Fig 1.
Progression of scan matching methods from traditional to proposed approach.
Fig 2.
Architecture of the proposed RANDT method: (a) Extended Normal Distribution Transform Algorithm with Outlier Removal (b) Incremental Scan Matching Framework using Extended Normal Distribution Transform for Point Density Uniformity.
Fig 3.
The effect of outlier elimination in point cloud (a) before outlier removal, (b) after outlier removal.
Fig 4.
Overall comparison of proposed algorithm on KITTI outdoor scene registration results with ML-NDT, One point RANSAC, and DeepSIR algorithms.
Fig 5.
Object-level registration on ModelNet40: Comparison of proposed method Classical NDT,ML-NDT,Piecewise ICP.
Table 1.
Performance comparison of registration methods on KITTI dataset.
Table 2.
Performance comparison of registration methods on ModelNet40 dataset.
Fig 6.
Convergance comparison of registration methods.
Fig 7.
Execution time comparison across registration methods.
Table 3.
The standard deviation of point density for the KITTI and ModelNet40 datasets.
Fig 8.
Comparison of point density uniformity across methods.
Fig 9.
Relationship between Point Density Uniformity (PDU) and registration accuracy.
Table 4.
Robustness evaluation under noise, outliers, and partial overlap.
Fig 10.
Robustness analysis under noise, outliers, and partial overlap.
Fig 11.
Comparison of registration accuracy before (a) and after(b) applying the proposed outlier elimination module within the RANDT framework.