Affiliation:
1. Tianjin University of Technology
Abstract
In this paper, feature extraction algorithm based on spherical perspective projection model for the matching fisheye image is proposed. The fisheye image is mapped to the image plane through spherical mapping. Then the diffusion equation is formed by convolution of the image projection and spherical Gaussian function. The feature points of image are extracted based on the SIFT at the scale of spherical correlation function. Compared with SURF(Speeded Up Robust Features), more feature points in a shorter time are obtained.
Publisher
Trans Tech Publications, Ltd.
Reference8 articles.
1. L. Meinel, M. Findeisen, M. Hes, A. Apitzsch and G. Hirtz: Automated real-time surveillance for ambient assisted living using an omnidirectional camera, 2014 IEEE International Conference on in Consumer Electronics (ICCE) (2014), pp.396-399.
2. J. P. Barreto, and D. Kostas: Fundamental matrix for cameras with radial distortion, Tenth IEEE International Conference on Computer Vision, Vol. 1, (2005).
3. X. Qin, S. Li: Finding scale-invariant corner feature from full-view image based on discrete spherical model, 2012 International Conference on Systems and Informatics (ICSAI), (2012).
4. In Chinese, J. Zhu, L. Ge, F. Han, N. Liu and B. Zhang: Research on Fisheye Lens Imaging Model Based on the Separated Parameters Calibration. Chinese Journal of Sensors and Actuators, Vol. 7, no. 015 (2013).
5. M. M. Fleck: Perspective projection: the wrong imaging model, Department of Computer Science, University of Iowa, (1995) pp.1-27.