一种基于局部灰度匹配的无人机图像拼接算法

Unmanned aerial vehicle image mosaic algorithm based on local gray fitting

结合无人机图像的特点,开发了一种基于局部灰度匹配的图像拼接算法。根据无人机的飞行数据,对无人机图像进行预处理,缩小了特征点的搜索范围。然后利用图像的局部灰度算法,对提取的角特征点进行快速匹配。最后在运行速度以及特征点的匹配准确率等方面对算法进行了评价。试验结果表明,该算法较大幅度地提高了程序的运行速度,能够达到无人机图像处理的实时性的要求。该算法对无人机图像的工业化处理有帮助。

Considering the feature of unmanned aerial vehicle （UAV） image, an image mosaic algorithm based on local gray fitting method was developed. According to the air data of UAV, UAV image was preconditioned, which can reduce the range for searching feature point. And then, the comer feature points abstracted were matched fast on the basis of square error of the gray value. Last, the algorithm was evaluated by speed and accurate rate. The results obtained from experiments and simulations show that this algorithm is effective and can adapt to the real-time character of UAV image stitching. The algorithm can provide some experience for the UAV image processing industrialization.