A multi-object segmentation algorithm based on background modeling and region growing

Authors:
Kun Zhang;Cuirong Wang;Baoyan Wang
Affiliations:
Northeastern University at Qinhuangdao, Hebei, China;Northeastern University at Qinhuangdao, Hebei, China;Northeastern University at Qinhuangdao, Hebei, China
Venue:
ISNN'12 Proceedings of the 9th international conference on Advances in Neural Networks - Volume Part I
Year:
2012

Citing 4
Cited 0

Erratum: A multi-Kalman filtering approach for video tracking of human-delineated objects in cluttered environments

Computer Vision and Image Understanding
Multivariate image segmentation using semantic region growing with adaptive edge penalty

IEEE Transactions on Image Processing
Multi-object Segmentation in a Projection Plane Using Subtraction Stereo

ICPR '10 Proceedings of the 2010 20th International Conference on Pattern Recognition
A Simple Single Seeded Region Growing Algorithm for Color Image Segmentation using Adaptive Thresholding

CSNT '11 Proceedings of the 2011 International Conference on Communication Systems and Network Technologies

Quantified Score

Hi-index	0.00

Visualization

Abstract

A multi-object segmentation algorithm based on Background Modeling and Region Growing (named as BMRG) algorithm is proposed in this paper. For multi-object segmentation, the algorithm uses Chebyshev inequality and the kernel density estimation method to do background modeling firstly. Then in order to classify image pixels as background points, foreground points and suspicious points, an adaptive threshold algorithm is proposed accordingly. After using background subtraction to get the ideal foreground image, region growing method is used for multi-object segmentation. Here, we improved the region growing method by introducing the growth seed concept for multi-object segmentation, which is calculated from the sparse matrix of quad-tree decomposition. Experimental results show that Chebyshev inequalities can quickly distinguish the foreground and background points. Multi-object segmentation results are satisfactory through seed-based region growing method. Comparison and analysis the experimental results show that the proposed BMRG algorithm is feasible, rapid and effective.