An Improved Optical Flow Algorithm Based on Mask-R-CNN and K-Means for Velocity Calculation
Source abstract
Abstract
Aiming at enhancing the accuracy and reliability of velocity calculation in vision navigation, an improved method is proposed in this paper. The method integrates Mask-R-CNN (Mask Region-based Convolutional Neural Network) and K-Means with the pyramid Lucas Kanade algorithm in order to reduce the harmful effect of moving objects on velocity calculation. Firstly, Mask-R-CNN is used to recognize the objects which have motions relative to the ground and covers them with masks to enhance the similarity between pixels and to reduce the impacts of the noisy moving pixels. Then, the pyramid Lucas Kanade algorithm is used to calculate the optical flow value. Finally, the value is clustered by the K-Means algorithm to abandon the outliers, and vehicle velocity is calculated by the processed optical flow. The prominent advantages of the proposed algorithm are (i) decreasing the bad impacts to velocity calculation, due to the objects which have relative motions; (ii) obtaining the correct optical flow sets and velocity calculation outputs with less fluctuation; and (iii) the applicability enhancement of the optical flow algorithm in complex navigation environment. The proposed algorithm is tested by actual experiments. Results with superior precision and reliability show the feasibility and effectiveness of the proposed method for vehicle velocity calculation in vision navigation system.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Advanced Vision and Imaging · Video Surveillance and Tracking Methods · Autonomous Vehicle Technology and Safety
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Key Laboratory of Instrumentation Science & Dynamic Measurement, Ministry of Education, School of Instrument and Electronics, North University of China, Taiyuan 030051, China
College of Energy & Electric Engineering, Hohai University, Nanjing 210098, China
Funding
National Natural Science Foundation of China · Scientific and Technology Innovation Programs of Higher Education Institutions in Shanxi · Shanxi Province Science Foundation for Youths · Program for the Top Young Academic Leaders of Higher Learning Institutions of Shanxi · Young Academic Leaders Foundation in North University of China · Science Foundation of North University of China · 15th Graduate Science and Technology Project of North University of China · Fund for Shanxi “1331 Project” Key Subjects Construction
References (41)
Show source-provided references
1. Determine turntable coordinate system considering its non-orthogonality10.1063/1.5084799 ↗
2. Robust Optical Flow Integration10.1109/TIP.2014.2336547 ↗
3. Note: Parallel processing algorithm of temperature and noise error for micro-electro-mechanical system gyroscope based on variational mode decomposition and augmented nonlinear differentiator10.1063/1.5037052 ↗
4. Underwater terrain-aided navigation system based on combination matching algorithm10.1016/j.isatra.2017.12.018 ↗
5. Cao, H., Zhang, Y., Shen, C., Liu, Y., and Wang, X. (2018). Temperature Energy Influence Compensation for MEMS Vibration Gyroscope Based on RBF NN-GA-KF Method. Shock Vib.10.1155/2018/2830686 ↗
6. An Unmanned Aerial Vehicle-Based Imaging System for 3D Measurement of Unpaved Road Surface Distresses10.1111/j.1467-8667.2011.00727.x ↗
7. Temperature drift modeling of MEMS gyroscope based on genetic-Elman neural network
8. Robust INS/GPS Sensor Fusion for UAV Localization Using SDRE Nonlinear Filtering10.1109/JSEN.2009.2034730 ↗
9. Effect of Spoofing on Unmanned Aerial Vehicle using Counterfeited GPS Signal10.11003/JPNT.2015.4.2.057 ↗
10. Shen, C., Liu, X., Cao, H., Zhou, Y., Liu, J., Tang, J., Guo, X., Huang, H., and Chen, X. (2019). Brain-like Navigation Scheme based on MEMS-INS and Place Recognition. Appl. Sci., 9.10.3390/app9081708 ↗
11. GPS/INS integration utilizing dynamic neural networks for vehicular navigation10.1016/j.inffus.2010.01.003 ↗
12. Guo, X., Tang, J., Li, J., Shen, C., and Liu, J. (2019). Attitude Measurement based on Imaging Ray Tracking Model and Orthographic Projection with Iteration Algorithm. ISA Trans.10.1016/j.isatra.2019.05.009 ↗
13. A novel hybrid fusion algorithm to bridge the period of GPS outages using low-cost INS10.1016/j.eswa.2013.09.015 ↗
14. A survey on vision-based UAV navigation10.1080/10095020.2017.1420509 ↗
15. Pole-Zero-Temperature Compensation Circuit Design and Experiment for Dual-mass MEMS Gyroscope Bandwidth Expansion10.1109/TMECH.2019.2898098 ↗
16. A temperature and humidity synchronization detection method based on microwave coupled-resonator10.1016/j.snb.2018.01.142 ↗
