Improved Pedestrian Detection Based on Extended Histogram of Oriented Gradients

Abstract:

In order to further improve pedestrian detection accuracy and avoid the disadvantage of original histogram of oriented gradients (HOG), differential template, overlap ratio and normalization method and so on are improved when HOG features are extracted, then more gradient information are extracted and feature description operators can be obtained which describe human detail features better in lager image regions or detection windows. Considering speed, we select support vector machine (SVM) using linear function kernel as a classifier. Multi-scale detection technique and non maxima suppression method are employed for precisely locating the pedestrians in the image. Experiments show that the human detection system improves detection accuracy and still maintains a relatively satisfactory speed.

• Periodical: Applied Mechanics and Materials (Volumes 347-350)
• Pages: 3815-3820
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.347-350.3815
• Online since: August 2013
• Authors: Li Hong Zhang, Lin Li
• Keywords: Histograms of Oriented Gradient (HOG), Multi-Scale Detection, Pedestrian Detection, Support Vector Machine (SVM)

References

[1] Li L, Leung MKH. Unsupervised learning of human perspective context using ME-DT for efficient human detection in surveillance. In: Proc. of the IEEE Int'l Conf. on Computer Vision and Pattern Recognition. 2008. pp.91-97.

DOI: 10.1109/cvpr.2008.4587725

[2] Wojek C, Walk S, Schiele B. Multi-Cue onboard pedestrian detection. In: Proc. of the IEEE Int'l Conf. on Computer Vision and Pattern Recognition. Miami: IEEE, 2009. pp.794-801.

DOI: 10.1109/cvpr.2009.5206638

[3] Xu R, Zhang B, Ye Q, Jiao J. Cascade L1-norm classifier for pedestrian detection. In: Proc. of the IEEE Int'l Conf. on Computer Vision and Pattern Recognition. 2010. pp.89-96.

DOI: 10.1109/cvpr.2010.5540224

[4] LI H J, LIN SH X, ZHANG Y D. A survey of video based human motion capture [J]. Journal of Computer Aided Design & Computer Graphics, 2006, 18(11). pp.1645-1651.

[5] Wu B, Nevatia R. Detection of multiple, partially occluded humans in a single image by Bayesian combination of Edgelet part detectors. In: Proc. of the IEEE Int'l Conf. on Computer Vision. Beijing, China, 2005. pp.90-97.

DOI: 10.1109/iccv.2005.74

[6] Viola P, Jones M. Rapid object detection using a boosted cascade of simple features. Conference on Computer Vision and Pattern Recognition (CVPR), Florida, USA: IEEE, 2001. pp.511-518.

DOI: 10.1109/cvpr.2001.990517

[7] Wu B, Nevatia R. Optimizing discrimination-efficiency tradeoff in integrating heterogeneous local features for object detection. (CVPR), 2008. pp.1-8.

DOI: 10.1109/cvpr.2008.4587749

[8] Dalal N, Triggs B. Histograms of oriented gradients for human detection. Conference on Computer Vision and Pattern Recognition (CVPR), 2005. pp.886-893.

DOI: 10.1109/cvpr.2005.177

[9] Munder S, Gavrila D. An experimental study on pedestrian classification. IEEE Trans. on Pattern Analysis and Machine Computer Vision. vol. 28, no. 11 Nov. 2007. pp.1863-1868.

DOI: 10.1109/tpami.2006.217

[10] Wu B, Nevatia R. Cluster boosted tree classifier for multi-view, multi-pose object detection. In: Proc. of the IEEE Int'l Conf. On Computer Vision. 2007. pp.1-8.

DOI: 10.1109/iccv.2007.4409006

[11] P. Felzenszwalb, R. Girshick, D. McAllester, D. Ramanan. Object detection with discriminatively trained part based models, JAIR, 29, (2009).

DOI: 10.1109/tpami.2009.167

[12] Cui Jianguo, Li Zhonghai. The application of support vector machine in pattern recognition [J]. IEEE Intrernational Confernce on Control and Automation, 2007, June. pp.3135-3138.

[13] Belousov A I, Verzakov S A, Von Frese J. A flexible classification approach with optimal generalization performance: support vector machines [J]. Chemometrics and Intelligent Laboratory Systems, 2002, 64. pp.15-25.

DOI: 10.1016/s0169-7439(02)00046-1