Detecting changes in land cover through time using remotely sensed imagery is a powerful application that has seen increased use as imagery has become more widely available and inexpensive. Before a time series of remotely sensed imagery can be used for change detection, images must first be standardized for effects outside of real surface change. This thesis established a validation protocol to evaluate the effectiveness of an automated technique for normalizing temporally separate but spatially coincident imagery. Using the concept of pseudo-invariant features between master-slave image pairs, spatially coincident dark and bright points are identified from images and a regression equation is calculated to normalize slave images to a master. I used two sets of imagery to test the performance of the standardization process, a spatially coincident, but temporally variable time series, and spatially and temporally variable images. I tested the underlying statistical assumptions of this approach, and performed simple image subtraction to validate the reduction of master-slave differences using invariant locations. In addition I tested the possibility of reducing between-sensor differences by applying simple linear regression to comparable bands of MSS and TM sensors. Image subtraction showed decreases in master-slave differences as a result of the standardization process, and the process behaved appropriately when there should be no difference between master and slave images (adjacent, but temporally identical imagery). I also found that comparable bands between MSS and TM sensors are similar enough that linear regression may not significantly reduce between-sensor differences.

Orientador: Rubens Augusto Camargo Lamparelli / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia Agricola / Made available in DSpace on 2018-08-13T09:46:25Z (GMT). No. of bitstreams: 1 Nery_LuisAntonio_M.pdf: 111375957 bytes, checksum: 05f2b3dc3fe0dece107d3e34a24a5a3e (MD5) Previous issue date: 2009 / Resumo: A importância econômica da produção brasileira de café no mercado mundial é notória e contribui com uma grande parcela na balança comercial de exportação do país. Minas Gerais se destaca como o centro da atual produção cafeeira no Brasil e tem na região sul do estado a grande concentração da lavoura de café (Coffea arabica), onde o seu cultivo é realizado em pequenas propriedades bastante dispersas pela região montanhosa. A necessidade de adequação da agricultura cafeeira por meio do planejamento, controle de custos e melhoria da produtividade, tem acelerado a procura por técnicas e ferramentas para a previsão da produção agrícola passando, necessariamente, pela localização e quantificação das áreas cultivadas. Neste contexto, o objetivo desta pesquisa foi avaliar o fornecimento de informações dos dados do sensor TM/Landsat 5 utilizando as técnicas de Análise por Principais Componentes (APC) e separação de classes de iluminação sobre as áreas de lavoura de café em região montanhosa. A área de estudo escolhida foi o município de Guaxupé/MG por conter uma forte lavoura cafeeira mantida sob um organizado sistema cooperativo. Foram utilizadas imagens dos satélites Landsat 7, Landsat 5 e do sensor MODIS para a aplicação das técnicas de processamento digital, para correção atmosférica e normalização radiométrica, visando a análise do cafeeiro nas datas de 15/08/2001, 05/12/2001 e 12/04/2002, que caracterizam os estágios fenológicos do cafeeiro como períodos de colheita (repouso e senescência dos ramos), florada e início do crescimento da gemas florais, respectivamente. Também foi utilizada uma máscara da área cafeeira obtida por método de interpretação visual extraído de imagens Ikonos. A utilização de um modelo digital de elevação gerado por imagens do sensor ASTER/TERRA possibilitou a aplicação da técnica de determinação do fator de iluminação, que consistiu na criação de classes de iluminação que contribuíram na identificação de áreas de lavoura e áreas de mata sombreadas pelo relevo. Dados de campo foram levantados para auxiliar na identificação da lavoura cafeeira, separadas pelas classes amostrais de café adensado e café adulto em função dos espaçamentos de ruas e linhas adotados no plantio. A Análise por Principais Componentes (APC) foi aplicada com o objetivo de reduzir a redundância dos dados obtidos das imagens orbitais de maneira a permitir a seleção de amostras de treinamento para a utilização em classificação supervisionada. Utilizando o método da Distância de Mahalanobis como classificador, as imagens nas datas selecionadas para a pesquisa, mostraram dados importantes quando comparados os resultados das classificações com a máscara da área de café extraída das imagens Ikonos do município. Os resultados dessas classificações foram validados por meio da determinação da Exatidão Global e coeficiente Kappa que mostraram os valores de EG= 0,78 e K= 0,40 para a imagem de 15/08/2001, EG= 0,81 e K= 0.29 para a imagem de 05/12/2001 e, EG= 0,76 e K= 0,24 para a data de 12/04/2002, confirmando que o período seco (maio até outubro) é favorável para a classificação do cafeeiro que, neste período está sob processo de colheita onde, a queda de folhas e remoção de frutos ocorre, diferenciando das outras coberturas do solo como as matas. Os maiores valores atingidos na validação dos dados ocorreram na classificação da imagem gerada pela composição dos resultados das três datas, atingindo valores de EG = 0,81 e Kappa = 0,56. O valor da área, quantificada como sendo da cultura do café, encontrado pelo método de soma dos resultados das classificações em cada data, produziu um valor 73,06 % do total de área quantificada na máscara de café, utilizada como referência. Neste sentido a metodologia se mostrou bem adequada na quantificação de áreas de café em relevo