Unsupervised Wildfire Change Detection based on Contrastive Learning
Unsupervised Wildfire Change Detection based on Contrastive Learning
复制标题
基于对比学习的无监督野火变化检测
DOI：
--
复制
发表时间：
2022
期刊：
影响因子：
--
通讯作者：
Vít Ruzicka
中科院分区：
文献类型：
--
作者：
Beichen Zhang;Huiqi Wang;Amani Alabri;K. Bot;Cole McCall;Dale A. Hamilton;Vít Ruzicka
准确描述野火事件的严重程度有助于描述火灾易发地区的燃料状况，并为灾害应对提供有价值的信息。这项研究的目的是开发一个建立在高分辨率多光谱卫星图像之上的自主系统，并采用先进的深度学习方法来检测烧毁面积的变化。这项工作提出了一个初步的探索使用无监督模型的野火场景中的特征提取。它基于对比学习技术Simplified，该技术经过训练以最小化图像增强之间的余弦距离。编码图像之间的距离也可以用于变化检测。我们建议改变这种方法，使其能够用于无监督的烧伤面积检测和后续的下游任务。我们表明，我们提出的方法优于测试的基线方法。
The accurate characterization of the severity of the wildfire event strongly contributes to the characterization of the fuel conditions in fire-prone areas, and provides valuable information for disaster response. The aim of this study is to develop an autonomous system built on top of high-resolution multispectral satellite imagery, with an advanced deep learning method for detecting burned area change. This work proposes an initial exploration of using an unsupervised model for feature extraction in wildfire scenarios. It is based on the contrastive learning technique SimCLR, which is trained to minimize the cosine distance between augmentations of images. The distance between encoded images can also be used for change detection. We propose changes to this method that allows it to be used for unsupervised burned area detection and following downstream tasks. We show that our proposed method outperforms the tested baseline approaches.