Active learning for interactive satellite image change detection

Abstract

We introduce in this paper a novel active learning algorithm for satellite image change detection. The proposed solution is interactive and based on a question and answer model, which asks an oracle (annotator) the most informative questions about the relevance of sampled satellite image pairs, and according to the oracle's responses, updates a decision function iteratively. We investigate a novel framework which models the probability that samples are relevant; this probability is obtained by minimizing an objective function capturing representativity, diversity and ambiguity. Only data with a high probability according to these criteria are selected and displayed to the oracle for further annotation. Extensive experiments on the task of satellite image change detection after natural hazards (namely tornadoes) show the relevance of the proposed method against the related work.

References in corpus (10)

• Matching Networks for One Shot Learning
• Bayesian Active Learning for Classification and Preference Learning
• S2Looking: A Satellite Side-Looking Dataset for Building Change Detection
• Self-supervised Learning of Motion Capture
• Preserving Semantic Relations for Zero-Shot Learning
• Automated Scalable Bayesian Inference via Hilbert Coresets
• Meta-Learning Transferable Active Learning Policies by Deep Reinforcement Learning
• Hi-UCD: A Large-scale Dataset for Urban Semantic Change Detection in Remote Sensing Imagery
• Epipolar-Guided Deep Object Matching for Scene Change Detection
• A Weakly Supervised Convolutional Network for Change Segmentation and Classification

Cited by in corpus (2)

• Semantic-aware Dense Representation Learning for Remote Sensing Image Change Detection
• Bayesian Active Meta-Learning for Few Pilot Demodulation and Equalization