A New Approach to Robust Human Motion Detection

Q. Li, J. You (PRC), and P. Bhattacharya (Canada)


Human Motion Detection, Image Understanding, Multi ple Feature Extraction, Image Feature Hierarchy, Fussy Set Theory


This paper presents an image understanding approach to monitor human movement and identify the abnormal cir cumstance by robust motion detection for the care of the elderly in a home-based environment. In contrast to the conventional approaches which apply either a fixed feature extraction scheme or a fixed object model for motion de tection and tracking, we introduce a multiple feature ex traction scheme for robust motion detection. The proposed algorithms include 1) multiple image feature extraction in cluding the detection of interesting points and color clus ters, 2)adaptive thresholding selection based on the com pactness measures of fuzzy sets in image feature space, 3) a flexible model of human motion adapted in both rigid and non-rigid conditions, and 4) an optimized algorithm for ob ject tracking and fuzzy decision making.

Important Links:

Go Back