一种不同地面分辨率图像的拼接算法

A mosaic method of different ground resolution images

航空遥感监测中变焦镜头的广泛应用以及飞行高度的变化会获得不同地面分辨率的航空影像,因此为了实时监控到飞行区域的整体情况,需要对不同地面分辨率的图像进行拼接。提出了针对不同地面分辨率图像进行拼接的方法。首先用四分法对图像进行分割,检测变化倍率,选择配准策略;然后用基于加速鲁棒性特征(Speeded-Up Robust Features,SURF)的算法对相邻图像进行拼接。获得了焦距变化从高倍过渡到低倍,或从低倍过渡到高倍的拼接图像。通过航空变焦模拟影像的拼接实验和地面变焦影像的拼接实验表明,该方法能够对不同地面分辨率的图像进行拼接,且拼接结果均以高分辨率为基准。

In the process of airborne remote sensing monitoring,zoom lens is widely used and altitude is frequently changed.The airborne images of different resolutions are got.Therefore,to mosaic the quick views of different resolution for real-time monitoring is needed.A new method to solve the problem is proposed.This method uses quartering theory to split images,detects zooming ratio,and selects registration strategy.Then SURF（Speeded-Up Robust Features）algorithm is applied to mosaic images.It can be used to get a mosaic image under the condition of continuous zooming with focal distance ranging from high power to low power or from low power to high power.The mosaic experiments of zooming airborne images and zooming ground images indicate this new method can be used to mosaic images with different resolutions,and the results are all based on the highest resolution.