Bootstrapping Rare Object Detection in High-Resolution Satellite Imagery (2403.02736v1)
Abstract: Rare object detection is a fundamental task in applied geospatial machine learning, however is often challenging due to large amounts of high-resolution satellite or aerial imagery and few or no labeled positive samples to start with. This paper addresses the problem of bootstrapping such a rare object detection task assuming there is no labeled data and no spatial prior over the area of interest. We propose novel offline and online cluster-based approaches for sampling patches that are significantly more efficient, in terms of exposing positive samples to a human annotator, than random sampling. We apply our methods for identifying bomas, or small enclosures for herd animals, in the Serengeti Mara region of Kenya and Tanzania. We demonstrate a significant enhancement in detection efficiency, achieving a positive sampling rate increase from 2% (random) to 30%. This advancement enables effective machine learning mapping even with minimal labeling budgets, exemplified by an F1 score on the boma detection task of 0.51 with a budget of 300 total patches.
- “Damage identification and assessment using image processing on post-disaster satellite imagery,” in 2017 IEEE Global Humanitarian Technology Conference (GHTC). IEEE, 2017, pp. 1–7.
- “Rapid building damage assessment workflow: An implementation for the 2023 rolling fork, mississippi tornado event,” arXiv preprint arXiv:2306.12589, 2023.
- “Synthetic imagery aided geographic domain adaptation for rare energy infrastructure detection in remotely sensed imagery,” in NeurIPS 2021 Workshop on Tackling Climate Change with Machine Learning, 2021.
- “Deep learning enables satellite-based monitoring of large populations of terrestrial mammals across heterogeneous landscape,” Nature communications, vol. 14, no. 1, pp. 3072, 2023.
- “Active learning for object detection in high-resolution satellite images,” arXiv preprint arXiv:2101.02480, 2021.
- “Few-shot object detection in remote sensing: Lifting the curse of incompletely annotated novel objects,” IEEE Transactions on Geoscience and Remote Sensing, 2024.
- “Harmonised global datasets of wind and solar farm locations and power,” Scientific data, vol. 7, no. 1, pp. 130, 2020.
- “Training rare object detection in satellite imagery with synthetic gan images,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021, pp. 2769–2776.
- “An improved aggregated-mosaic method for the sparse object detection of remote sensing imagery,” Remote Sensing, vol. 13, no. 13, pp. 2602, 2021.
- “Fine-grain uncommon object detection from satellite images,” in 2015 IEEE Applied Imagery Pattern Recognition Workshop (AIPR). IEEE, 2015, pp. 1–6.
- “Landscape dynamics (landdx) an open-access spatial-temporal database for the kenya-tanzania borderlands,” Scientific Data, vol. 9, no. 1, pp. 8, 2022.
- “A generalizable and accessible approach to machine learning with global satellite imagery,” Nature communications, vol. 12, no. 1, pp. 4392, 2021.
- “Deep residual learning for image recognition,” in Proceedings of the IEEE conference on computer vision and pattern recognition, 2016, pp. 770–778.
- “A comparison of document clustering techniques,” 2000.
- “A density-based algorithm for discovering clusters in large spatial databases with noise,” in kdd, 1996, vol. 96, pp. 226–231.
- Peter J Rousseeuw, “Silhouettes: a graphical aid to the interpretation and validation of cluster analysis,” Journal of computational and applied mathematics, vol. 20, pp. 53–65, 1987.
- “U-net: Convolutional networks for biomedical image segmentation,” in Medical Image Computing and Computer-Assisted Intervention–MICCAI 2015: 18th International Conference, Munich, Germany, October 5-9, 2015, Proceedings, Part III 18. Springer, 2015, pp. 234–241.
- “Aggregated residual transformations for deep neural networks,” in Proceedings of the IEEE conference on computer vision and pattern recognition, 2017, pp. 1492–1500.
- “Entropy regularization.,” .