• JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
 
  Bookmark and Share
 
 
Doctoral Thesis
DOI
https://doi.org/10.11606/T.74.2016.tde-03112016-160723
Document
Author
Full name
Thiago Pereira Motta
E-mail
Institute/School/College
Knowledge Area
Date of Defense
Published
Pirassununga, 2016
Supervisor
Committee
Luz, Pedro Henrique de Cerqueira (President)
Avanzi, Junior Cesar
Baesso, Murilo Mesquita
Heinrichs, Reges
Silva, Silvia Helena Modenese Gorla da
Title in Portuguese
Índices espectrais de imagens coloridas na avaliação do estado nutricional de fósforo, enxofre e magnésio no capim-marandu
Keywords in Portuguese
Banda do verde
Banda do vermelho
Nutrição da pastagem
Abstract in Portuguese
O presente trabalho foi conduzido com o objetivo de avaliar a nutrição da Urochloa brizantha (Syn. Brachiaria brizantha) cv. Marandu para P, Mg e S através do desenvolvimento das plantas, sintomas visuais e índices espectrais das bandas do visível. Os experimentos foram conduzidos em casa de vegetação localizada no campus da FZEA/USP. As plantas foram cultivadas em solução nutritiva, onde foram fornecidas as seguintes doses de P, Mg e S: 6%, 20%, 100% e 200% do recomendado, em blocos casualizados. As imagens digitais de scanner das folhas diagnósticas foram processadas e classificadas utilizando MATLAB® R2012b com pacote de ferramentas para processamento de imagens. Foram utilizados vetores de características baseados nas bandas do visível, sendo calculados os seguintes índices: excesso de verde; vermelho normalizado; verde normalizado; e razão verde-vermelho. Na classificação foi utilizado classificador bayesiano. As doses de P, Mg e S causaram aumento significativo (P<0,05) no acúmulo de massa seca da parte aérea (MSPA) e nos teores foliares dos nutrientes dosados. As doses de P, S e Mg promoveram manifestações dos sintomas de deficiência característicos que se traduzem em modificações no perfil de coloração das folhas. A Porcentagem de Acerto Global (PAG) e o índice Kappa atingiram valores acima de 80% e 0,81, respectivamente, sendo consideradas classificações excelentes. As melhores classificações foram obtidas quando se utilizou mais de um índice como vetor de característica. Índices espectrais do verde e do vermelho são bons descritores do status da nutrição do P, Mg e S no capim-marandu.
Title in English
Spectral indices of color images in the evaluation of phosphorus, súlfur and magnesium nutritional status in Marandu grass
Keywords in English
Green band
Nutrition of pastureland
Red band
Abstract in English
The present study was conducted to assess the nutrition of Urochloa brizantha (SYN. Brachiaria brizantha) CV. Marandu to P, Mg and S through the development of plants, Visual symptoms and spectral indices of the visible bands. The experiments were conducted in a greenhouse located on the campus of the FZEA/USP. The plants were grown in nutrient solution, where they were provided the following doses of P, Mg, and S: 6%, 20%, 100% and 200% of the recommended, in randomized blocks. The digital images from scanner diagnostic sheets were processed and classified using MATLAB ® R2012b with package of tools for processing images. Characteristic vectors were used based on bands of the visible, being calculated the following indexes: excess Green; standard red; standardised Green; Red-Green ratio. The classification has been used Bayesian classifier. The doses of P, Mg, and S have caused significant increase (P < 0.05) in acúmul shoot dry mass (MSPA) and foliar nutrient levels dosed. The doses of P, S and Mg promoted demonstrations of characteristic deficiency symptoms that result in changes in the profile of colouring of leaves. The Global hit Percentage (PAG) and the Kappa index reached values above 80%, and 0.81, respectively, being considered excellent ratings. The best ratings were obtained when using more than one index as a feature vector. Spectral indices of green and Red are good descriptors of the nutrition status of P, Mg and S marandu grass.
 
WARNING - Viewing this document is conditioned on your acceptance of the following terms of use:
This document is only for private use for research and teaching activities. Reproduction for commercial use is forbidden. This rights cover the whole data about this document as well as its contents. Any uses or copies of this document in whole or in part must include the author's name.
DO8193290COR.pdf (3.04 Mbytes)
Publishing Date
2016-11-11
 
WARNING: Learn what derived works are clicking here.
All rights of the thesis/dissertation are from the authors
CeTI-SC/STI
Digital Library of Theses and Dissertations of USP. Copyright © 2001-2024. All rights reserved.