Project Field : Machine Vision
Project Title : The Adaptive Markov Chain Monte Carlo Technique to Estimate Vehicle Motion.
Project : MEng in Electrical and Electronic Engineering
Project Duration : January 2010 –
Project Supervisor : Kenneth Teo, Ismail Saad
Description of Project :
The main objective of this research is to reduce the chain length of Markov Chain Monte Carlo (MCMC) to track maneuvering vehicles that undergoes overlapping situations. Overlapping situation will cause to the lost of observable vehicle information whereas maneuvering situation will gives varying vehicle outlook which enhances the tracking difficulties. MCMC is capable of tracking objects under various conditions by estimating the position of target object with the sampling of probability distributions. Each estimated position sample will be computed with Metropolis-Hasting algorithm where better samples will be accepted to form the Markov Chain. Therefore computations of MCMC are highly depending on the sampling process where the tracking error will be escalated if the sampling is not computed accurately. Consequently, situations occurred during the tracking process will affect the convergence speed of MCMC and therefore longer chain length is required for the MCMC to converge. As a result, conventional MCMC with fixed chain length is facing difficulties to determine the appropriate length to accurately track the target vehicle undergoing various situations. Short chain length will provide insufficient tracking information whereas longer chain length will provide better tracking accuracy but consume higher computational power. Thus convergence diagnostic algorithm is embedded into MCMC to quantitatively and qualitatively determine the steady state of MCMC samples. The diagnostic algorithm is capable of adaptively sampling the MCMC at optimal chain length without affecting the accuracy of the tracking algorithm. In addition, introduction of genetic operators to the adapted MCMC has further reduced the chain length by improving the convergence speed of the MCMC. Evaluation and assessment of the MCMC tracking algorithm have been carried out under multiple overlapping and maneuvering situations and results have shown that adaptive MCMC has tracked the target vehicle accurately with shorter chain length.
Contact Person : Kow Wei Yeang
Contact Email : wei_yeang@hotmail.com
Publications:
International Journal
2011
[J1101] W.Y. Kow, W.L. Khong, H.T. Yew, I. Saad, K.T.K. Teo. Adapting Markov Chain Monte Carlo with CUSUM Path Plot for Overlapped Vehicle Tracking. International Journal of Simulation, Systems, Science and Technology, IJSSST. ISSN: 1473-804x. June 2011. vol. 12, no. 3. [URL] [published]
[J1102] W.L. Khong, W.Y. Kow, L. Angeline, I. Saad, K.T.K. Teo. Overlapped Vehicle Tracking via Enhancement of Particle Filter with Adaptive Resampling Algorithm. International Journal of Simulation, Systems, Science and Technology, IJSSST. ISSN: 1473-804x. June 2011. vol. 12, no. 3. [URL] [published]
International Proceeding
2012
[P1201] M.Y. Choong, W.L. Khong, W.Y. Kow, L. Angeline, K.T.K. Teo. Graph-based Image Segmentation using K-Means Clustering and Normalised Cuts. 4th International Conference on Computational Intelligence, Communication Systems and Networks [Phuket, Thailand]. Proceeding of CICSyN 2012. ISBN: 978-0-7695-4821-0. 24-26 July 2012. pp. 307-312. [pdf] [published]
doi: 10.1109/CICSyN.2012.64 [IEEE Xplore, Scopus]
[P1202] W.Y. Kow, W.L. Khong, Y.K. Chin, I. Saad, K.T.K. Teo. Enhancement of Markov Chain Monte Carlo Convergence Speed in Vehicle Tracking Using Genetic Operator. 4th International Conference on Computational Intelligence, Modelling & Simulation [Kuantan, Malaysia]. Proceeding of CIMSim 2012. ISBN: 978-0-7695-4871-5. 25-27 September 2012. pp. 270-275. [published]
doi: 10.1109/CIMSim.2012.61 [IEEE Xplore, Scopus]
[P1203] L. Angeline, W.Y. Kow, W.L. Khong, M.Y. Choong, K.T.K. Teo. License Plate Character Recognition Via Signature Analysis and Features Extraction. 4th International Conference on Computational Intelligence, Modelling & Simulation [Kuantan, Malaysia]. Proceeding of CIMSim 2012. ISBN: 978-0-7695-4871-5. 25-27 September 2012. pp. 1-6. [published]
doi: 10.1109/CIMSim.2012.66 [IEEE Xplore, Scopus]
[P1204] W.L. Khong, W.Y. Kow, Y.K. Chin, M.Y. Choong, K.T.K. Teo. Enhancement of Particle Filter Resampling in Vehicle Tracking via Genetic Algorithm. UKSim-AMSS 6th European Modelling Symposium on Mathematical Modelling and Computer Simulation [Malta]. Proceeding of EMS 2012. ISBN: 978-0-7695-4926-2. 14-16 November 2012. pp. 230-235.
