| | |
Summary: A REAL-TIME ADAPTIVE THRESHOLDING FOR VIDEO CHANGE DETECTION
Chang Su and Aishy Amer
Concordia University, Electrical and Computer Engineering, MontrŽeal, QuŽebec, Canada
Email: {chan su, amer}@ece.concordia.ca
ABSTRACT
A real-time adaptive non-parametric thresholding algorithm
for change detection is proposed in this paper. Based on the
estimation of the scatter of regions of change in a difference
image, a threshold of each image block is computed discrimi-
natively, then the global threshold is obtained by averaging all
the thresholds for image blocks. The block threshold is calcu-
lated differently for regions of change and background. Ex-
perimental results show the proposed thresholding algorithm
performs well for change detection with high efficiency.
Index Terms-- Image processing, object detection, im-
age segmentation
1. INTRODUCTION
Change detection is widely used in video processing and anal-
ysis [1, 2, 3]. Image differencing followed by thresholding is
a popular method for change detection [1, 2, 4, 5, 6]. Thresh-
|
