We found a match
Your institution may have rights to this item. Sign in to continue.
- Title
Estimativa de severidade do mofo-branco em lavouras de feijão utilizando-se sensores hiper e multiespectral.
- Authors
Machado, Marley L.; Pinto, Francisco A. C.; de Queiroz, Daniel M.; de Paula Júnior, Trazilbo J.; Vieira, Rogério F.
- Abstract
This study aimed to identify wavelengths and spectral ranges from hyper and multispectral reflectance using PLS regression; and to promote comparative evaluation of these methods and ten vegetation indices to determine those that best estimate levels of white mold severity in common beans. Experiments were implemented in the municipalities of Viçosa and Oratorios, Minas Gerais state. Hyperspectral reflectance measurements were acquired with the spectroradiometer, whose useful reading range was between 440 and 900 nm. Multispectral reflectance measurements were obtained from camera images comprising five bands (red, green, blue, red-edge and infrared). The indexes of disease severity were low. In Viçosa the average was 5.8% and in Oratorios, 7.4%. Mathematical models using hyperspectral reflectance performed better for estimating white mold. The red-edge band presented the wavelengths that best estimate the severity of white mold. Vegetation index resistant to the soil reflectance effects was better to estimate white mold than the other indices.
- Publication
Revista Brasileira de Engenharia Agricola e Ambiental - Agriambi, 2015, Vol 19, Issue 5, p426
- ISSN
1807-1929
- Publication type
Article
- DOI
10.1590/1807-1929/agriambi.v19n5p426-432