Technology of UAV remote sensing image mosaic technology combining SIFT features and poisson fusion

doi:10.13474/j.cnki.11-2246.2021.0017

Abstract

Abstract: Aiming at the research of UAV remote sensing image mosaic technology, a seamless mosaic method of UAV remote sensing image combining scale-invariant SIFT feature points with overlapping transition Poisson fusion is proposed. The method firstly uses SIFT algorithm to extract the feature points of the image. the feature points are coarsely matched according to the euclidean distance between the feature descriptors. Then the random sampling consistency (RANSAC) algorithm is used to remove the mismatched point pairs. In order to prevent the calculated single adaptive matrix linear structure instability, the LM (Levenbeg-Marquardt) algorithm is introduced to optimize the single adaptive matrix. Finally, the overlapped transition poisson fusion algorithm is used to achieve the seamless mosaic of the image. The experimental results show that this method has advantages in UAV remote sensing image mosaic and can obtain good mosaic image.

Key words: SIFT feature, RANSAC algorithm, LM algorithm, Poisson fusion, image mosaic

CLC Number:

P237

ZHOU Jie, XIE Kun, FU Chao, SHI Kun. Technology of UAV remote sensing image mosaic technology combining SIFT features and poisson fusion[J]. Bulletin of Surveying and Mapping, 2021, 0(1): 94-98.

References

[1] 赵业婷. 电力巡线单航带无人机影像拼接关键技术[D].成都:成都理工大学,2017.
[2] PELEG S, ROUSSO B, RAV-ACH A A, et al. Mosaicing on adaptive manifolds[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2000, 22(10):1144-1154.
[3] KIM D W, HONG K S. Real-time mosaic using sequ ential graph[J]. Journal of Electronic Imaging, 2006, 15(2):023005-023013.
[4] 何敬,李永树. 基于特征点和最优路径的无人机影像拼接方法[J].遥感技术与应用,2012,27(2):173-176,219.
[5] 岳春宇,江万寿. 几何约束和改进SIFT的SAR影像和光学影像自动配准方法[J].测绘学报,2012,41(4):570-576.
[6] LOWE D G. Object recognition from local scale-invariant features[C]//Proceedings of the 7th IEEE International Conference on Computer Vision.[S.l.]:IEEE, 1999.
[7] LOWE D G. Distinctive image features from scale-invariant keypoints[J].International Journal of Computer Vision,2004,60(2):91-110.
[8] 宋佳乾,汪西原. 改进SIFT算法和NSCT相结合的遥感图像匹配算法[J].测绘通报,2018(9):34-38.
[9] 李孚煜,叶发茂. 基于SIFT的遥感图像配准技术综述[J].国土资源遥感,2016,28(2):14-20.
[10] 索效荣,齐苑辰. 基于SIFT算法的无人机航空遥感影像匹配[J].测绘科学,2012,37(1):89-91.
[11] 许志方,王双亭,王春阳,等. 尺度不变特征变换的UHD185高光谱影像拼接[J].遥感信息,2017,32(1):95-99.
[12] 徐志远,刘嘉文,王山东,等. 低空无人机影像的快速拼接技术研究[J].地理空间信息,2016,14(4):39-41.
[13] 张彪,邹哲,陈书界,等.基于仿射变换与Levenberg-Marquardt算法的织物图像配准[J].光学学报,2017,37(1):351-359.
[14] 赵书睿. 全景图像拼接关键技术研究[D].成都:电子科技大学,2013.
[15] 张小利,李雄飞,李军. 融合图像质量评价指标的相关性分析及性能评估[J].自动化学报,2014,40(2):306-315.