Talk · IEEE International Conference on Intelligent Systems · September 1, 2018
Deep Learning Approach for Building Detection in Satellite Multispectral Imagery
Geesara Kulathunga, Ilya M. Afanasyev
Abstract
Building detection from satellite multispectral imagery data is being a fundamental but a challenging problem mainly because it requires correct recovery of building footprints from high-resolution images. In this work, we propose a deep learning approach for building detection by applying numerous enhancements throughout the process. Initial dataset is preprocessed by 2-sigma percentile normalization. Then data preparation includes ensemble modelling where 3 models were created while incorporating OpenStreetMap data. Binary Distance Transformation (BDT) is used for improving data labeling process and the U-Net (Convolutional Networks for Biomedical Image Segmentation) is modified by adding batch normalization wrappers. Afterwards, it is explained how each component of our approach is correlated with the final detection accuracy. Finally, we compare our results with winning solutions of SpaceNet 2 competition for real satellite multispectral images of Vegas, Paris, Shanghai and Khartoum, demonstrating the importance of our solution for achieving higher building detection accuracy.
More by Ilya Afanasyev
- Geometric State Fusion for Autonomous Agents: A Comparative Analysis of Dual Quaternion Observer and Kalman FiltersAdaptive Agents and Multi-Agents Systems · 2026
- Beyond Kalman Filtering: a Dual Quaternion Geometric Observer for Robust 6DoF Pose EstimationConference of Open Innovations Association · 2026
- Efficient Document Image Dewarping via Hybrid Deep Learning and Cubic Polynomial Geometry RestorationResearch Square · 2025 · Citations: 1
- Modern Methods of Map Construction Using Optical Sensors FusionProceedings of International Conference on Artificial Life and Robotics · 2023 · Citations: 6
