Research on UAV Image Registration Based on Sift Algorithm Acceleration

Considering the low real-time performance and the large amount of false matches exist in the feature matching stage of traditional Scale Invariant Feature Transform(SIFT) algorithm in unmanned aerial vehicle (UAV) remote sensing image registration. In this paper, a series of optimization methods for traditional SIFT algorithms are proposed, including SIFT execution process optimization, changing the parameters of scale space construction, Simplified method for judging the construction area of feature descriptor and construction of bidirectional matching filters. Experiments on UAV remote sensing images show that the optimization method can significantly improve the matching efficiency compared with traditional methods, and the comprehensive acceleration ratio is about 35% to 40%, which proves the effectiveness of the acceleration method.

[1]  Jan-Michael Frahm,et al.  USAC: A Universal Framework for Random Sample Consensus , 2013, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[2]  Xiaorun Li,et al.  Medium-low resolution multisource remote sensing image registration based on SIFT and robust regional mutual information , 2018 .

[3]  Paolo Pirjanian,et al.  SIFT-ing through features with ViPR , 2006, IEEE Robotics & Automation Magazine.

[4]  Yan Zhang,et al.  A SIFT Algorithm Based on DOG Operator , 2018, 2018 International Conference on Intelligent Transportation, Big Data & Smart City (ICITBS).

[5]  Cordelia Schmid,et al.  A Performance Evaluation of Local Descriptors , 2005, IEEE Trans. Pattern Anal. Mach. Intell..

[6]  Yaxin Bi,et al.  KNN Model-Based Approach in Classification , 2003, OTM.

[7]  Luo Juan,et al.  A comparison of SIFT, PCA-SIFT and SURF , 2009 .

[8]  David G. Lowe,et al.  Distinctive Image Features from Scale-Invariant Keypoints , 2004, International Journal of Computer Vision.

[9]  S. Govindarajulu,et al.  A Comparison of SIFT, PCA-SIFT and SURF , 2012 .

[10]  Matthijs C. Dorst.  Distinctive Image Features from Scale-Invariant Keypoints , 2011 .

[11]  José-Emilio Guerrero-Ginel,et al.  SIFT optimization and automation for matching images from multiple temporal sources , 2017, Int. J. Appl. Earth Obs. Geoinformation.