Estimativa de área de soja por classificação de imagens normalizada pela matriz de erros

Antunes, J. F. G.; Mercante, Erivelto; Esquerdo, J. C. D. M.; Lamparelli, Rubens Augusto de Camargo; Rocha, Jansle Vieira

doi:10.1590/s0100-204x2012000900014

Cited by 16 publications

(8 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…However, the official statistics are based on research obtained from farmers and cooperatives and agricultural finance data. These values are usually generalized, resulting in relative and imprecise data (Pino, 2001;Antunes et al, 2012;Johann et al, 2012). Antunes, Mercante, Esquerdo, Lamparelli, and Rocha (2012) verified that estimates given by official bodies such as SEAB/DERAL are obtained subjectively using planting intention surveys performed by sampling.…”

Section: Resultsmentioning

confidence: 99%

“…These values are usually generalized, resulting in relative and imprecise data (Pino, 2001;Antunes et al, 2012;Johann et al, 2012). Antunes, Mercante, Esquerdo, Lamparelli, and Rocha (2012) verified that estimates given by official bodies such as SEAB/DERAL are obtained subjectively using planting intention surveys performed by sampling. According to Pino (2001), the IBGE uses subjective municipal information based on opinions of technical and economic agents.…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Mapping of Winter Crops and Second-Crop Corn in the Paraná State-Brazil, Using Multitemporal Images From MODIS Sensor

Santos¹,

Mercante²,

Johann³

et al. 2019

JAS

Self Cite

View full text Add to dashboard Cite

The use of effective technologies for the monitoring of agricultural crops should seek methodologies that provide information regarding crop development, preferably before harvesting. The study of the monitoring and/or estimation of areas using vegetation indices derived from multitemporal data from MODIS sensors is being studied in the search for greater objectivity of the generated values. In this context, the objective of this study was to map areas with winter and second-crop corn using EVI/MODIS time series from the Terra and Aqua satellites, for the seasons from 2012 to 2014 in the Paraná state of Brazil. Accuracy analysis of the mappings was performed in spatial resolution images of 30 m (LISS-III and Landsat-8), to identify and validate the masks the crops of interest. The accuracy of the mapping obtained values of global precision 87.5%, 79.5%, and 82.0%, with Kappa index of 0.81, 0.69, and 0.73, in the 2012, 2013, and 2014 harvests, respectively. Comparing with data from the Brazilian Institute of Geography and Statistics (IBGE), the areas obtained by the mappings were underestimated for the second-crop corn in the 2012 and 2013 seasons and overestimated in 2014. The winter crops were overestimated for the three seasons investigated. The use of remote sensing data and techniques can contribute to a quick estimation of crop area information, and can assist in the surveys conducted by official institutions.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

Mapping of Winter Crops and Second-Crop Corn in the Paraná State-Brazil, Using Multitemporal Images From MODIS Sensor

Santos¹,

Mercante²,

Johann³

et al. 2019

JAS

Self Cite

View full text Add to dashboard Cite

show abstract

“…Utilizing the reflectances obtained from the hyperspectral bands for the strips: 3, 4, 5, 6 e 7 of the Landsat 8 sensors, the image was classified through the pixel supervised method, employing the maximum likelihood algorithm (Maxver) (Tisot et al, 2007;Leite et al, 2011;Antunes et al, 2012;Fernandes et al, 2012;Mello et al, 2012;Prishchepov et al, 2012;Wrublack et al, 2013), considering 11 classes of predefined land features, with an average of ten samplings for feature training, and ten more samplings for the validation of the proposed model in each class. proposed classes: grape crop "Grape", mango crop "Mango", coconut crop "Coconut", banana crop "Banana", guava crop "Guava", acerola crop "Acerola", other crops "Others", areas with exposed soil "Solo_exp", urban areas "Area_urb", areas with natural vegetation "Veg_ nat" and water bodies "Water".…”

Section: Methodsmentioning

confidence: 99%

Mapping of the current land use in part of the Irrigated Perimeter Nilo Coelho, Petrolina-PE, Brazil

Ramos

Júnior

2019

Com. Sci.

