MAPPING OF WEEDS IN SWEET POTATO CROPS USING HIGH SPATIAL RESOLUTION IMAGES

• Grant number: 23/13500-3
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: December 01, 2023
• End date: November 30, 2024
• Field of knowledge: Physical Sciences and Mathematics - Geosciences - Geodesy
• Principal Investigator: Fernanda Sayuri Yoshino Watanabe
• Grantee: Alexandre Sossae de Souza
• Host Institution: Faculdade de Ciências e Tecnologia (FCT). Universidade Estadual Paulista (UNESP). Campus de Presidente Prudente. Presidente Prudente , SP, Brazil
• Associated research grant: 21/06029-7 - High resolution remote sensing for digital agriculture, AP.TEM

Abstract

The presence of weeds is a common problem in crops, which can cause a drop in agricultural production, as this vegetation starts to compete with the crop for resources necessary for plant development, such as water, nutrients and light. Weed control is generally done through the use of herbicides, which increases food production costs, in addition to generating impacts on public health and the environment, as well as a drop in agricultural yield when applied in large quantities. In this context, the early detection of these plants in the crop allows an appropriate management plan to be implemented to control these plants, preventing an infestation from occurring throughout the crop, reducing the use of herbicides. The use of high-resolution aerial images is becoming increasingly popular in weed detection. In this sense, the objective of the present project is to identify weeds in a commercial sweet potato cultivation area, on a rural property in the western region of the State of São Paulo, based on the classification of aerial images acquired from the Agrowing multispectral camera, from 14 bands, attached to a drone. The Random Forest machine learning algorithm will be trained for weed detection and mapping.