| 基于CSIFT算法的UAV影像配准技术研究 |
| |
| 引用本文： | 朱佳文,屈新原,高小伟.基于CSIFT算法的UAV影像配准技术研究[J].首都师范大学学报(自然科学版),2013,34(4):69-74. |
| |
| 作者姓名： | 朱佳文 屈新原 高小伟 |
| |
| 作者单位： | 1. 首都师范大学资源环境与旅游学院,北京100048;城市环境过程与数字模拟国家重点实验室培育基地,北京100048;三维信息获取与应用教育部重点实验室,北京100048;资源环境与地理信息系统北京市重点实验室,北京 100048
2. 北京硅盾安全技术有限公司,北京10048 |
| |
| 摘 要： | 无人机(Unmanned Aerial Vehicle,UAV)影像配准作为无人机数据处理的必要环节,一直是业界研究的关键技术.在传统配准中,由于只对灰度影像进行处理难免会产生特征混淆、误匹配等问题.本文分析了UAV影像的成像特点,采用CSIFT(Colored scale invariant feature transform)算法对UAV影像进行配准处理,并结合分块策略对配准结果进行优化.实验证明,CSIFT算法比传统SIFT(Scale invariant feature transform)算法在提取困难特征点时有非常明显的优势,尤其是在影像有较大倾角或较多水体等情况下,可以有效提高UAV影像配准效率.
|
| 关 键 词： | UAV 影像配准 CSIFT 分块策略 |
Study on UAV Image Registration Technique Based on CSIFT |
| |
| Zhu Jiawen
,
Qu Xinyuan
,
Gao Xiaowei.Study on UAV Image Registration Technique Based on CSIFT[J].Journal of Capital Normal University(Natural Science Edition),2013,34(4):69-74. |
| |
| Authors: | Zhu Jiawen Qu Xinyuan Gao Xiaowei |
| |
| Institution: | 1. College of Resource Environment and Tourism, Capital Normal University, Beijing 100048, China; 2. Key Laboratory of Resources Envknnmcnt and GIS of Beijing Municipal, Bcijing 100048, Chlna; 3. Key Lab of 3D Information Acquisition and Application, Ministry of Education, Beijing 100048, China; 4. State Key Laboratory Incubation Base of Urban Environmental Processes and Digital Simulation, Bcijing 10004g, China; 5. Beijing SiShield Security Co. Ltd, Beijing 100048,China) |
| |
| Abstract: | UAV （Unmanned Aerial Vehicle） image registration as the essential link of UAV data processing, has been the key technology in the industry research. In the traditional registration, the gray image processing will inevitably produce characteristics of confusion and mismatch problems. This paper analyzes the UAV image imaging features, using the CSIFT （Colored scale invariant feature transform） algorithm for UAV image registration process, and combined with Image strategy of block match to get higher match efficiency. Experiments prove that CSIFT has more obvious advantages in more difficult to extract feature points in the image, such as Camera shot in the angled and large areas of water body in the Image,which can improve the efficiency of UAV image registration effectively. |
| |
| Keywords: | Unmanned Aerial Vehicle Image registration CSIFT Image strategy of block match |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
