A feature-based algorithm for image mosaics with moving objects

A feature-based algorithm for image mosaics with moving objects is proposed. Firstly, the algorithm eliminates feature points which are selected on moving objects during similarity transform, then outliers are eliminated using RANSAC algorithm. In the image blend phase, difference image between two...

Full description

Saved in:
Bibliographic Details
Published in2010 IEEE International Conference on Intelligent Computing and Intelligent Systems Vol. 1; pp. 26 - 29
Main Authors Zhang Dakun, Li Zhaoxin, Jiang Guiyuan
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.10.2010
Subjects
Computer vision
Difference Image
Feature extraction
Image Mosaics
Image segmentation
Moving Objects
Pattern recognition
Prediction algorithms
Random variables
Region Growth
Similarity Transform
Transforms
Online AccessGet full text
ISBN9781424465828
1424465826
DOI10.1109/ICICISYS.2010.5658658

Cover

More Information
Summary:A feature-based algorithm for image mosaics with moving objects is proposed. Firstly, the algorithm eliminates feature points which are selected on moving objects during similarity transform, then outliers are eliminated using RANSAC algorithm. In the image blend phase, difference image between two images is created using homogeneous transformation matrix. Then region growth algorithm is used to segment the moving object region. Finally, a piecewise mapping algorithm for generating panoramic images is utilized. Experimental results show the effectiveness of the algorithm.
ISBN:9781424465828
1424465826
DOI:10.1109/ICICISYS.2010.5658658