Modelos MatemÃticos para Estimativa da Ãrea Foliar de um Cafeeiro por Meio de AnÃlise de Imagens / Estimation models of leaf area in a coffee tree using image analysis.

AUTOR(ES)

Marlon Marcon

DATA DE PUBLICAÇÃO

2009

RESUMO

One feature of most horticultural crop plants that is biologically relevant to their yield and productivity is total leaf area. However, the direct methods of estimation of the leaf area cause damages to the plants and are time consuming, whereas the indirect methods based on image analysis, which are non-destructive, demand accuracy in the set up of the measurement procedure, which can be difficult to achieve. The development of computer tools and their accessible characteristics help the search for new indirect approaches which can be reliable for producers and researchers, with a good association with the real values produced by the direct methods. This work aims to build a model based on indirect measurements to estimate the leaf area in coffee trees using image analysis, light intensity measuring and the leaf area index (LAI) obtained by LAI-2000. The reference data were the total leaf area values obtained by destructive method of measuring area Li-3100c. The experiments were conducted at the Federal University of Lavras (Minas Gerais / Brazil) by using trees of an experimental crop of the Topazio cultivar with a canopy height ranging from 0.43 to 1.14 m. Two models were developed using images and some features from the trees were collected, such as the height and width and their projected area. Fisheye lens adapted to commercial digital camera were used to produce views from the plants, when two distinct models were created, one using an image captured above the plant and another using four images from below. In addition, three models were developed using data from a luximeter and from the LAI equipment (LAI 2000), where the light intensities were captured in the middle and in the bottom of the plant. Comparisons among the models and the actual figures showed R2 of 0.94 for the model using the projected area, 0.82 to plant height and width and 0.83 when the model was created with the above images. These results suggest that these models can be used to estimate the leaf area of a coffee tree. The other models, obtained by using luximeters and the LAI, presented low values and further studies need to be done, so that they can be used.

ASSUNTO(S)

1. leaf area index, 2. digital image processing, 3. fisheye lens, 4. digital luximeter, 5. optical metrology modelagem de sistemas biologicos 1. Ãndice de Ãrea foliar, 2. processamento de imagens digitais. 3. lente olho-de-peixe, 4. luximetro digital, 5. metrologia Ãptica.

Documentos Relacionados

• Modelos matemÃticos em anÃlise financeira de empresas, de setores industriais e de crÃdito.
• Modelos MatemÃticos para Esquistossomose
• Modelos matemÃticos para a previsÃo de cheias fluviais.
• Estimativa da área foliar total de culturas perenes por meio de análise de imagens
• Modelos matemÃticos e computacionais para simulaÃÃo do escoamento em motores de combustÃo interna.