Burned area semantic segmentation: A novel dataset and evaluation using convolutional networks

Abstract

Wildfires have significant impacts on the environment, society, and economy. Consequently, understanding its dynamics is crucial to evaluate such effects. Nonetheless, monitoring and measuring the burned area by traditional, non-automatic methods remains time-consuming and challenging. For several years, automatic semantic segmentation models have been used to describe natural phenomena, but deep learning models have recently achieved very competitive results. However, this new breed of models typically needs annotated datasets of significant dimensions. Nonetheless, datasets for real-time burnt area segmentation are often scarce. In this article, we create tools to support the benchmarking for testing and validating burned area segmentation models in a wildfire context. As such, we propose a new manually annotated dataset for segmentation of forest fire burned area based on a video captured by a UAV to train and evaluate semantic segmentation models. We suggest specific temporal consistency metrics to validate burned area polygons generated by the models in successive frames of non-annotated data. We also explore deep learning-based techniques and establish baselines, including IoU values superior to 95% on the test set.