中文摘要：

为了及时、准确地反映测区情况，需要在拍摄现场将获得的无人机遥感影像进行实时拼接。基于SURF的无人机遥感影像拼接算法具有尺度和旋转不变性，拼接效果较好，但该算法稳定性较差，拼接过程中大量误特征点被提取出来，导致计算量大、匹配效率低，无法满足实时拼接的要求。该文首先利用SURF算法粗提取特征点，然后利用RANSAC算法剔除被误提取的特征点，实现特征点的快速匹配和图像实时拼接。实验结果表明该文的算法不但可以得到很好的图像拼接效果，而且极大地提高了拼接效率。

英文摘要：

In order to reflect the situation in flying area, the UAV images must be mosaiced real time. The algorithm of SURF （Speed Up Robust Features）has such features as scale invariant and rotate invariant, the mosaicing effect is very good, but the algorithm is unstable and a lot of error feature points are extracted which lead to large amount of calculation and low mosaicing efficiency. To meet the demands of feature points matching quickly and images mosaicing real time, in this paper, the improved algorithm of SURF is proposed, firstly the feature points are detected by SURF algorithm, then the error feature points are elim- inated by RANSAC（Random Sample Consensus） algorithm. The experiment results show that the algorithm proposed in this paper can acquire very good mosaicing effect, and the mosaicing efficiency is very high.