Previous [ 1] [ 2] [ 3] [ 4] [ 5] [ 6] [ 7] [ 8] [ 9] [ 10] [ 11] [ 12] [ 13] [ 14] [ 15] [ 16] [ 17] [ 18] [ 19]

¡@

Journal of Information Science and Engineering, Vol. 25 No. 3, pp. 945-959 (May 2009)

Analysis and Comparison of Various Shape Mirrors Used in Omnidirectional Cameras with Parameter Variation*

Kuo-Yang Tu
Institute of System Information and Control
National Kaohsiung First University of Science and Technology
Kaohsiung, 811 Taiwan

The studies on omnidirectional cameras usually focus on design and/or development of the complicated structure for a wide range of view. However, difficult calibration and expensive cost arisen from the complicated structure of omnidirectional cameras are considerable. In this paper, imprecise parameters resulted from missing calibration of the reflective mirrors of various profiles in omnidirectional cameras are studied. The relationship between an object surface and an image plane, related to camera parameters, is thus derived. In addition, an algorithm of more perfect unwarped images in omnidirectional cameras is proposed. In practical, it is difficult and costly to make an omnidirectional camera consisted of the reflective mirrors of various profiles. Therefore, a simulator based on the derived relationship is designed to compare the omnidirectional camera of various profile mirrors influenced by parameter variations. There are five kinds of reflective mirrors with distinct profiles, spherical, parabolic, conical, tangent and sine, respectively, compared in same size. In addition, the profile to compare the variation of these five mirrors is included.

Keywords: catadioptric sensors, parameter calibration, mirror profiles, unwarped images, image processing

Full Text (¥þ¤åÀÉ) Retrieve PDF document (200905_19.pdf)

Received June 27, 2007; revised April 2 & June 16 & August 26, 2008; accepted September 11, 2008.
Communicated by Pau-Choo Chung.
* This paper was partially supported by the National Science Council of Taiwan, R.O.C. under grants No. NSC 94-2213-E-327-014-.