跳到正文
研踪PaperTrail
← 研踪 PaperTrail

arXiv:2611.20036 · v1

Machine Learning for Near Real-Time Identification of Wildfire Burned Areas

首次提交 2026-11-15 22:29
版本更新 2026-11-15 22:29
雷达观测 2026-11-17 12:06
arXiv ↗PDF ↗

摘要

In this article, we present a near real-time method that estimates burned probability anywhere in the world using satellite images. Our approach takes advantage of the state-of-the-art pix2pix model to translate a preprocessed index calculated from Landsat 8/9 images into an approximate of USGS Burn Probability indicating per pixel burn probability. Since this is a machine learning-based approach, it can be automated with a minimum amount of human intervention with significantly less time delays and geographic restrictions compared to other contemporary methods. We include results from our experiments and compare our approximation with the USGS Burn Probability using RMSE score.

AI 论文解读

基于摘要 · 非全文审阅

此版本尚未生成 AI 解读。解读由后台任务生成并按论文版本保存,登录后共享已有结果。

版本差异

目前只收录了一个版本。收录下一版后,可以对照标题、作者、分类和摘要。

观测到的版本

相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。

v1 · 2026-11-15 22:29 · 最新观测版本

Machine Learning for Near Real-Time Identification of Wildfire Burned Areas

In this article, we present a near real-time method that estimates burned probability anywhere in the world using satellite images. Our approach takes advantage of the state-of-the-art pix2pix model to translate a preprocessed index calculated from Landsat 8/9 images into an approximate of USGS Burn Probability indicating per pixel burn probability. Since this is a machine learning-based approach, it can be automated with a minimum amount of human intervention with significantly less time delays and geographic restrictions compared to other contemporary methods. We include results from our experiments and compare our approximation with the USGS Burn Probability using RMSE score.

查看此版本解读 →查看此版本原文 ↗