Active learning for interactive satellite image change detection
This work addresses the challenge of efficient annotation for satellite image change detection, particularly after natural hazards, but appears incremental as it builds on existing active learning concepts.
The authors tackled the problem of satellite image change detection by introducing an interactive active learning algorithm that selects the most informative image pairs for annotation, achieving relevance against related work in experiments on tornado-affected areas.
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.