Author Affiliations
1 Zhejiang Provincial Key Laboratory of Information Processing, Communication and Networking, Hangzhou, Zhejiang 310027, China2 College of Information Science & Electronic Engineering, Zhejiang University, Hangzhou, Zhejiang 310027, China;show less
Fig. 1. Framework of system
Fig. 2. Flowchart of optimization algorithm
Fig. 3. Example of stereo matching in KITTI01 sequence. (a) Features in left image; (b) features in right image
Fig. 4. Example of image segmentation in KITTI01 sequence
Fig. 5. Feature plane extracted in an image block
Fig. 6. Example of feature re-matching
Fig. 7. Binocular stereoscopic imaging model
Fig. 8. Scenes of KITTI 01 sequence
Fig. 9. Comparison of the number of local stereo match points in left and right images (a) before and (b) after optimization
Fig. 10. Number of stereo matching points in KITTI 01 sequence
Fig. 11. Comparison of error of KITTI 01 sequence
Fig. 12. Accuracy comparison of different algorithms in KITTI 11 sequence
Fig. 13. (a) Electric hunting cart; (b) BumbleBee2 binocular camera
Fig. 14. Part of the experimental scene. (a) Weak texture environment; (b) moving pedestrians and vehicles; (c) shadow, exposure and other light changes
Fig. 15. Results of actual scene experiment
Method | Average translation error /% | Average rotation error /[(°)/m] |
---|
SOFT2 | 0.66 | 0.0014 | LG_SLAM | 0.82 | 0.0020 | RotRocc+ | 0.83 | 0.0026 | FPVO | 1.10 | 0.0023 | ORB_SLAM2 | 1.15 | 0.0027 | S-PTAM | 1.19 | 0.0025 |
|
Table 1. Precision comparison of different algorithms in KITTI benchmark
Operation | t /s |
---|
Feature extraction and matching | 0.0390 | Optimized by local feature | 0.0052 | Feature tracking and pose optimization | 0.0380 | Total | 0.0820 |
|
Table 2. System runtime statistics