ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Download
Publications Copernicus
Download
Citation
Articles | Volume X-3/W2-2022
ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci., X-3/W2-2022, 17–22, 2022
https://doi.org/10.5194/isprs-annals-X-3-W2-2022-17-2022
ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci., X-3/W2-2022, 17–22, 2022
https://doi.org/10.5194/isprs-annals-X-3-W2-2022-17-2022
 
27 Oct 2022
27 Oct 2022

A NOVEL REMOTE SENSING IMAGE REGISTRATION ALGORITHM BASED ON THE ADAPTIVE PCNN SEGMENTATION

J. F. Ge1,2, Y. S. Zhang1, X. J. Li1,2,3, H. Li1, and Y. K. Li1,2,3 J. F. Ge et al.
  • 1Faculty of Geomatics, Lanzhou Jiaotong University, Lanzhou, China
  • 2National-Local Joint Engineering Research Center of Technologies and Applications for National Geographic State Monitoring, Lanzhou, China
  • 3Gansu Provincial Engineering Laboratory for National Geographic State Monitoring, Lanzhou, China

Keywords: Image Registration, Slime Mould Algorithm, PCNN, Remote Sensing Image, Segmentation, Feature Detection

Abstract. The appropriate feature segmentation can further improve the remote sensing image registration. The paper proposes a novel adaptive region-based registration method for remote sensing image, which combines the PCNN segmentation and feature-based method. Specifically, the parameters of PCNN are adaptively optimized by the slime mould algorithm. The reference and the input image are matched by the similar regions of PCNN segmentation, which is insensitive to the geometric and photometric changes. Then, two images are registered by the regional matching. Since the segmentation regions of the PCNN agree with the human visual system, and more stable. The proposed method achieves better registration performance. Experimental results conducted on UAV and GaoFen-2 remote sensing image pairs indicate that the proposed method outperforms the SIFT, SURF, Harris-Laplace, MSER methods.