unsupervised classification algorithms under different landscape patterns found in Rondônia, including (1) areas of mid-size farms, (2) fish-bone settlements and (3) a gradient of forest and Cerrado (Brazilian savannah). Comparison with a reference map based on the kappa statistics resulted in good to superior indicators (best results - K-means: k=0.68; k=0.77; k=0.64 and MaxVer: k=0.71; k=0.89; k=0.70 respectively for three areas mentioned). Results show that choosing a specific algorithm requires to take into account both its capacity to discriminate among various spectral signatures under different landscape patterns as well as a cost/benefit analysis considering the different steps performed by the operator performing a land cover/use map. it is suggested that a more systematic assessment of several options of implementation of a specific project is needed prior to beginning a land use/cover mapping job.

Mapeamento de semeaduras de soja (Glycine max (L.)Merr.) mediante dados MODIS/Terra E TM/Landsat 5: um comparativo

The objective of this work was to compare the soybean crop mapping in the western of Parana State by MODIS/Terra and TM/Landsat 5 images. Firstly, it was generated a soybean crop mask using six TM images covering the crop season, which was used as a reference. The images were submitted to Parallelepiped and Maximum Likelihood digital classification algorithms, followed by visual inspection. Four MODIS images, covering the vegetative peak, were classified using the Parallelepiped method. The quality assessment of MODIS and TM classification was carried out through an Error Matrix, considering 100 sample points between soybean or not soybean, randomly allocated in each of the eight municipalities within the study area. The results showed that both the Overall Classification (OC) and the Kappa Index (KI) have produced values ranging from 0.55 to 0.80, considered good to very good performances, either in TM or MODIS images. When OC and KI, from both sensors were compared, it wasn't found no statistical difference between them. The soybean mapping, using MODIS, has produced 70% of reliance in terms of users. The main conclusion is that the mapping of soybean by MODIS is feasible, with the advantage to have better temporal resolution than Landsat, and to be available on the internet, free of charge.

Características espectrais da soja ao longo do ciclo vegetativo com imagens landsat 5/TM em área agrícola no oeste do Paraná

The objective of this study was to analyze changes in the spectral behavior of the soybean crop through spectral profiles of the vegetation indexes NDVI and GVI, expressed by different physical values such as apparent bi-directional reflectance factor (BRF), surface BRF, and normalized BRF derived from images of the Landsat 5/TM. A soybean area located in Cascavel, Paraná, was monitored by using five images of Landsat 5/TM during the 2004/2005 harvesting season. The images were submitted to radiometric transformation, atmospheric correction and normalization, determining physical values of apparent BRF, surface BRF and normalized BRF. NDVI and GVI images were generated in order to distinguish the soybean biomass spectral response. The treatments showed different results for apparent, surface and normalized BRF. Through the profiles of average NDVI and GVI, it was possible to monitor the entire soybean cycle, characterizing its development. It was also observed that the data from normalized BRF negatively affected the spectral curve of soybean crop, mainly, during the phase of vegetative growth, in the 12-9-2004 image.

Albedo à superfície a partir de imagens Landsat 5 em áreas de cana-de-açúcar e cerrado

Propôs-se, neste trabalho, estimar dados de albedo à superfície terrestre usando-se o sensor Thematic Mapper (TM) do satélite LANDSAT 5 e compará-lo com dados de duas estações agrometeorológicas localizadas em região de Cerrado e a outra em cultivo da cana-de-açúcar. A região de estudo está localizada no município de Santa Rita do Passa Quatro, SP, Brasil. Para a realização do estudo obtiveram-se seis imagens orbitais do satélite Landsat 5 sensores TM, na órbita 220 e ponto 75, nas datas de 22/02, 11/04, 29/05, 01/08, 17/08 e 21/11, todas do ano de 2005, a que correspondem os dias juliano de 53, 101, 149, 213, 229 e 325, respectivamente. As correções geométricas para as imagens foram realizadas e geradas as cartas de albedo. O algoritmo SEBAL estimou satisfatoriamente os valores de albedo de superfícies sobre áreas de cerrado e de cana-de-açúcar, na região de Santa Rita do Passa Quatro, SP, consistentes com observações realizadas do albedo à superfície.

