DocumentCode :
423824
Title :
Auto method for ambient light independent panorama mosaics
Author :
Xiao, Fu ; Wu, Hui-Zhong ; Xiao, Liang ; Tang, Yang ; Ma, Wen-Jun
Author_Institution :
Dept. of Comput. Sci. &Technology, NanJing Univ. of Sci. & Technol., China
Volume :
6
fYear :
2004
fDate :
26-29 Aug. 2004
Firstpage :
3851
Abstract :
Image mosaic is one of the key elements for panorama construction. Existing mosaic methods generally assume that mosaic pieces are captured with the same illumination. In this paper, we present a novel method for image mosaics with notable illumination difference. It first uses statistical parameters to adjust the illumination of one source image to another, and then implements image registration by extracting feature points and drawing characteristic curves for matching. Thus a solution is found to solve illumination difference, accurate matching, noise points disturbance - key problems in image mosaics. Experimental results are given in the end to demonstrate the effectiveness of this method in image mosaics with notable illumination difference.
Keywords :
feature extraction; image registration; image segmentation; lighting; ambient light independent panorama mosaic; feature extraction; illumination; image mosaic; image registration; panorama construction; statistical parameter; Application software; Cameras; Computer graphics; Computer science; Image sequences; Layout; Lighting; Pixel; Robot vision systems; Virtual reality;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Machine Learning and Cybernetics, 2004. Proceedings of 2004 International Conference on
Print_ISBN :
0-7803-8403-2
Type :
conf
DOI :
10.1109/ICMLC.2004.1380509
Filename :
1380509
Link To Document :
بازگشت