• Laser & Optoelectronics Progress
  • Vol. 57, Issue 10, 101509 (2020)
Bin Wu and Xuri Wang*
Author Affiliations
  • State Key Laboratory of Precision Measuring Technology and Instruments, Tianjin University, Tianjin 300072, China
  • show less
    DOI: 10.3788/LOP57.101509 Cite this Article Set citation alerts
    Bin Wu, Xuri Wang. Inertial Navigation Aided Image Feature Matching Method[J]. Laser & Optoelectronics Progress, 2020, 57(10): 101509 Copy Citation Text show less
    Process of three-dimensional rotation. (a) Process of rotation; (b) three-dimensional rotation
    Fig. 1. Process of three-dimensional rotation. (a) Process of rotation; (b) three-dimensional rotation
    Process of coordinate transformation
    Fig. 2. Process of coordinate transformation
    Calculation process of displacement variation in navigation coordinate system
    Fig. 3. Calculation process of displacement variation in navigation coordinate system
    Epipolar geometric constraint
    Fig. 4. Epipolar geometric constraint
    Process of inertial navigation aided image feature matching
    Fig. 5. Process of inertial navigation aided image feature matching
    Experimental equipment
    Fig. 6. Experimental equipment
    Source images. (a) First image of experimental table; (b) second image of experimental table; (c) first image of laboratory; (d) second image of laboratory; (e) first image of office; (f) second image of office
    Fig. 7. Source images. (a) First image of experimental table; (b) second image of experimental table; (c) first image of laboratory; (d) second image of laboratory; (e) first image of office; (f) second image of office
    Brute-force match results of three sets of images
    Fig. 8. Brute-force match results of three sets of images
    Possible areas. (a) Four feature points selected in the first image of experimental table; (b) possible areas corresponding to the feature points in the first image of experimental table; (c) four feature points selected in the first image of laboratory; (d) possible areas corresponding to the feature points in the first image of laboratory; (e) four feature points selected in the first image of office; (f) possible areas corresponding to the feature points in the first image of office
    Fig. 9. Possible areas. (a) Four feature points selected in the first image of experimental table; (b) possible areas corresponding to the feature points in the first image of experimental table; (c) four feature points selected in the first image of laboratory; (d) possible areas corresponding to the feature points in the first image of laboratory; (e) four feature points selected in the first image of office; (f) possible areas corresponding to the feature points in the first image of office
    Match results of three sets of images after adding constraints
    Fig. 10. Match results of three sets of images after adding constraints
    Match results of three sets of images after removing outliers by RANSAC
    Fig. 11. Match results of three sets of images after removing outliers by RANSAC
    SceneRotation matrix
    Experimentaltable0.948665-0.0059610.3162280.0081230.999952-0.005519-0.3161800.0078040.948667
    Laboratory0.9985340.004594-0.0539260.0020840.9923890.1231240.054082-0.1230560.990925
    Office0.996481-0.048260-0.0685320.0616190.9760500.2086390.056821-0.2121280.975589
    Table 1. Rotation matrix
    SceneTranslation vector /mm
    Experimentaltable13.4731210.0045002.186387T
    Laboratory9.200683-4.7711470.862319T
    Office4.8232901.268367-0.356531T
    Table 2. Translation vector
    SceneFundamental matrix
    Experimental table-1.565037×10-9-1.791871×10-71.964372×10-45.191469×10-7-9.682601×10-9-0.004100-5.179737×10-40.004090-0.100741
    Laboratory-2.130416×10-8-2.201791×10-8-1.334185×10-32.979019×10-89.308407×10-8-2.758886×10-31.364791×10-32.548693×10-39.785125×10-1
    Office7.710623×10-96.469725×10-93.591123×10-4-5.158159×10-88.523534×10-8-1.365610×10-3-2.318040×10-41.266611×10-39.907891×10-1
    Table 3. Fundamental matrix
    SceneMatching points of inertial navigationaided image feature matching methodMatching pointsof RANSACAccuracy rate /%
    Experimental table14513492.4
    Laboratory29328396.6
    Office17116194.2
    Table 4. Matching results of three sets of images
    Bin Wu, Xuri Wang. Inertial Navigation Aided Image Feature Matching Method[J]. Laser & Optoelectronics Progress, 2020, 57(10): 101509
    Download Citation