Using FCD Mapper Software and Landsat Images to Assess Forest Canopy Density in Landscapes in Australia and the Philippines

Using Landsat imagery, forest canopy density (FCD) estimated with the FCD Mapper®, was correlated with predominant height (PDH, measured as the average height of the tallest 50 trees per hectare) for 20 field plots measured in native forest at Noosa Heads, south-east Queensland, Australia. A corresponding image was used to calculate FCD in Leyte Island, the Philippines and was validated on the ground for accuracy. The FCD Mapper was produced for the International Tropical Timber Organisation and estimates FCD as an index of canopy density using reflectance characteristics of Landsat Enhanced Thematic (ETM) Mapper images. The FCD Mapper is a ‘semi-expert’ computer program which uses interactive screens to allow the operator to make decisions concerning the classification of land into bare soil, grass and forest. At Noosa, a positive strong nonlinear relationship (r2 = 0.86) was found between FCD and PDH for 15 field plots with variable PDH but complete canopy closure. An additional five field plots were measured in forest with a broken canopy and the software assessed these plots as having a much lower FCD than forest with canopy closure. FCD estimates for forest and agricultural land in the island of Leyte and subsequent field validation showed that at appropriate settings, the FCD Mapper differentiated between tropical rainforest and banana or coconut plantation. These findings suggest that in forests with a closed canopy this remote sensing technique has promise for forest inventory and productivity assessment. The findings also suggest that the software has promise for discriminating between native forest with a complete canopy and forest which has a broken canopy, such as coconut or banana plantation.

Mapping changes in coastline geomorphic features using Landsat TM and ETM+ imagery: examples in southeastern Brazil

Multitemporal Landsat Thematic Mapper (TM) and Enhanced Thematic Mapper Plus (ETM+) imagery was used to assess coastline morphological changes in southeastern Brazil. A spectral linear mixing approach (SLMA) was used to estimate fraction imagery representing amounts of vegetation, clean water (a proxy for shade) and soil. Fraction abundances were related to erosive and depositional features. Shoreline, sandy banks (including emerged and submerged banks) and sand spits were highlighted mainly by clean water and soil fraction imagery. To evaluate changes in the coastline geomorphic features, the fraction imagery generated for each data set was classified in a contextual approach using a segmentation technique and ISOSEG, an unsupervised classification. Evaluation of the classifications was performed visually and by an error matrix relating ground-truth data to classification results. Comparison of the classification results revealed an intense transformation in the coastline, and that erosive and depositional features are extremely dynamic and subject to change in short periods of time.

Wetland Habitat Studies using various Classification Techniques on Multi-Spectral Landsat Imagery: Case study: Tram chim National Park, Dong Thap Vietnam

Dissertation submitted in partial fulfilment of the requirements for the Degree of Master of Science in Geospatial Technologies

Spatial metrics and Landsat data for urban landuse change detection: case of Addis Ababa, Ethiopia.

Dissertation submitted in partial fulfilment of the requirements for the Degree of Master of Science in Geospatial Technologies.

Interpretação tectónica de um conjunto de lineamentos WNW-ESE observados em imagens landsat da margem oeste ibérica

The existence of satellite images ofthe West Iberian Margin allowed comparative study of images as a tool applied to structural geology. Interpretation of LANDSAT images of the Lusitanian Basin domain showed the existence of a not previously described WNW-ESE trending set oflineaments. These lineaments are persistent and only observable on small scale images (e.g. approx. 11200000 and 11500 000) with various radiometric characteristics. They are approximately 20 km long, trend l200±15° and cross cut any other families oflineaments. The fact that these lineaments are perpendicular to the Quaternary thrusts of the Lower Tagus Valley and also because they show no off-set across them, suggests that they resulted from intersection oflarge tensile fractures on the earth's surface. It is proposed in this work that these lineaments formed on a crustal flexure of tens ofkm long, associated with the Quaternary WNW-ESE oriented maximum compressive stress on the West Iberian Margin. The maximum compressive stress rotated anticlockwise from a NW -SE orientation to approximately WNW-ESE, from Late Miocene to Quaternary times (RIBEIRO et aI., 1996). Field inspection of the lineaments revealed zones of norm~1.J. faulting and cataclasis, which are coincident with the lineaments and affect sediments of upper Miocene up to Quaternary age. These deformation structures show localized extension perpendicular to the lineaments, i.e. perpendicular to the maximum compressive direction, after recent stress data along the West Portuguese Margin (CABRAL & RIBEIRO, 1989; RIBEIRO et at., 1996). Also, on a first approach, the geographical distribution of these lineaments correlates well with earthquake epicenters and areas of largest Quaternary Vertical Movements within the inverted Lusitanian Basin (CABRAL, 1995).

Analysis of classification algorithms for crop detection using LANDSAT 8 images

Remote sensing - the acquisition of information about an object or phenomenon without making physical contact with the object - is applied in a multitude of different areas, ranging from agriculture, forestry, cartography, hydrology, geology, meteorology, aerial traffic control, among many others. Regarding agriculture, an example of application of this information is regarding crop detection, to monitor existing crops easily and help in the region’s strategic planning. In any of these areas, there is always an ongoing search for better methods that allow us to obtain better results. For over forty years, the Landsat program has utilized satellites to collect spectral information from Earth’s surface, creating a historical archive unmatched in quality, detail, coverage, and length. The most recent one was launched on February 11, 2013, having a number of improvements regarding its predecessors. This project aims to compare classification methods in Portugal’s Ribatejo region, specifically regarding crop detection. The state of the art algorithms will be used in this region and their performance will be analyzed.

