Image Processing Method for Underwater SLAM
Project Description
SLAM is of utmost importance in underwater environments as it enables autonomous navigation, mapping, and localization for underwater vehicles, facilitating tasks such as exploration, environmental monitoring, infrastructure inspection, and scientific research. Limited by the underwater environment, the images acquired by the camera are usually blurry and distorted. This project aims to develop an underwater image processing method for underwater robot navigation and mapping.
Supervisor
YEUNG, Sai Kit
Quota
5
Course type
UROP1000
UROP1100
UROP2100
UROP3100
UROP3200
UROP4100
Applicant's Roles
Learn and implement image processing methods. Apply traditional methods to the underwater field.
Applicant's Learning Objectives
Students will learn image processing methods for underwater SLAM, such as Generative Adversarial Network (GAN), Intrinsic Image Decomposition, Low-Light Image Enhancement and so on.
Complexity of the project
Moderate