montanhoso. / Abstract: The economic importance of Brazilian coffee growing in the world market is notorious and makes up significant portion of the country's foreign trade exports. Minas Gerais stands out as the core of Brazilian coffee growing, with most of the planting areas (Coffea arabica) concentrated in the south, where it is grown in small plots widely spread throughout the hills. The need to adequate coffee agriculture by planning, cost control and productivity improvement has increased the search for techniques and tools for the prediction of agricultural production, necessarily involving the location and quantification of cultivated areas. In this context, the goal of this research has been to evaluate data from the TM/Landsat-5 remote sensor, providing information about coffee growing areas in hilly regions. The city of Guaxupé/MG/Brazil was chosen for this study due to its strong coffee growing, kept under an organized cooperate system. Images from the Landsat-7 and Landsat-5 satellites and from the MODIS sensor have been employed for the purpose of using digital processing tools for atmospheric correction and radiometric normalization, in order to analyze coffee crops in three dates: 08/15/2001, 12/05/2001 and 04/12/2002, characterizing phenological stages such as harvesting periods (rest and senescence of boughs), flowering and beginning of flower bud growing, respectively. The use of a digital elevation model generated from ASTER/TERRA sensor enabled the use of a lighting factor determination technique, consisting in the creation of lighting classes that contributed in the identification of crop areas and terrain-shadowed vegetated areas. Field data were gathered to help identifying coffee plantation separated by sample classes of dense coffee and adult coffee as a function of the spacing of the field foods and lines used in planting. PCA (Principal Component Analysis) was applied in order to reduce the redundancy of the data obtained from orbital imaging in a way that allows the selection of training samples for use in supervised classification. Using the Mahalanobis Distance as a classifier, the images in the selected dates showed highly positive result when the classification was compared to the coffee area mask extracted from Ikonos images. The results of these classifications were validated through the determination of Global Accuracy and Kappa Index, which showed values of GA= 0.78 and K= 0.40 for the 08/15/2001 image, GA= 0.81 and K= 0.29 for the 12/05/2001 image, and GA= 0.76 and K= 0.24 for 04/12/2002, confirming that the dry season (May through October) is favorable for the classification of coffee, which is under the harvesting process in this period, during which the falling of leaves and remotion of fruit separates it from other ground cover such as vegetation. The spectral data obtained from satellite imaging through digital processing have proven themselves adequate for the location of coffee-growing areas in hilly regions when aided by a digital elevation model. The value of the area as being coffee crop, calculated by sum of the areas found from each date classification, produced 73,06% of the agreement with coffee mask considered as a reference data. Due this the methodology showed very suitable to quantify coffee areas in hilly region. / Mestrado / Planejamento e Desenvolvimento Rural Sustentável / Mestre em Engenharia Agrícola

There has been rapid change in the landcover/landuse in the Mekong delta, Viet Nam. The landcover/landuse has changed very fast due to intense population pressure, agriculture/aquaculture farming and timber collection in the coastal areas of the delta. The changing landuse pattern in the coastal areas of the delta is threatened to be flooded by sea level rise; sea level is expected to rise 33 cm until 2050; 45 cm until 2070 and 1 m until 2100. The coastline along the eastern Mekong delta has never been static, but the loss of mangrove forests along the coast has intensified coastline change. The objective of the present study is to map the changes in landcover/landuse along the eastern coast of the Mekong delta; and to detect the changes in position of the eastern coastline over the time period from 1989 to 2002.To detect changes in landuse, two satellite images of the same season, acquired by the TM sensor of Landsat 5 and the ETM+ sensor of Landsat 7 were used. The TM image was acquired on January 16, 1989 and ETM+ image was acquired on February 13, 2002. The landcover/landuse classes selected for the study are water, forest, open vegetation, soil and shrimp farms. Image differencing and post classification comparison are used to detect the changes between two time periods. Image to image correction technique is used to align satellite images. Maximum likelihood supervised classification technique is used to classify images. The result of the classification consists of five classes for 1989 and 2002, respectively. Overall accuracies of 87.5% and 86.8%, with kappa values of 0.85 and 0.84 are obtained for landuse 1989 and landuse 2002, respectively. The overall accuracy for the change map is 82% with kappa value 0.80. Post classification comparison is carried out in this study based on the supervised classification results. According to the results obtained from the post classification comparison, a significant decrease of 48% in forest and a significant increase of 74% in open vegetation and 21% in shrimp farms area observed over the entire study area. The coastline obtained by the combination of histogram thresholding and band ratio showed an overall advancement towards the South China Sea. The results showed that new land patches emerged along the eastern coast. The amount of new land patches appeared along the coast of the Mekong delta is approximately 2% of the entire study area.