Building Damage Mapping with Self-PositiveUnlabeled Learning

2021/11/04 by Xia, Junshi, Yokoya, Naoto, Adriano, Bruno · 1 citation
#Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #FOS: Electrical engineering #Image and Video Processing (eess.IV) #Machine Learning (cs.LG) #electronic engineering #information engineering

paper · doi:10.48550/arxiv.2111.02586

Abstract

Humanitarian organizations must have fast and reliable data to respond to disasters. Deep learning approaches are difficult to implement in real-world disasters because it might be challenging to collect ground truth data of the damage situation (training data) soon after the event. The implementation of recent self-paced positive-unlabeled learning (PU) is demonstrated in this work by successfully applying to building damage assessment with very limited labeled data and a large amount of unlabeled data. Self-PU learning is compared with the supervised baselines and traditional PU learning using different datasets collected from the 2011 Tohoku earthquake, the 2018 Palu tsunami, and the 2018 Hurricane Michael. By utilizing only a portion of labeled damaged samples, we show how models trained with self-PU techniques may achieve comparable performance as supervised learning.

Cited by

• Noisy-Pair Robust Representation Alignment for Positive-Unlabeled Learning

Related

• Self-Supervised Learning for Building Damage Assessment from Large-scale xBD Satellite Imagery Benchmark Datasets
• Assessing Post-Disaster Damage from Satellite Imagery using Semi-Supervised Learning Techniques
• Effective Damage Data Generation by Fusing Imagery with Human Knowledge Using Vision-Language Models
• Disaster mapping from satellites: damage detection with crowdsourced\n point labels
• DeepDamageNet: A two-step deep-learning model for multi-disaster building damage segmentation and classification using satellite imagery