| Texto completo | |
| Autor(es): |
Alves, Larissa Mioni Vieira
;
Negri, Rogerio Galante
;
Dias, Maurcio Araujo
Número total de Autores: 3
|
| Tipo de documento: | Artigo Científico |
| Fonte: | TRANSACTIONS IN GIS; v. 29, n. 3, p. 12-pg., 2025-05-01. |
| Resumo | |
Change detection is a type of technique applied to remotely sensed data to map temporal changes. This approach serves as a vital tool for assessing the impacts of disasters, offering large-scale data acquisition at a lower cost. This study introduces a novel fully automatic and computationally efficient change detection technique designed for large multispectral remote sensing image series. The proposed method exploits the concept of thresholding deviations observed in time series data. To demonstrate the effectiveness of the proposed technique, case studies were conducted on two disaster-affected areas in 2023: one in Brazil, impacted by a landslide, and the other in Mozambique, affected by flooding, using data from Landsat-8 and Sentinel-2, respectively. The results showed superior accuracy compared to an alternative technique reported in the literature, with F1-Scores of 22.09% and 0.28% higher in the first and second study areas, respectively. Additionally, qualitative analyses revealed that the developed method more effectively identifies disaster-affected areas, requiring significantly less processing time than the alternative method. (AU) | |
| Processo FAPESP: | 24/01610-1 - Abordagens baseadas em redes neurais profundas para detecção de mudanças via séries de imagens de sensoriamento remoto |
| Beneficiário: | Rogério Galante Negri |
| Modalidade de apoio: | Auxílio à Pesquisa - Regular |
| Processo FAPESP: | 21/01305-6 - Avanços teóricos em detecção de anomalias e construção de sistemas de monitoramento ambiental |
| Beneficiário: | Rogério Galante Negri |
| Modalidade de apoio: | Auxílio à Pesquisa - Regular |