17. Gibson, J.J. (1950). The Perception of the Visual World, Houghton Mifflin Company. [1st ed.].
18. Visual control orientation behavior in the fly: Part II. Towards underlying neural interactions10.1017/S0033583500002535 ↗
19. Determining optical flow10.1016/0004-3702(81)90024-2 ↗
20. A survey of variational and CNN-based optical flow techniques10.1016/j.image.2018.12.002 ↗
21. A Database and Evaluation Methodology for Optical Flow10.1007/s11263-010-0390-2 ↗
22. Dynamically Removing False Features in Pyramidal Lucas-Kanade Registration10.1109/TIP.2014.2331140 ↗
23. Landing a VTOL Unmanned Aerial Vehicle on a Moving Platform Using Optical Flow10.1109/TRO.2011.2163435 ↗
24. Application of Parameter Optimized Variational Mode Decomposition Method in Fault Diagnosis of Gearbox10.1109/ACCESS.2019.2909300 ↗
25. Large Displacement Optical Flow: Descriptor Matching in Variational Motion Estimation10.1109/TPAMI.2010.143 ↗
26. A Quantitative Analysis of Current Practices in Optical Flow Estimation and the Principles behind Them10.1007/s11263-013-0644-x ↗
27. Forster, C., Pizzoli, M., and Scaramuzza, D. (June, January 31). SVO: Fast semi-direct monocular visual odometry. Proceedings of the 2014 IEEE International Conference on Robotics and Automation (ICRA), Hong Kong, China.10.1109/ICRA.2014.6906584 ↗
28. Traffic flow detection and statistics via improved optical flow and connected region analysis10.1007/s11760-017-1135-2 ↗
29. Long-Term Temporal Convolutions for Action Recognition10.1109/TPAMI.2017.2712608 ↗
30. He, K., Gkioxari, G., Dollar, P., and Girshick, R. (2007, January 17–22). Mask R-CNN. Proceedings of the 2007 IEEE Conference on Computer Vision and Pattern Recognition, Minneapolis, MN, USA.
31. Research and Application of Improved Adaptive MOMEDA Fault Diagnosis Method10.1016/j.measurement.2019.03.033 ↗
32. An efficient line symmetry-based K-means algorithm10.1016/j.patrec.2005.11.006 ↗
33. Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks10.1109/TPAMI.2016.2577031 ↗
34. Lu, S., Ren, C., Zhang, J., Zhai, Q., and Liu, W. (2018). A Novel Approach to Droplet’s 3D Shape Recovery Based on Mask R-CNN and Improved Lambert(-)Phong Model. Micromachines (Basel), 9.10.3390/mi9090462 ↗
35. A novel Fault Diagnosis Method of Gearbox Based on Maximum Kurtosis Spectral Entropy Deconvolution10.1109/ACCESS.2019.2900503 ↗
36. Shen, C., Zhang, Y., Tang, J., Cao, H., and Liu, J. Dual-optimization for a MEMS-INS/GPS System during GPS Outages Based on the Cubature Kalman Filter and Neural Networks. Mech. Syst. Signal Process., In press.
37. Video Face Detection Based on Deep Learning10.1007/s11277-018-5311-7 ↗
38. Wang, Z., Zheng, L., and Wenhua, D. (2019). A novel method for intelligent fault diagnosis of bearing based on capsule neural network. Complexity.10.1155/2019/6943234 ↗
39. A tutorial on spectral clustering10.1007/s11222-007-9033-z ↗
40. Wang, Z., Wang, J., Cai, W., Zhou, J., Du, W., Wang, J., He, G., and He, H. (2019). Application of an Improved Ensemble Local Mean Decomposition Method for Gearbox Composite Fault diagnosis. Complexity.10.1155/2019/1564243 ↗
41. A new camera calibration technique based on circular points10.1109/34.888718 ↗