| 异分辨率高光谱图像的空间配准 |
| 投稿时间：2018-04-02 修订日期：2018-04-09 点此下载全文 |
| 引用本文：汪雪洁,王跃明,王磊.异分辨率高光谱图像的空间配准[J].红外,2018,39(7):14~18 |
| 摘要点击次数: 593 |
| 全文下载次数: 358 |
|
| 基金项目:国家自然科学基金项目（51371007）；南京“321”计划项目（2013B01008） |
|
| 中文摘要:近年来，高光谱成像仪迅速发展，可应用于越来越多更复杂的应用场景。人们对遥感图像处理精度等方面提出了更高的要求。图像配准作为遥感图像处理的重要步骤，对遥感图像的后续处理至关重要。遥感图像的配准方式很多，目前比较成熟的配准方式大多集中在不同传感器、相同分辨率下的配准；对于不同传感器、不同分辨率的遥感图像配准成果很少，所以本文对此进行了研究。提出了基于边缘特征寻找同名点的遥感图像配准算法，得到了图像配准结果，并对其进行了验证。相对于现有的配准算法，本文算法使得不同传感器、不同分辨率的遥感图像的配准精度大大提高。 |
| 中文关键词:图像配准 边缘检测 连通域 仿射变换 |
| |
| Registration of Hyperspectral Images with Different Resolution |
|
|
| Abstract:In recent years, hyperspectral imagers have developed rapidly. With their use in more and more complicated applications, higher requirements are put forward for the processing precision of remote sensing images. As an important step in the processing of remote sensing images, image registration is crucial to the subsequent processing of remote sensing images. There are many registration methods for remote sensing images. At present, most mature registration methods are those methods which conduct registration in the case of different sensors and same resolution. There are few registration methods for the remote sensing images in the case of different sensors and different resolution. A new registration algorithm for extracting point features on the basis of image edge features is proposed. Image registration results are obtained and validated. Compared with the existing registration algorithms, the registration precision of the algorithm is improved greatly. |
| keywords:image registration edge detection connected domain affine transformation |
| 查看全文 HTML 查看/发表评论 下载PDF阅读器 |