[P1205] Y.K. Chin, W.Y. Kow, W.L. Khong, M.K. Tan, K.T.K. Teo. Q-Learning Traffic Signal Optimization Within Multiple Intersections Traffic Network. 2nd International Conference on Intelligent Systems and Informatics [Bandung, Indonesia]. Proceeding of ISI 2012. 14-16 November 2012. [published]
[P1206] M.Y. Choong, W.Y. Kow, Y.K. Chin, L. Angeline, K.T.K. Teo. Image Segmentation via Normalised Cuts and Clustering Algorithm. International Conference on Control System, Computing and Engineering [Penang, Malaysia]. Proceeding of ICCSCE 2012. ISBN: 978-1-4673-3141-8. 23-25 November 2012. [pdf] [published]
[P1207] W.Y. Kow, Y.K. Chin, W.L. Khong, H.K. Lau, K.T.K. Teo. Implementing Manifold Learning in Adaptive MCMC for Tracking Vehicle under Disturbances. International Conference on Control System, Computing and Engineering [Penang, Malaysia]. Proceeding of ICCSCE 2012. ISBN: 978-1-4673-3141-8. 23-25 November 2012. [pdf] [published]
2011
[P1101] W.Y. Kow, W.L. Khong, F. Wong, I. Saad, K.T.K. Teo. Adaptive Tracking of Overlapping Vehicles via Markov Chain Monte Carlo with CUSUM Path Plot Algorithm.26-28 July 2011, 3rd International Conference on Computational Intelligence, Communication Systems and Networks [Bali, Indonesia]. Proceeding of CICSyN 2011 (IEEE), ISBN: 978-1-4577-0975-3. pp. 253-258. [URL] [published]
[P1102] W.L. Khong, W.Y. Kow, F. Wong, I. Saad, K.T.K. Teo. Enhancement of Particle Filter Approach for Vehicle Tracking via Adaptive Resampling Algorithm. 26-28 July 2011, 3rd International Conference on Computational Intelligence, Communication Systems and Networks [Bali, Indonesia]. Proceeding of CICSyN 2011 (IEEE), ISBN: 978-1-4577-0975-3. pp. 259-263. [URL] [published]
[P1103] W.L. Khong, W.Y. Kow, H.T. Tan, H.P. Yoong, K.T.K. Teo. Kalman Filtering Based Object Tracking in Surveillance Video System. 8-9 November 2011, 3rd International Conference on Science and Engineering [Miri, Malaysia]. Proceeding of CUTSE 2011, ISBN: 978-983-44176-0-4. [URL] [published]
[P1104] W.L. Khong, W.Y. Kow, Y.K. Chin, I. Saad, K.T.K. Teo. Overlapping Vehicle Tracking via Adaptive Particle Filter with Multiple Cues. 25-27 November 2011, International Conference on Control System, Computing and Engineering [Penang, Malaysia]. Proceeding of ICCSCE 2011 (IEEE), ISBN: 978-1-45771-641-6. [URL] [published]
[P1105] W.Y. Kow, W.L. Khong, Y.K. Chin, I. Saad, K.T.K. Teo. CUSUM-Variance Ratio Based Markov Chain Monte Carlo Algorithm in Overlapped Vehicle Tracking. 4-7 December 2011, International Conference on Computer Applications & Industrial Electronics [Penang, Malaysia]. Proceeding of ICCAIE 2011 (IEEE), ISBN: 978-1-4577-2057-4. pp. 45-50. [URL] [published]
[P1106] Y.S. Chia, W.Y. Kow, W.L. Khong, A. Kiring, K.T.K. Teo. Kernel-Based Object Tracking via Particle Filter and Mean Shift Algorithm. 5-8 December 2011, 11th International Conference on Hybrid Intelligent Systems [Malacca, Malaysia]. Proceeding of HIS 2011 (IEEE). [URL] [accepted]
2010
[P1001] K.T.K. Teo, W.Y. Kow, Y.K. Chin. Optimization of Traffic Flow within an Urban Traffic Light Intersection with Genetic Algorithm. 28-30 September 2010, 2nd International Conference on Computational Intelligence, Modelling and Simulation [Bali, Indonesia]. Proceeding of CIMSim 2010 (IEEE), ISBN: 978-0-7695-4262-1. pp. 172-177. [URL] [published]
Project Field : Machine Vision
Modelling, Simulation & Computing Laboratory (mscLab) 