Banca de QUALIFICAÇÃO: JHIORRANNI FREITAS SOUZA

Uma banca de QUALIFICAÇÃO de MESTRADO foi cadastrada pelo programa.
STUDENT : JHIORRANNI FREITAS SOUZA
DATE: 03/12/2020
TIME: 09:30
LOCAL: Google meet
TITLE:

USE OF COMPUTATIONAL VIEW OF MACHINES FOR ESTIMATION OF QUALITY ATTRIBUTES OF TIFTON 85 (Cynodon spp.)

KEY WORDS:

Multivariate statistics; digital images; spectral indices; physical-chemical attributes

PAGES: 20
BIG AREA: Ciências Agrárias
AREA: Engenharia Agrícola
SUBÁREA: Máquinas e Implementos Agrícolas
SUMMARY:

The world forage trade is expanding due to the demand for roughage to supply the animals' food. The application of technology from computer vision can be an alternative to traditional methods of analysis, making the process cheaper and allowing obtaining non-destructive samples with results instantly. The present work aims to estimate quality attributes of Tifton 85 grass (Cynodon spp.) And hay produced using a computer vision system, relating the colorimetric indices to the physical-chemical and nutritional attributes of the crop. The experiment will be conducted in two parts, in an agricultural production area located at the Federal Rural University of Rio de Janeiro, in the municipality of Seropédica, RJ. In the first experiment, the colorimetric indices obtained through digital images (in the visible and near infrared spectral region) of the cultivar Tifton-85 (Cynodon spp.) Will be used to estimate the moisture content, dry matter productivity, level of nitrogen, leaf area index, leaf / stem ratio, crude fiber, crude energy and crude protein. 50 images will be obtained at different cutting ages (14, 28, 42 and 56 days), totaling 200 samples. For the development of the model, the principal component regression method (PCR) will be applied. In the second experiment, colorimetric indices will be obtained from digital images (in the visible and near infrared spectral region) of hay from cultivar Tifton 85 (Cynodon spp.) To estimate the moisture and dry matter content. 500 samples obtained at different times during the collection period and the ideal storage condition will be used. For the development of the model, the principal component regression method (PCR) and regression by partial minimum squares (PLS) will be applied. It is expected at the end of the project to obtain the colorimetric characteristics represented by the spectral indices and quality attributes of Tifton 85 grass (Cynodon spp.) At different cutting ages and their hays stored in different moisture and dry matter contents. Thus, comparing their ability to distinguish and measure by an artificial vision system, and by traditional laboratory methods, using multivariate statistical analysis techniques.

BANKING MEMBERS:
Presidente - 2161955 - ANDERSON GOMIDE COSTA
Externo ao Programa - 386710 - CARLOS ALBERTO ALVES VARELLA
Externo ao Programa - 1827405 - MARCUS VINICIUS MORAIS DE OLIVEIRA
Externa à Instituição - AMELIA LAÍSY DO NASCIMENTO - UFRPE