Songhan Luan

With a background in geography and commerce, I joined the PhD program in geography to expand my knowledge and expertise. In addition to my academic pursuits, I have hands-on experience in urban planning, which allows me to approach remote sensing applications from multiple perspectives, considering both societal and spatial dimensions. My passion lies in improving SAR image quality through advanced super-resolution techniques and effective denoising strategies, similar to those used in Vision Mamba. By focusing on preserving key details while reducing noise, I aim to produce clearer and more insightful imagery. This research is crucial for environmental monitoring, providing accurate SAR data that supports critical insights in various fields, including urban development and natural disaster management. I am particularly interested in integrating these technologies into real-world applications to promote better decision-making and enable sustainable planning efforts.

Super-Resolution for SAR Images with Speckle Noise Based on Deep Learning April 2024 – present

• Enhance the clarity of the SAR images using Vision Mamba-based model
• Denoise the SAR images for better interpretation of real-world scenarios
• Improvement in current deep learning architectures and reduce computational costs

LiDAR Detection of Road Guard Rails September 2023 – April 2024 

• Denoise the raw 3D point cloud data using traditional and machine learning outlier removal methods
• Segment and classify the 3D point cloud into groups or categories based on their similar attributes
• Align the 3D point cloud with real-world coordinates to enable safe autonomous driving

Boot Camp Projects 2021 – 2022 

• Innovate new sports for seniors and retirement housing plans using business models
• Promote skills in business selling
• Workshops to enhance presentation and communication skills