Banca de DEFESA: JOYCE DE AGUIAR CARVALHO

Uma banca de DEFESA de MESTRADO foi cadastrada pelo programa.
STUDENT : JOYCE DE AGUIAR CARVALHO
DATE: 30/08/2022
TIME: 08:30
LOCAL: Virtual
TITLE:

SPECTRAL ANALYSIS OF WEEDS IN SORGHUM CULTURE (BICOLOR SORGHUM) IN THE VISIBLE REGION

KEY WORDS:

Spectral signature, Precision agriculture, Multivariate analysis.

PAGES: 59
BIG AREA: Ciências Agrárias
AREA: Engenharia Agrícola
SUBÁREA: Máquinas e Implementos Agrícolas
SUMMARY:

Sorghum is a rustic plant that stands out for its high biomass production and tolerance to water deficit, making it a good alternative for agricultural diversification. A latent factor that causes problems in sorghum production is the presence of weeds, which compete for nutrients and water with the crop. The study of spectral analysis of plants by spectroradiometer can be an ideal tool for decision making, obtaining faster results and in a non-destructive way for crop management. The present work aims to define the appropriate spectral range to differentiate sorghum weeds in the RGB spectral region by means of the spectroradiometer. The experiment was carried out at UFRRJ, located in Seropédica RJ, in an area of 1 ha of sorghum production. Forty control points were analyzed on five different dates for collecting the spectral signatures of the plants. Spectral analysis was performed for weeds and sorghum by an ASD FIELDSPEC® 4 spectroradiometer. The spectral readings were analyzed using a multivariate statistical technique, principal component analysis (PCA), in order to obtain consistent classifications. regarding the discrimination of the spectral bands for the plant species in the area. As a result of the study, it was observed that the plants showed different behavior for the five days of collection, mainly in the green and red regions. The principal component analysis allowed the reduction of the data size and the discrimination of the spectral bands that differentiated the groups of weed species and Sorghum. The green region showed greater differentiation between plants for all collections. The ideal spectral ranges for species distinction in the red region were the wavelengths 689-693nm for the first collection, 650-662nm for the second collection and 657-660nm for the last collection and in the green region of 570-580nm . This spectral difference between plants is due to the pigment content of each species, such as chlorophyll and carotenoids. The results obtained contributed to the discrimination of weed and sorghum species in the RGB spectral region, showing potential for use to discriminate species, which makes it a more accessible alternative.

BANKING MEMBERS:
Presidente - 1050615 - MURILO MACHADO DE BARROS
Interno - 2161955 - ANDERSON GOMIDE COSTA
Externo à Instituição - FLÁVIO CASTRO DA SILVA - UFF