Title

Activity Segmentation of Infrared Images Using Fuzzy Clustering Techniques

Document Type

Conference Proceeding

Publication Date

5-2011

Abstract

Every year, many older adults are at risk for falling, especially in the dark. Infrared lighting provides a nonintrusive lighting in the dark and our research shows a technique of segmenting human activities using fuzzy clustering of image moments even in the dark. While our research is still in the preliminary stages, it shows promise of being able to detect several different activities and in the future might prevent several falls from taking place.

Comments

Presented at the World Conference on Soft Computing, San Francisco, CA, May 23-26, 2011.