Title :
A robust image registration algorithm used for panoramic image mosaic
Author :
Gao Bo ; Jianzhong Cao ; Zuofeng Zhou ; Enshi Qu ; Wang Hua ; Huinan Guo
Author_Institution :
Xi´an Inst. of Opt. & Precision Mech., Xi´an, China
Abstract :
The panoramic image has been widely used in social production, and has become an important topic in research on the field of image processing. For complex images with multiple scenes and other elements, the algorithm that is based on Harris cannot make image registration effectively. This paper proposes a method that combines Harris with SIFT, using the Harris algorithm with adaptive threshold to extract the corners and the SIFT descriptor to make the registration. It improved registration efficiency sub-effectively for the image in complex scenes, and generated a 360-degree panoramic image quickly. The experiments showed that the algorithm is adaptable and robust.
Keywords :
feature extraction; image registration; image segmentation; Harris algorithm; SIFT descriptor; corner extraction; image processing; image registration algorithm; panoramic image mosaic; scale-invariant feature transform; Algorithm design and analysis; Feature extraction; Histograms; Image registration; Robustness; Signal processing algorithms; Vectors; Adaptive Harris corners; SIFT descriptor; cylindrical projection; panoramic images;
Conference_Titel :
Image Analysis and Signal Processing (IASP), 2012 International Conference on
Conference_Location :
Hangzhou
Print_ISBN :
978-1-4673-2547-9
DOI :
10.1109/IASP.2012.6425046