Abstract
The kernel density estimate does not need to have the characteristic distribution hypothesis to the background, it also does not require the estimation parameter, and it can deal with the moving target detection under the complex background, but the kernel function bandwidth choice uniformly puzzles the algorithm application. To solve this problem, this paper proposes a fusion method of human body characteristics and kernel density estimation for pedestrian detection. Firstly, the kernel function bandwidth is chosen by the prior information of moving target, then the foreground (moving target) is extracted based on kernel density estimation, finally, using human features to detect video pedestrians. The experimental results show that the calculation of kernel density estimation is reduced by comparing introduction of prior information with traditional methods, and the pedestrian and no pedestrian can be detected accurately by the interference of light variation and noise.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Shao, L., Liu, Y., Zhang, J.: Human action segmentation and recognition via motion and shape analysis. Pattern Recogn. Lett. 33(4), 438–445 (2012)
Matsubara, T., Hyon, S.-H., Morimoto, J.: Real-time stylistic prediction for whole-body human motions. Neural Netw. 25, 191–199 (2012)
Chen, D., Ren, S., Wei, Y., Cao, X., Sun, J.: Joint cascade face detection and alignment. In: Fleet, D., Pajdla, T., Schiele, B., Tuytelaars, T. (eds.) ECCV 2014. LNCS, vol. 8694, pp. 109–122. Springer, Cham (2014). https://doi.org/10.1007/978-3-319-10599-4_8
Elgammal, A., Harwood, D., Davis, L.: Non-parametric model for background subtraction. In: Vernon, D. (ed.) ECCV 2000. LNCS, vol. 1843, pp. 751–767. Springer, Heidelberg (2000). https://doi.org/10.1007/3-540-45053-X_48
Wang, L., Weiming, H., Tan, T.: Identification based on gait. J. Comput. Sci. 26(3), 353–360 (2003)
Wang, L., Hu, W., Tan, T.: Recent developments in human motion analysis. Pattern Recogn. 36(3), 585–601 (2003)
Zhang, L., Yu, H.: Tracking method of human body occlusion under dual camera. Chin. J. Graph. Graph. 16(4), 606–612 (2011)
Sun, Y., Weng, P.: Study on pedestrian recognition and tracking based on particle swarm optimization algorithm. Comput. Eng. Des. 32(3), 988–994 (2011)
Sun, J., Chen, Y., Ji, Z.: Kernel density estimation background modeling method based on key frames. Opt. Technol. 34(5), 699–701 (2008)
Yu, J., Xu, D., Liao, Q.: Research progress of image fog technology. Chin. J. Graph. 16(9), 1561–1576 (2011)
Jin, T., Zhou, F., Bai, X.: Detection of motion target for space-based video based on kernel density estimation. Infrared Laser Eng. 40(1), 153–157 (2011)
Liu, Y.: Research on algorithm of moving target detection and tracking based on super pixel. Master’s degree thesis of China University of Science and Technology, Hefei (2013)
Rui, L., Lei, Z.: Design of syntactic parsing unit for structural pattern recognition system based on relational algebra. J. Dalian Jiaotong Univ. 33(3), 59–63 (2012)
Chen, F., Shui, A., Li, L.: Small sample pattern recognition method for pipeline plugging in storage and transportation process. Comput. Appl. Res. 31(7), 2031–2034 (2014)
Peng, D.: Research on brain activity of paired association learning in visual haptic cross pattern. Doctoral dissertations of East China Normal University, Shanghai (2015)
Cheng, X.: Research on pattern recognition framework APRF based on multi agent. Doctoral dissertations of Nanjing University of Science and Technology, Nanjing (2006)
Di, J., Yin, J.: Application of Wavelet Analysis. Science Press, Beijing (2017)
Haken, H. (Yang Jia-ben translated): Work with Computer and Cognitive - Top-Down Method of Neural Network. Tsinghua University Press, Beijing (1994)
Acknowledgment
This project is supported by Anhui University Natural Science Research Project (No. KJ2018A0431), Jiangsu Modern Educational Technology Research Project (No. 2017-R-54131), Nantong Science and Technology Project (No. MS12016036), Research on Teaching Reform at Nantong University (No. 2018043).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2019 Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Cheng, S., Zhou, M., Lu, C., Li, Y., Wang, Z. (2019). Detection of Pedestrians Based on the Fusion of Human Characteristics and Kernel Density Estimation. In: Peng, H., Deng, C., Wu, Z., Liu, Y. (eds) Computational Intelligence and Intelligent Systems. ISICA 2018. Communications in Computer and Information Science, vol 986. Springer, Singapore. https://doi.org/10.1007/978-981-13-6473-0_31
Download citation
DOI: https://doi.org/10.1007/978-981-13-6473-0_31
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-13-6472-3
Online ISBN: 978-981-13-6473-0
eBook Packages: Computer ScienceComputer Science (R0)