CAJ | 학술논문

无人机低空拍摄的图像具有成本低、方便快捷的优点，但也存在重叠度高、覆盖范围小等问题，需通过图像拼接来获得更大范围的图像。目前，无人机图像拼接效率不高。针对此问题，提出了一种结合 Harris 与 SURF 算法优点的改进 Harris-SURF 算法，并将该算法应用于无人机图像的自动配准；采用加权平滑融合算法消除亮度差异，较好地实现了无人机图像的无缝拼接。实验结果表明，该算法在效率方面较 SURF 算法有较大提高，对存在旋转、光照及视角变化的图像均有较好的匹配效果，实现了无人机图像的快速拼接。
무인궤저공박섭적도상구유성본저、방편쾌첩적우점，단야존재중첩도고、복개범위소등문제，수통과도상병접래획득경대범위적도상。목전，무인궤도상병접효솔불고。침대차문제，제출료일충결합 Harris 여 SURF 산법우점적개진 Harris-SURF 산법，병장해산법응용우무인궤도상적자동배준；채용가권평활융합산법소제량도차이，교호지실현료무인궤도상적무봉병접。실험결과표명，해산법재효솔방면교 SURF 산법유교대제고，대존재선전、광조급시각변화적도상균유교호적필배효과，실현료무인궤도상적쾌속병접。
UAV images captured in low altitude have the advantages of low cost and convenient , but they also have the problems with the high overlap degree and the small-scale coverage. It needs image mosaic to get the wider range images. Currently , UAV image stitching is not efficient. To solve the problem, this paper proposes an improved Harris-SURF algorithm which combines the advantages of the Harris and the SURF and is applied to automatic registration for UAV images. It uses the weighted smoothing method to eliminate the differences in brightness and implementes the seamless mosaic for UAV images. The experimental results show that the algorithm in terms of efficiency has been improved greatly over the SURF algorithm , has a good match result for the images that exist rotation, illumination and multi-angle differences, and implements rapid stitching for UAV images.