View full text Add to dashboard Cite

The objective of this study was to quantify and spatialize the agricultural composition and current land use in part of the Irrigation perimeter Nilo Coelho from the Landsat 8 satellite digital image classification. At first we performed the unsupervised classification image related to January 2014, identifying the separability within natural vegetation, water bodies, agricultural crops and anthropized areas. Then the supervised classification Maxver model was performed where it was observed that the fruit classes “Grape” (5.41%) “Acerola” (3.75%), “Coconut” (6.14%) “Guava” (6.17%), “Banana” (2.29%) and “Mango” (8.18%) showed dominance in the study area. These classes stood out in terms of yield in the area, pointing out that in that expansion which occurs in an area with agricultural crops in the irrigated Nilo Coelho, crops of mango, coco, guava, acerola, banana and grape also grow. In addition, we observed the emergence of new areas with passion fruit crops, melon, watermelon, corn, beans, onions, pumpkin, sweet potatoes, tomatoes, bell peppers, among other crops. Furthermore, we also found the occurrence of pastures containing buffel and elephant grasses, intercropping or not. The estimated overall accuracy was 80.52%, and the mapping defined as very good, with Kappa index equal to 0.7783.

show abstract

“…-soma dos elementos da matriz em sua diagonal principal x i+ -soma total das observações para as linhas x +i -soma total das observações para as colunas N -número total de observações Para determinar os valores do Kappa e da Exatidão Global foram utilizados 5000 pontos amostrais distribuídos aleatoriamente, gerados no TerraView 4.2.0 que foram analisados como soja ou não-soja (áreas ocupadas pela vegetação natural ou por outros usos antrópicos) com base no conhecimento dos intérpretes e com apoio da segmentação orientada em geo-objeto e séries-temporais (Freitas et al, 2011). Além disto, foi medida a precisão de cada categoria (soja e não-soja), utilizando-se o erro do ponto de vista do produtor e do consumidor, por meio da análise dos erros de inclusão e dos erros de omissão presentes nas classificações (Antunes et al, 2012).…”

Section: Protocolo Experimental E Análise Estatísticaunclassified

Discriminação de áreas de soja por meio de imagens EVI/MODIS e análise baseada em geo-objeto

Silva¹,

Frank²,

Rodrigues³

2014

Rev. bras. eng. agríc. ambient.

View full text Add to dashboard Cite

R E S U M ONeste trabalho objetivou-se estimar e mapear áreas plantadas com soja [Glycine max (L.) Merr.] por meio de imagens multitemporais EVI/MODIS e classificação de imagens baseada em geo-objeto. A área de estudo compreendeu o Sul do estado do Maranhão. Para o mapeamento das lavouras de soja foram utilizados o índice de vegetação realçado (EVI) e o índice de valorização das culturas (CEI) para a classificação das imagens do sistema-sensor Terra/MODIS. Para tal cálculo foram utilizadas doze imagens compreendendo entressafra e safra da cultura, conforme calendário agrícola do Estado. Além disto, foi empregada a segmentação utilizando-se parâmetros de escala 250, os algoritmos "classification" e "merge region" e extração de atributos para classificação baseada em geo-objeto. Foram empregados, para avaliar a precisão da classificação, os parâmetros Kappa e Exatidão Global e nas suas resultantes foi aplicado o teste Z; logo, foram estabelecidos, como hipótese nula (H0) a igualdade dos índices e o inverso para suas diferenças (H1), a um nível de 0,05 de significância. Os resultados obtidos indicam que a metodologia proposta se mostrou eficiente para mapeamento da soja, com 0,89 para o parâmetro Kappa. Discrimination of soybean areas through images EVI/MODIS and analysis based on geo-object A B S T R A C TThis study aimed to estimate and map areas planted with soybean [Glycine max (L.) Merr.] through multitemporal images EVI/MODIS and classification of images based on geo-object. The study area comprised the southern part of the State of Maranhão. For the mapping of the soybean crop the Enhanced Vegetation Index (EVI) and the Crop Enhancement Index (CEI) for image classification sensor-system Terra/MODIS was used. For this calculation twelve images were used, including offseason and harvest of the crop, as per the state agricultural calendar. In addition, the segmentation was employed using scaling parameters 250, the algorithms "classification" and "merge region", and extracting attributes for classification GEOgraphic-Object-Based Image Analysis (GEOBIA). To assess the accuracy of the classification, parameters Kappa and Accuracy Global and its resulting Z test were applied. A null hypothesis (H0) of equal and opposite rates for their differences (H1) at 0.05 level of significance was established. The results indicate that the proposed methodology is efficient for mapping the soybean crop, with 0.89 for the parameter Kappa.Discriminação de áreas de soja por meio de imagens EVI/MODIS e análise baseada em geo-objeto

show abstract

Estimativa de área de soja por classificação de imagens normalizada pela matriz de erros

Cited by 16 publications

References 18 publications

Mapping of Winter Crops and Second-Crop Corn in the Paraná State-Brazil, Using Multitemporal Images From MODIS Sensor

Mapping of Winter Crops and Second-Crop Corn in the Paraná State-Brazil, Using Multitemporal Images From MODIS Sensor

Mapping of the current land use in part of the Irrigated Perimeter Nilo Coelho, Petrolina-PE, Brazil

Discriminação de áreas de soja por meio de imagens EVI/MODIS e análise baseada em geo-objeto

Contact Info

Product

Resources

About