Analysis of approaches for filling gaps in remote sensing image series

• Grant number: 24/13136-2
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: October 01, 2024
• End date: September 30, 2025
• Field of knowledge: Physical Sciences and Mathematics - Geosciences - Geophysics
• Principal Investigator: Rogério Galante Negri
• Grantee: Julia Bertoldo Ribeiro
• Host Institution: Instituto de Ciência e Tecnologia (ICT). Universidade Estadual Paulista (UNESP). Campus de São José dos Campos. São José dos Campos , SP, Brazil

Abstract

Environmental analyses have become increasingly important in recent years, providing a detailed understanding of the impacts of human activities on ecosystems and biodiversity. Remote sensing has emerged as an essential technology, enabling comprehensive spatial and temporal analyses. Change detection is particularly relevant for environmental studies, as it identifies regions subjected to alterations over time. However, atmospheric and geometric factors can disrupt the continuous availability of images, creating gaps in the time series that may compromise the analyses. This research project aims to study and analyze different approaches for filling gaps in remote sensing image series, enhancing support for environmental applications. Approaches based on machine learning models and autoregressive models for irregularly sampled data will be implemented and tested. The effectiveness of these approaches will be compared indirectly through the accuracy of the results in an environmental case study, mapping areas affected by wildfires in the region of Cáceres, Mato Grosso, using Landsat-8 image series from 2014 to 2024.