25 October 2016 Infrared image mosaic using point feature operators
Author Affiliations +
Proceedings Volume 10157, Infrared Technology and Applications, and Robot Sensing and Advanced Control; 101570L (2016) https://doi.org/10.1117/12.2244493
Event: International Symposium on Optoelectronic Technology and Application 2016, 2016, Beijing, China
In this paper, we study infrared image mosaic around a single point of rotation, aiming at expanding the narrow view range of infrared images. We propose an infrared image mosaic method using point feature operators including image registration and image synthesis. Traditional mosaic algorithms usually use global image registration methods to extract the feature points in the global image, which cost too much time as well as considerable matching errors. To address this issue, we first roughly calculate the image shift amount using phase correlation and determine the overlap region between images, and then extract image features in overlap region, which shortens the registration time and increases the quality of feature points. We improve the traditional algorithm through increasing constraints of point matching based on prior knowledge of image shift amount based on which the weighted map is computed using fade in-out method. The experimental results verify that the proposed method has better real time performance and robustness.
© (2016) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Zhen Huang, Zhen Huang, Shaoyuan Sun, Shaoyuan Sun, Zhenyi Shen, Zhenyi Shen, Junjie Hou, Junjie Hou, Haitao Zhao, Haitao Zhao, } "Infrared image mosaic using point feature operators", Proc. SPIE 10157, Infrared Technology and Applications, and Robot Sensing and Advanced Control, 101570L (25 October 2016); doi: 10.1117/12.2244493; https://doi.org/10.1117/12.2244493

Back to Top