Effectiveness of Machine Learning in environmental pollution from remote sensing images (#1402)

Authors

Chafloque-Llontop, Carlos Efraín

Sánchez-Rentería, Raúl Armando

Dios-Castillo, Christian

Abstract

Machine learning algorithms for the analysis of remotely sensed images allow identifying and tracking sources of environmental contamination effectively. The objective or purpose of this research was to evaluate the effectiveness of these techniques in the identification and tracking of environmental pollution sources through remote sensing images. The methodology proposed was the systematic review of scientific articles on the subject in question, supported by PICO questions and the PRISMA procedure for the analysis of these articles. The information was obtained from the Scopus database, obtaining 58 open access documents from the last 5 years. The results found show that machine learning algorithms allow a more accurate identification of sources of environmental pollution compared to conventional methods. Thus, it is concluded that these techniques are a valuable tool for the identification and monitoring of environmental pollution sources, which can contribute to decision making and the implementation of measures to reduce environmental pollution. Keywords: machine learning, remote sensing, environmental contamination, deep learning