Banca de QUALIFICAÇÃO: THAÍS MACHADO DE SOUZA

Uma banca de QUALIFICAÇÃO de DOUTORADO foi cadastrada pelo programa.
STUDENT : THAÍS MACHADO DE SOUZA
DATE: 02/07/2026
TIME: 13:30
LOCAL: Galpão de Máquinas Agrícolas - IT/DE
TITLE:

AUTOMATIC CLASSIFICATION OF AGRICULTURAL CROPS (MAIZE AND SORGHUM) USING COMPUTER VISION, DEEP LEARNING, AND INTEGRATION INTO A QGIS ENVIRONMENT

KEY WORDS:

artificial intelligence, remote sensing, geotechnologies, precision agriculture, weeds.

PAGES: 28
BIG AREA: Ciências Agrárias
AREA: Engenharia Agrícola
SUBÁREA: Máquinas e Implementos Agrícolas
SUMMARY:

Weed management remains a major challenge in modern agriculture due to its direct impacts on crop productivity, production costs, and the broadcast application of herbicides. In this context, precision agriculture combined with artificial intelligence has emerged as a strategic approach to optimize crop monitoring and management in maize and sorghum production systems. This project proposes the development of a computer vision and deep learning-based system capable of automatically recognizing the main crop present in an agricultural field and identifying as weeds those plants exhibiting characteristics distinct from the target crop. The system will be developed in a Python environment using deep learning architectures applied to image classification, object detection, and semantic segmentation of images acquired by Remotely Piloted Aircraft Systems, enabling high spatial resolution and detailed characterization of the analyzed areas. In addition, images acquired at different phenological stages of both crops and weeds will be considered to increase the robustness and generalization capability of the computational models under real field conditions. Subsequently, the proposed solution will be integrated into the QGIS environment through a plug-in capable of generating georeferenced weed infestation maps, enabling spatial analyses and supporting decisionmaking for site-specific weed management. The model is expected to contribute to sitespecific herbicide application, improved operational efficiency, and the strengthening of sustainable agricultural practices. Furthermore, this study aims to integrate artificial intelligence, remote sensing, geotechnologies, and precision agriculture into an applied, reproducible tool with potential use by researchers, farmers, and agricultural professionals.

COMMITTEE MEMBERS:
Presidente - 1050615 - MURILO MACHADO DE BARROS
Interno - 2161955 - ANDERSON GOMIDE COSTA
Externa ao Programa - 3132531 - PRISCILA DE LIMA E SILVA - UFRRJ