People Counting in Conference Scenes in Shearlet Domain

Nguyen Thanh Binh

Nature of Computation and Communication. Second International Conference, ICTCC 2016, Rach Gia, Vietnam, March 17-18, 2016, Revised Selected Papers

Research Article

People Counting in Conference Scenes in Shearlet Domain

Download

259 downloads

Cite: BibTeX Plain Text

@INPROCEEDINGS{10.1007/978-3-319-46909-6_33,
    author={Nguyen Thanh Binh},
    title={People Counting in Conference Scenes in Shearlet Domain},
    proceedings={Nature of Computation and Communication. Second International Conference, ICTCC 2016, Rach Gia, Vietnam, March 17-18, 2016, Revised Selected Papers},
    proceedings_a={ICTCC},
    year={2017},
    month={1},
    keywords={People counting Shearlet transform Saliency map Color smoothing},
    doi={10.1007/978-3-319-46909-6_33}
}

Nguyen Thanh Binh
Year: 2017
People Counting in Conference Scenes in Shearlet Domain
ICTCC
Springer
DOI: 10.1007/978-3-319-46909-6_33

Nguyen Thanh Binh¹^,*

1: Ho Chi Minh City University of Technology, VNU-HCM

*Contact email: ntbinh@hcmut.edu.vn

Abstract

People counting is an important task in visual-based surveillance system. The task of people counting is not easy to solve problems. In this paper, the author has proposed a method for people counting which identify the objects present in a scene of conference into two classes: empty seat and non-empty seat. The proposed method based on saliency map and color smoothing in shearlet domain. The author uses shearlet transform and combine of adaboost with support vector machine for classifiers and people counting. The proposed method is simple but the accuracy of people counting is high.

Keywords: People counting Shearlet transform Saliency map Color smoothing

Published: 2017-01-04
Appears in: SpringerLink

: http://dx.doi.org/10.1007/978-3-319-46909-6_33