| 基于云台标定的快速视频拼接算法 |
| |
| 引用本文： | 谢小竹,崔建峰,吴克寿.基于云台标定的快速视频拼接算法[J].新乡学院学报(自然科学版),2013(2):123-125,135. |
| |
| 作者姓名： | 谢小竹 崔建峰 吴克寿 |
| |
| 作者单位： | 厦门理工学院 计算机与信息工程学院,福建 厦门 364102 |
| |
| 摘 要： | 针对视频监控成本高、性能提升受到限制的问题,提出了基于云台标定的快速视频拼接算法:首先利用相位相关法对云台进行标定,利用标定公式能快速计算出任意两个视频帧之间的粗估平移量,然后,在粗估的重叠区域(ROR)用加速鲁棒特性(SURF,Speeded-Up Robust Features)特征匹配算法进行拼接.从实验结果看,该方法不仅大大提高了视频拼接的效率,而且提高了精准度.
|
| 关 键 词： | 视频拼接 云台标定 粗估重叠区域(ROR) SURF |
Rapid Video Stitching Algorithm Based on Pan-tilt Calibration |
| |
| Authors: | XIE Xiao-zhu CUI Jian-feng WU Ke-shou |
| |
| Institution: | (School of Computer and Information Engineering, Xiamen University of Technology, Xiamen 361024, China) |
| |
| Abstract: | A rapid video stitching algorithm base on Pan-Tilt calibration is proposed in the paper, aiming at problems about high cost of video surveillance and performance improvement restrained. First, phase correlation is used for Pan-Tilt calibration. The rough overlapped region（ROR） can be rapidly computed between any two frames in the video by calibration formula. Then it extracts by SURF（Speeded-Up Robust Features） matching algorithm in ROR. The experimental results show that the approach can not only greatly improve the stitching efficiency but also the precision. |
| |
| Keywords: | video mosaic Pan-Tilt calibration rough overlapping region（ROR） SURF |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
