Discipline: Computer Sciences and Information Management
Subcategory: Computer Science & Information Systems
Tori Wilbon - Elizabeth City State University
Co-Author(s): Omar Owens, Winston Salem State University, Winston Salem, NC
The focus of this project is to explore learning algorithms for automatically detecting layer boundaries from images collected from the Polar Regions, specifically Antarctica images. Layer boundaries are incorporated into climate models for forecasting a rise in sea level but are difficult to extract from noisy images. Currently, glaciologists manually identify layer boundaries, which is time-consuming and requires sparse hand selection. An active contours model will be explored for detecting layer boundaries. An active contours model (snake) is used in computer vision for object tracking, shape recognition, segmentation, and edge detection. The snake algorithm is an energy optimization spline guided by an outside constraint force and influenced by image forces, which would pull it toward layer boundaries.
Funder Acknowledgement(s): Indiana University-Summer Research Opportunities in Computing (IU-SROC) Dr. Geoffrey Fox
Faculty Advisor: Jerome Mitchell,