Research Interests
My work is mainly concentrated on potential deep learning methods that can achieve excellent performance in building rooftop extraction. The CNN-based methods have shown their high performance and have been dominant in building rooftops extraction. However, due to size and shape diversities, occlusions, and complex scenarios, extracting building rooftops from aerial images accurately and efficiently can still be a great challenge. In order to address the obvious limitations, further deep learning methods need exploration. Therefore, my work will seek for new methods and try to improve the performance of these methods based on method-comparison. Moreover, my recent work is also focused on the application of deep learning methods on single-image super-resolution (SISR) for building dataset. The main focus of the work will be developing novel deep learning methods to improve the spatial resolution of the building dataset.
Education
- M.Eng., Mineral Investigation and Exploration, China University of Geosciences (Beijing), Sept 2017 – June 2021
- MES, Geography, University of Waterloo, Sept 2019 – Oct 2020
- B.Eng., Resources Exploration Engineering, Chengdu University of Technology, Sept 2013 – June 2017
Publication
- Cai, Y., He, H., Yang, K., Fatholahi, S. N., Li, J., 2021. A comparative study of deep learning approaches to rooftop detection in aerial images. Canadian Journal of Remote Sensing (3), 1-19.
- Wang, H., Zhang, F., Cai, Y., 2020. Sedimentary evolution characteristics and controlling factors of passive continental margin in the northern sub-basin of Senegal basin, Marine Geology and Quaternary Geology, 40(4):11.
- Jin, X., Zhang, G., Cai, Y., et al. 2017. Forming conditions and distribution of shallow gas in Daqing Placanticline. Petroleum Geology and Development of Daqing, 36(3):7-12.