You have requested a machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Neither SPIE nor the owners and publishers of the content make, and they explicitly disclaim, any express or implied representations or warranties of any kind, including, without limitation, representations and warranties as to the functionality of the translation feature or the accuracy or completeness of the translations.
Translations are not retained in our system. Your use of this feature and the translations is subject to all use restrictions contained in the Terms and Conditions of Use of the SPIE website.
14 May 2015The omni-directional image assisted optical surveillance system
This paper presents a high resolution optical surveillance system which integrated an omni-directional imager as an event finder/ system trigger. The omni-directional optics, a fish-eye camera in this study, provides a wider field of view (FOV) which can monitor widely range continuously without scanning mechanism but offers sufficient information which includes sign of field event and direction and then drive high resolution surveillance camera for detail imaging. To archive an optical triggering surveillance system, the scale-invariant feature transform (SIFT) is implemented to detect features both from images taken by omni-directional imager and the high resolution surveillance camera. Considering the FOV of high resolution surveillance system is narrow, to ensure the pointing of high resolution surveillance system, feature matching is also implemented in this system to identify the images obtained by high resolution surveillance system are identical to the existing omni-directional image obtained from fish-eye camera. This provides a robust and accurate solution to the problem of optical radar surveillance system localization in unknown environments. An experiment is performed on outdoor image sequences with demonstrating the efficiency of our algorithm.