JP

Organization

Research Division

Co-creation Center for Disaster Resilience
Associate Professor
Ph.D.(Engineering)
Research Subject(s)
My research interests including Photogrammetry & Remote Sensing, Computervision and Machine Learning. I mainly focus on Aerial/Satellite image based image processing such as automatic DEM/DSM generation, stereo matching, urban infrastructure extraction and change detection with application to Urban Monitering and Disaster Science.
Key Words
Photogrammetry / Remote Sensing / Machine Learning / Data Fusion / Computer Vision
Research Activities

The development of an efficient and robust method for dense image-matching has been a technical challenge due to high variations in illumination and ground features of aerial images of large areas. In this study, I propose a method for the dense matching of aerial images using an optical flow field and a fast-guided filter which is beneficial for automatic DSM/DEM generation.

Building detection is a fundamental and significant task that aims to locate all buildings in the remote sensing image. It can be an upstream task in various applications, such as urban planning, environment monitoring, and land resource utilization. I proposed  a combined CNN and transformer hybrid neural network for precision building detection from high-resolution remote sensing images which can be utilized in urban change monitoring and disaster assesment.

Aerial/Satellite image based specific feature extraction is crucial to a lot of applications such as autonomous driving, urban planning, and disaster assesment. However most of exsiting studies are focusing on pixel-level based segmentation, which are inadequate for practical applictions. In this study we propose CNN and Transformer based neural networks for automatic man-made infrastructures extraction and vectorization.

Selected Works

Yuan, W., Ran, W., Shi, X., and Shibasaki, R. (2023). Multi-Constraint Transformer based Automatic Building Extraction from High Resolution Remote Sensing Images. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,16, 9164-9174, 2023, doi: 10.1109/JSTARS.2023.3319826

Yuan, W., Yuan, X., Cai, Y., and Shibasaki, R. (2023). Fully automatic DOM generation method based on optical flow field dense image matching. Geo-spatial Information Science, 26(2),242-256, doi: 10.1080/10095020.2022.2159886

Yuan, W., Yuan, X., Fan, Z., Guo, Z., Shi, X., Gong, J., and Shibasaki, R. (2021). Graph neural network based multi-feature fusion for building change detection. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 43, 377-382. doi:10.5194/isprs-archives-XLIII-B3-2021-377-2021

Yuan, W., Yuan, X., Xu, S., Gong, J., and Shibasaki, R. (2019). Dense image-matching via optical flow field estimation and fast-guided filter refinement. Remote Sensing, 11(20), 2410. doi:10.3390/rs11202410

Selected Memberships
  • International Society of Photogrammetry and Remote Sensing
  • IEEE Society of Geoscience and Remote Sensing
  • IEEE
Selected Awards
  • Best paper award, ISPRS 2022 (co-author) (2022.7)
  • Outstanding graduates award, Hubei Privince, China (2012.6)