Realization of Traffic Video Surveillance on DM3730 Chip

Xin Zhang; Hang Dong

Machine Learning and Intelligent Communications. Second International Conference, MLICOM 2017, Weihai, China, August 5-6, 2017, Proceedings, Part II

Research Article

Realization of Traffic Video Surveillance on DM3730 Chip

Download

112 downloads

Cite: BibTeX Plain Text

@INPROCEEDINGS{10.1007/978-3-319-73447-7_44,
    author={Xin Zhang and Hang Dong},
    title={Realization of Traffic Video Surveillance on DM3730 Chip},
    proceedings={Machine Learning and Intelligent Communications. Second International Conference, MLICOM 2017, Weihai, China, August 5-6, 2017, Proceedings, Part II},
    proceedings_a={MLICOM},
    year={2018},
    month={2},
    keywords={Multi-objects tracking DM3730 OpenCV},
    doi={10.1007/978-3-319-73447-7_44}
}

Xin Zhang
Hang Dong
Year: 2018
Realization of Traffic Video Surveillance on DM3730 Chip
MLICOM
Springer
DOI: 10.1007/978-3-319-73447-7_44

Xin Zhang¹^,*, Hang Dong¹^,*

1: Tongji University

*Contact email: mic_zhangxin@tongji.edu.cn, dh@tongji.edu.cn

Abstract

A general method for traffic video surveillance task involves foreground detecting and moving objects’ tracking. The Gaussian mixture model is generally used in detecting foreground and the Kalman filter is used in multi-objects tracking. This paper has implemented a multi-objects tracking system using DM3730 development board as the hardware platform, which is powerful at image processing and analysis. This paper will adopt an Open Computer Vision library (OpenCV) to efficiently implement the overall system. The OpenCV library with a large amount of optimized algorithms in computer vision and machine learning will facilitate the realization of the system. The testing results demonstrate the effectiveness of the system through tracking of vehicles.

Keywords: Multi-objects tracking DM3730 OpenCV

Published: 2018-02-09
Appears in: SpringerLink

: http://dx.doi.org/10.1007/978-3-319-73447-7_44