Image sequence segmentation based on 2D temporal entropic thresholding

Authors:
Jianping Fan;Rong Wang;Liming Zhang;Dingjia Xing;Fuxi Gan
Affiliations:
Shanghai Institute of Optics and Fine Mechanics, Academia Sinica, P.O. Box 800-216, Shanghai 201800, PR China;Shanghai Institute of Optics and Fine Mechanics, Academia Sinica, P.O. Box 800-216, Shanghai 201800, PR China;Department of Electronical Engineering, Fudan University, Shanghai 200433, PR China;Department of Computer Science, Fudan University, Shanghai 200433, PR China;Shanghai Institute of Optics and Fine Mechanics, Academia Sinica, P.O. Box 800-216, Shanghai 201800, PR China
Venue:
Pattern Recognition Letters
Year:
1996

Citing 2
Cited 4

Motion compensating field interpolation using a hierarchically structured displacement estimator

Signal Processing
Automatic thresholding of gray-level pictures using two-dimensional entropy

Computer Vision, Graphics, and Image Processing

Accessing Video Contents through Key Objects over IP

Multimedia Tools and Applications
A fast automatic VOP generation using boundary block segmentation

Real-Time Imaging
Length-speed ratio (LSR) as a characteristic for moving elements real-time classification

Real-Time Imaging
Spatio-temporal reasoning for the classification of satellite image time series

Pattern Recognition Letters

Quantified Score

Hi-index	0.10

Visualization

Abstract

This paper reports a novel technique for image sequence segmentation based on maximizing the 2D temporal entropy. The 2D temporal entropy is formed by the scatterplot of frame difference contrast and local variance contrast. Experimental results show that this segmentation-based coding scheme is more efficient than normal fixed-size coding algorithms.