TOWARDS TRUSTWORTHY WILDFIRE DETECTION: INTEGRATING EXPLAINABILITY AND UNCERTAINTY IN DEEP LEARNING MODELS
Ключевые слова:
trustworthy artificial intelligence; wildfire detection; Grad-CAM; uncertainty calibration; deep learning.Аннотация
Лесные пожары стали круглогодичной, усиливаемой
климатом угрозой, что повышает роль глубокого обучения для их раннего
визуального обнаружения. В статье представлена система BlazeVeritas AI,
объединяющая CNN, ResNet-18 и DenseNet-121 с объяснимостью Grad-CAM
и оценкой неопределённости на основе Monte Carlo Dropout и Temperature
Scaling, реализованная средствами FastAPI и Streamlit. Результаты
показывают, что совместное применение объяснимости и калиброванной
неопределённости повышает надёжность систем поддержки принятия
решений.
Библиографические ссылки
Jain, P., Coogan, S. C. P., Subramanian, S. G., Crowley, M., Taylor, S., & Flannigan, M. D. (2020). A review of machine learning applications in wildfire science and management. Environmental Reviews, 28(4), 478–505.
Khan, S., Khan, A., Maqsood, M., Aadil, F., & Ghazanfar, M. A. (2022). Convolutional neural networks for fire and smoke detection: A comprehensive survey. Multimedia Tools and Applications, 81(24), 35143–35180.
Guo, C., Pleiss, G., Sun, Y., & Weinberger, K. Q. (2017). On calibration of modern neural networks. In Proceedings of the 34th International Conference on Machine Learning (ICML) (pp. 1321–1330).
Frizzi, S., Kaabi, R., Bouchouicha, M., Ginoux, J.-M., Moreau, E., & Fnaiech, F. (2016). Convolutional neural network for video fire and smoke detection. In Proceedings of the 42nd Annual Conference of the IEEE Industrial Electronics Society (IECON) (pp. 877–882).
Park, M., Tran, D. Q., Jung, D., & Park, S. (2020). Wildfire-detection method using DenseNet and CycleGAN data augmentation-based remote camera imagery. Remote Sensing, 12(22), Article 3715.
Gal, Y., & Ghahramani, Z. (2016). Dropout as a Bayesian approximation: Representing model uncertainty in deep learning. In Proceedings of the 33rd International Conference on Machine Learning (ICML) (pp. 1050–1059).
Selvaraju, R. R., Cogswell, M., Das, A., Vedantam, R., Parikh, D., & Batra, D. (2017). Grad-CAM: Visual explanations from deep networks via gradient-based localization. In Proceedings of the IEEE International Conference on Computer Vision (ICCV) (pp. 618–626).
He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (pp. 770–778).
Naeini, M. P., Cooper, G. F., & Hauskrecht, M. (2015). Obtaining well calibrated probabilities using Bayesian binning. In Proceedings of the 29th AAAI Conference on Artificial Intelligence (pp. 2901–2907).