Adaptive Target Tracking based on Kernel SVM and HOG Features

  IJETT-book-cover  International Journal of Recent Engineering Science (IJRES)          
  
© 2018 by IJRES Journal
Volume-5 Issue-1
Year of Publication : 2018
Authors : Yarui Wang, Zuolei Sun, Xiaoyu Wang
  10.14445/23497157/IJRES-V5I1P103

MLA 

MLA Style: Yarui Wang, Zuolei Sun, Xiaoyu Wang  "Adaptive Target Tracking based on Kernel SVM and HOG Features" International Journal of Recent Engineering Science 5.1(2018):11-15. 

APA Style: Yarui Wang, Zuolei Sun, Xiaoyu Wang. Adaptive Target Tracking based on Kernel SVM and HOG Features.  International Journal of Recent Engineering Science, 5(1),11-15.

Abstract
The traditional target tracking algorithm is popular with its real-time. Based on its discriminative model, this paper extracts HOG features into the Kernel SVM classifier to distinguish the foreground and background, which also uses a sliding window with an adaptive scale to divide the image into different scale image blocks. Meanwhile, aiming to improve the speed of calculation, we perform Fast Fourier Transform on the input image to process it in the frequency domain and then obtain rotating invariant gradient information as its feature map to distinguish the target and background the kernel SVM classifier. We performed our experiments on six challenging videos from the visual tracker benchmark and compared them with two well-known KCF and CT tracking algorithms. The experiment results show that our approach can track the targets accurately in a fast way, even though there are challenges such as motion blur and occlusion in the testing videos.

Reference

[1] Lei Z, Huimin L. Application of Projection Pursuit Model in residential building comprehensive performance assessment[C]//Intelligent Computing and Intelligent Systems (ICIS), 2010 IEEE International Conference on. IEEE, 2010, 3: 851-855.
[2] Stone, Lawrence D., et al. Bayesian multiple target tracking. Artech House, 2013.
[3] Nägeli, Tobias, et al. "Real-Time Motion Planning for Aerial Videography With Real-Time With Dynamic Obstacle Avoidance and Viewpoint Optimization." IEEE Robotics and Automation Letters 2.3 (2017): 1696-1703.
[4] Kakoty N M, Kaiborta M, Hazarika S M. Electromyographic grasp recognition for a five-fingered robotic hand[J]. IAES International Journal of Robotics and Automation, 2013, 2(1): 1.
[5] Jebelli A, Yagoub M C E, Dhillon B S. Design and Control of Underwater Robots with Rotating Thrusters[J]. IAES International Journal of Robotics and Automation (IJRA), 2016, 5(4): 284-294.
[6] Xu Y, Yu G, Wang Y, et al. A hybrid vehicle detection method based on viola-jones and HOG+ SVM from UAV images[J]. Sensors, 2016, 16(8): 1325.
[7] Dadi H S, Pillutla G K M, Makkena M L. Face Recognition and Human Tracking Using GMM, HOG and SVM in Surveillance Videos[J]. Annals of Data Science, 2017: 1-23.
[8] Ouerhani, Yousri, Ayman Alfalou, and Christian Brosseau. "Road mark recognition using HOG-SVM and correlation." Optics and Photonics for Information Processing XI. Vol. 10395. International Society for Optics and Photonics, 2017.
[9] Kim, Sanggi, and Dong Seog Han. "Real Time Traffic Light Detection Algorithm Based on Color Map and Multilayer HOG-SVM." 방송공학회논문지 22.1 (2017): 62-69.
[10] Wei W, Jia Q. Weighted feature Gaussian kernel SVM for emotion recognition[J]. Computational intelligence and neuroscience, 2016, 2016: 11.
[11] Rao T C S, Sreenivasulu G. A Kernel SVM Classifier for Classification of Brain Tumors in Magnetic Resonance Images[J]. i-manager's Journal on Image Processing, 2016, 3(3): 34.
[12] Prasad S, Ramkumar B. Passive copy-move forgery detection using SIFT, HOG and SURF features[C]//Recent Trends in Electronics, Information & Communication Technology (RTEICT), IEEE International Conference on. IEEE, 2016: 706-710.
[13] Wang Q. Ore image edge detection using HOG-index dictionary learning approach[J]. The Journal of Engineering, 2017, 1(1).
[14] Lampert C H, Blaschko M B, Hofmann T. Beyond sliding windows: Object localization by efficient subwindow search[C]//Computer Vision and Pattern Recognition, 2008. CVPR 2008. IEEE Conference on. IEEE, 2008: 1-8.
[15] Wang W, Liu Y, Tang X, et al. A novel IRA-like codes construction method based on cyclic matrix permutation[C]//Image, Vision and Computing (ICIVC), 2017 2nd International Conference on. IEEE, 2017: 909-912.
[16] Doucet A, Godsill S, Andrieu C. On sequential Monte Carlo sampling methods for Bayesian filtering[J]. Statistics and computing, 2000, 10(3): 197-208.
[17] Siji Joseph, Arun Pradeep "Object Tracking using HOG and SVM", International Journal of Engineering Trends and Technology (IJETT), V48(6),321-325 June 2017.
[18] R. Rifkin, G. Yeo, and T. Poggio, “Regularised least-squares classification,” Nato Sci. Ser. Sub Ser. III, vol. 190, pp. 131–154, 2003.
[19] Weickert J, Grewenig S, Schroers C, et al. Cyclic schemes for PDE-based image analysis[J]. International Journal of Computer Vision, 2016, 118(3): 275-299.
[20] Siji Joseph, Arun Pradeep "Object Tracking using HOG and SVM", International Journal of Engineering Trends and Technology (IJETT), V48(6),321-325 June 2017.
[21] Cheng Y. Mean shift, mode seeking, and clustering[J]. IEEE transactions on pattern analysis and machine intelligence, 1995, 17(8): 790-799.
[22] Doucet A, Godsill S, Andrieu C. On sequential Monte Carlo sampling methods for Bayesian filtering[J]. Statistics and computing, 2000, 10(3): 197-208.
[23] Jazwinski A H. Stochastic processes and filtering theory[M]. Courier Corporation, 2007.
[24] Du W, Piater J. Tracking by cluster analysis of feature points using a mixture particle filter[C]//Advanced Video and Signal Based Surveillance, 2005. AVSS 2005. IEEE Conference on. IEEE, 2005: 165-170.
[25] Orellana L H, Rodriguez-R L M, Konstantinidis K T. ROCker: accurate detection and quantification of target genes in short-read metagenomic data sets by modeling sliding-window bitscores[J]. Nucleic acids research, 2017, 45(3): e14-e14.
[26] Wu Y, Lim J, Yang M H. Online object tracking: A benchmark[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2013: 2411-2418.
[27] Everingham M, Zisserman A, Williams C K I, et al. The PASCAL visual object classes challenge 2007 (VOC2007) results[J]. 2007.
[28] R Sindhoori, "Digital image processing. Multi feature face recognition in PSO -SVM" SSRG International Journal of Electrical and Electronics Engineering 1.3 (2014): 1-6.
[29] Cortes C, Vapnik V. Support-vector networks[J]. Machine learning, 1995, 20(3): 273-297.
[30] Jolliffe I. Principal component analysis[M]. John Wiley & Sons, Ltd, 2002.

Keywords
Object tracking, HOG feature, kernel SVM