Pedestrian Detection Optimization Algorithm Based on Low-Altitude UAV

Abstract:

Pedestrian detection is one of the critical benchmarks for object detection in computer vision. In recent years, more effective detectors and features, such as Histograms of Oriented Gradients (HOG) have been proposed. The process of HOG features calculation is slow, and the features cannot satisfy represent the human body. Therefore, we adopt the multi-channel features, and propose a new improved method for accelerated integral image, the execution time of which is less than the original method. In addition, we apply novel multi-scales detection to detect new scenario, which is based on the low-altitude UAV. Under such scenario our algorithm can handle the changing in pedestrian posture and occlusion cues. The experimental results indicate that our algorithm is rapid and efficient under dynamic camera, comparing with other methods in INRIA dataset.

• Periodical: Applied Mechanics and Materials (Volumes 571-572)
• Pages: 757-763
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.571-572.757
• Online since: June 2014
• Authors: Yong De Guo*, Zhi Gang Xie, Hong Bing Ma
• Keywords: Accelerated Integral Image, Dynamic Camera, Low-Altitude UAV, Multi-Channel Features, Multi-Scales Detection

References

[1] Viola, P., & Jones, M.J., Rapid object detection using a boosted cascade of simple features. Proc. of IEEE Conf. on CVPR, Kauai, USA, pp.511-518, (2001).

DOI: 10.1109/cvpr.2001.990517

[2] Dalal, N., & Triggs, B., Histograms of oriented gradients for human detection. Proc. of IEEE Conf. on CVPR, San Diego, USA, pp.886-893, (2005).

DOI: 10.1109/cvpr.2005.177

[3] Zhu, Q., Yeh, M.C., Cheng, K.T., & Avidan, S., Fast human detection using a cascade of histograms of oriented gradients. Proc. of IEEE Conf. on CVPR, New York, USA, pp.1491-1498, (2006).

DOI: 10.1109/cvpr.2006.119

[4] Dollar, P., Tu, Z., Perona, P., & Belongie, S., Integral Channel Features. Proc. of BMVC, London, UK 2(3), pp.5-15, (2009).

[5] Dollar, P., Belongie, S., & Perona, P., The Fastest Pedestrian Detector in the West. Proc. of BMVC, Wales, UK, pp.7-17, (2010).

DOI: 10.5244/c.24.68

[6] Benenson, R., Mathias, M., Timofte, R., & Gool, V.L., Pedestrian detection at 100 frames per second. Proc. of IEEE Conf. on CVPR, Providence, USA, pp.2903-2910, (2012).

DOI: 10.1109/cvpr.2012.6248017

[7] Zhang, C., & Viola P.A., Multiple-Instance Pruning For Learning Efficient Cascade Detectors. Proc. of Conf. on NIPS, Vancouver, CA, pp.3-10, (2007).

[8] Felzenszwalb, P.F., et al. Object detection with discriminatively trained part-based models. IEEE Trans. on PAMI, 32(9), pp.1627-1645, (2010).

DOI: 10.1109/tpami.2009.167