Enhancing Visual SLAM Localization Accuracy through Dynamic Object Detection and Adaptive Feature Filtering

Zhang Qiang; Wang Tao

doi:10.53941/ijndi.2025.100018

Abstract

To address the critical challenge of localization accuracy degradation in visual Simultaneous Localization and Mapping (SLAM) systems, primarily caused by dynamic feature point interference in complex environments, we propose an advanced visual SLAM framework that integrates deep learningbased object detection with an optimized feature-point filtering strategy. The proposed methodology follows a two-stage processing pipeline. First, a YOLOv5-based object detection module accurately identifies and segments dynamic objects in the operational environment. Second, a probabilistic dynamic feature-point elimination mechanism refines localization precision by selectively retaining reliable static features. To validate the framework, we conduct comprehensive experiments using datasets collected from an Automated Guided Vehicle (AGV) system operating in port environments. Comparative results demonstrate that our approach achieves a 50% improvement in localization accuracy over the conventional ORB-SLAM3 framework in dynamic scenarios. These findings not only confirm the effectiveness of our method but also provide valuable insights for developing robust SLAM systems in real-world engineering applications.

References

1.
Abaspur Kazerouni, I.; Fitzgerald, L.; Dooly, G.; et al. A survey of state-of-the-art on visual SLAM. Expert Syst. Appl., 2022, 205: 117734. doi: 10.1016/j.eswa.2022.117734
2.
Sahili, A.R.; Hassan, S.; Sakhrieh, S.M.; et al. A survey of visual SLAM methods. IEEE Access, 2023, 11: 139643−139677. doi: 10. 1109/ACCESS.2023.3341489
3.
Yan, X.Q.; Mao, Y.Q.; Ye, Y.D.; et al. Cross-modal clustering with deep correlated information bottleneck method. IEEE Trans. Neural Netw. Learn. Syst., 2024, 35: 13508−13522. doi: 10.1109/TNNLS.2023.3269789
4.
Xie, T.; Sun, Q.H.; Sun, T.; et al. DVDS: A deep visual dynamic slam system. Expert Syst. Appl., 2025, 260: 125438. doi: 10.1016/j. eswa.2024.125438
5.
He, J.M.; Li, M.R.; Wang, Y.Y.; et al. OVD-SLAM: An online visual SLAM for dynamic environments. IEEE Sens. J., 2023, 23: 13210−13219. doi: 10.1109/JSEN.2023.3270534
6.
Hu, W.; Gao, X.; Cheung, G.; et al. Feature graph learning for 3D point cloud denoising. IEEE Trans. Signal Process., 2020, 68: 2841−2856. doi: 10.1109/TSP.2020.2978617
7.
Wen, S.H.; Li, X.F.; Liu, X.; et al. Dynamic SLAM: A visual SLAM in outdoor dynamic scenes. IEEE Trans. Instrum. Meas., 2023, 72: 5028911. doi: 10.1109/TIM.2023.3317378
8.
Sun, H.L.; Fan, Q.W.; Zhang, H.Q.; et al. A real-time visual SLAM based on semantic information and geometric information in dynamic environment. J. Real-Time Image Process., 2024, 21: 169. doi: 10.1007/s11554-024-01527-4
9.
Yi, J.Z.; Chen, A.B.; Cai, Z.X.; et al. Facial expression recognition of intercepted video sequences based on feature point movement trend and feature block texture variation. Appl. Soft Comput., 2019, 82: 105540. doi: 10.1016/j.asoc.2019.105540
10.
Lv, C.L.; Lin, W.S.; Zhao, B.Q. Intrinsic and isotropic resampling for 3D point clouds. IEEE Trans. Pattern Anal. Mach. Intell., 2022, 45: 3274−3291. doi: 10.1109/TPAMI.2022.3185644
11.
Lee, O.; Joo, K.; Sim, J.Y. Learning-based reflection-aware virtual point removal for large-scale 3D point clouds. IEEE Robot. Autom. Lett., 2023, 8: 8510−8517. doi: 10.1109/LRA.2023.3329365
12.
Wang, J.W.; Zhuang, Y.; Liu, Y.S. FSS-NET: A fast search structure for 3D point clouds in deep learning. Int. J. Netw. Dyn. Intell., 2023, 2: 100005. doi: 10.53941/ijndi.2023.100005
13.
Li, R.H.; Wang, S.; Gu, D.B. Ongoing evolution of visual SLAM from geometry to deep learning: Challenges and opportunities. Cognit. Comput., 2018, 10: 875−889. doi: 10.1007/s12559-018-9591-8
14.
Qin, Y.; Yu, H.D. A review of visual SLAM with dynamic objects. Ind. Rob., 2023, 50: 1000−1010. doi: 10.1108/IR-07-2023-0162
15.
Li, W.Y.; Yang, F.W. Information fusion over network dynamics with unknown correlations: An overview. Int. J. Netw. Dyn. Intell., 2023, 2: 100003. doi: 10.53941/ijndi0201003
16.
Liu, F.Y.; Cao, Y.; Cheng, X.H.; et al. A visual SLAM method assisted by IMU and deep learning in indoor dynamic blurred scenes. Meas. Sci. Technol., 2024, 35: 025105. doi: 10.1088/1361-6501/ad03b9
17.
Zhang, X.Y.; Abd Rahman, A.H.; Qamar, F. Semantic visual simultaneous localization and mapping (SLAM) using deep learning for dynamic scenes. PeerJ Comput. Sci., 2023, 9: e1628. doi: 10.7717/peerj-cs.1628
18.
Wu, W.X.; Guo, L.; Gao, H.L.; et al. YOLO-SLAM: A semantic SLAM system towards dynamic environment with geometric constraint. Neural Comput. Appl., 2022, 34: 6011−6026. doi: 10.1007/s00521-021-06764-3
19.
Pu, H.Y.; Luo, J.; Wang, G.; et al. Visual SLAM integration with semantic segmentation and deep learning: A review. IEEE Sen. J., 2023, 23: 22119−22138. doi: 10.1109/JSEN.2023.3306371
20.
Iqra; Giri, K.J. SO-YOLOV8: A novel deep learning-based approach for small object detection with yolo beyond coco. Expert Syst. Appl., 2025, 280: 127447. doi: 10.1016/j.eswa.2025.127447
21.
Yuan, Y.J.; Li, Y.; Fang, X.X.; et al. Automatic velocity picking based on improved mask R-CNN. IEEE Trans. Geosci. Remote Sens., 2023, 61: 5923312. doi: 10.1109/TGRS.2023.3335250
22.
He, K.M.; Gkioxari, G.; Dollár, P.; et al. Mask R-CNN. In Proceedings of2017 IEEE International Conference on Computer Vision, Venice, Italy, 22—29 October 2017; IEEE: New York, 2017, pp. 2980–2988. doi:10.1109/ICCV.2017.322
23.
Guo, C.S.; Cai, M.; Ying, N.; et al. ANMS: Attention-based non-maximum suppression. Multimed. Tools Appl., 2022, 81: 11205−11219. doi: 10.1007/s11042-022- 12142-5
24.
Tang, X. S.; Xie, X.L.; Hao, K.R.; et al. A line-segment-based non-maximum suppression method for accurate object detection. Knowl.-Based Syst., 2022, 251: 108885. doi: 10.1016/j.knosys.2022.108885

Scilight Press

Author Information

Abstract

Keywords

References

About Scilight

Journals

Publishing Policies

Contact Us