Bibliographic citations
Fernandez, K., Mendoza, J. (2024). Análisis de clasificadores de imágenes satelitales en deslizamiento de tierra con técnicas de machine learning aplicado al distrito de Ancash [Trabajo de Suficiencia Profesional, Universidad Peruana de Ciencias Aplicadas (UPC)]. http://hdl.handle.net/10757/675217
Fernandez, K., Mendoza, J. Análisis de clasificadores de imágenes satelitales en deslizamiento de tierra con técnicas de machine learning aplicado al distrito de Ancash [Trabajo de Suficiencia Profesional]. PE: Universidad Peruana de Ciencias Aplicadas (UPC); 2024. http://hdl.handle.net/10757/675217
@misc{renati/502444,
title = "Análisis de clasificadores de imágenes satelitales en deslizamiento de tierra con técnicas de machine learning aplicado al distrito de Ancash",
author = "Mendoza Bernardillo, Jhon Brayan",
publisher = "Universidad Peruana de Ciencias Aplicadas (UPC)",
year = "2024"
}
The detection of landslides through satellite images represents a significant advance in the prevention of natural disasters. Therefore, the implementation of machine learning classifiers is a complex process that begins with the careful selection and preprocessing of high-resolution images from reliable satellites, such as Landsat 8 or Sentinel. These images are essential to accurately identify areas at risk, where the implementation of machine learning classifiers requires meticulous selection and preprocessing of high-resolution images from trusted sources such as Landsat 8 or Sentinel. Since, according to the algorithms evaluated in this project, the Support Vector Machine (SVM) has proven to be exceptionally effective, with an overall precision index of 0.972, outperforming alternative models such as Random Forest, decision trees, Naive Bayes and Neural Networks. . Where, the reliability of these classifiers is measured by comparing their predictions with historical data and records of real events, which allows their predictive capacity to be refined. Likewise, the continuous evaluation and improvement process not only increases the precision in the identification of areas susceptible to landslides, but also reinforces risk management strategies. This is crucial for effective preparation and response to possible disasters, allowing the implementation of preventive measures and the rapid mobilization of resources in the event of an emergency. Ultimately, these efforts help protect lives, infrastructure and ecosystems, mitigating the impact of landslides on vulnerable communities.
This item is licensed under a Creative Commons License