Mapping the Pasture Steppe in Bayankhongor, Mongolia: comparison of classification methods, using Landsat-8 and geophysical data

Grasslands in semi-arid regions, like Mongolian steppes, are facing desertification and degradation processes, due to climate change. Mongolia’s main economic activity consists on an extensive livestock production and, therefore, it is a concerning matter for the decision makers. Remote sensing and Geographic Information Systems provide the tools for advanced ecosystem management and have been widely used for monitoring and management of pasture resources. This study investigates which is the higher thematic detail that is possible to achieve through remote sensing, to map the steppe vegetation, using medium resolution earth observation imagery in three districts (soums) of Mongolia: Dzag, Buutsagaan and Khureemaral. After considering different thematic levels of detail for classifying the steppe vegetation, the existent pasture types within the steppe were chosen to be mapped. In order to investigate which combination of data sets yields the best results and which classification algorithm is more suitable for incorporating these data sets, a comparison between different classification methods were tested for the study area. Sixteen classifications were performed using different combinations of estimators, Landsat-8 (spectral bands and Landsat-8 NDVI-derived) and geophysical data (elevation, mean annual precipitation and mean annual temperature) using two classification algorithms, maximum likelihood and decision tree. Results showed that the best performing model was the one that incorporated Landsat-8 bands with mean annual precipitation and mean annual temperature (Model 13), using the decision tree. For maximum likelihood, the model that incorporated Landsat-8 bands with mean annual precipitation (Model 5) and the one that incorporated Landsat-8 bands with mean annual precipitation and mean annual temperature (Model 13), achieved the higher accuracies for this algorithm. The decision tree models consistently outperformed the maximum likelihood ones.

Avaliação de técnicas de segmentação e classificação automática de imagens landsat-tm no mapeamento do uso do solo na Amazônia1

O mapeamento do uso da terra é fundamental para o entendimento dos processos de mudanças globais, especialmente em regiões como a Amazônia que estão sofrendo grande pressão de desenvolvimento. Tradicionalmente estes mapeamentos têm sido feitos utilizando técnicas de interpretação visual de imagens de satélites, que, embora de resultados satisfatórios, demandam muito tempo e alto custo. Neste trabalho é proposta uma técnica de segmentação da imagens com base em um algoritmo de crescimento de regiões, seguida de uma classificação não-supervisionada por regiões. Desta forma, a classificação temática se refere a um conjunto de elementos (pixels da imagem), beneficiando-se portanto da informação contextual e minimizando as limitações das técnicas de processamento digital baseadas em análise pontual (pixel-a-pixel). Esta técnica foi avaliada numa área típica da Amazônia, situada ao norte de Manaus, AM, utilizando imagens do sensor "Thematic Mapper" - TM do satélite Landsat, tanto na sua forma original quanto decomposta em elementos puros como vegetação verde, vegetação seca (madeira), sombra e solo, aqui denominada imagem misturas. Os resultados foram validados por um mapa de referência gerado a partir de técnicas consagradas de interpretação visual, com verificação de campo, e indicaram que a classificação automática é viável para o mapeamento de uso da terra na Amazônia. Testes estatísticos indicaram que houve concordância significativa entre as classificações automáticas digitais e o mapa de referência (em tomo de 95% de confiança).

Mapeamento da antiga cobertura vegetal de várzea do Baixo Amazonas a partir de imagens históricas (1975-1981) do Sensor MSS-Landsat

Este estudo apresenta um mapa da cobertura vegetal da planície de inundação do Rio Amazonas entre as cidades de Parintins (AM) e Almeirim (PA), com base em imagens Landsat-MSS adquiridas entre 1975 e 1981. O processamento digital dessas imagens envolveu a transformação para imagens-fração de vegetação, solo e água escura (sombra), seguido da aplicação de técnicas de segmentação e classificação por região. O mapa resultante da classificação foi organizado em quatro classes de cobertura do solo: floresta de várzea, vegetação não-florestal de várzea, solo exposto e água aberta. A precisão do mapa foi estimada a partir de dois tipos de informações coletadas em campo: 1) pontos de descrição: para validação das classes de cobertura não sujeitas a grandes alterações, como é o caso dos corpos d'água permanentes, e identificação de indicadores dos tipos de cobertura original presentes na paisagem na ocasião da obtenção das imagens (72 pontos); 2) entrevistas com moradores antigos para a recuperação da memória sobre a cobertura vegetal existente há 30 anos (44 questionários). Ao todo foram coletadas informações em 116 pontos distribuídos ao longo da área de estudo. Esses pontos foram utilizados para calcular o Índice Kappa de concordância entre os dados de campo e o mapa resultante da classificação automática, cujo valor (0,78) indica a boa qualidade do mapa de cobertura vegetal da várzea. Os resultados mostram que a região possuía uma cobertura florestal de várzea de aproximadamente 8.650 km2 no período de aquisição das imagens.