Off-position detection based on convolutional neural network

DOI

10.2991/aest-16.2016.96How to use a DOI?

Keywords

off-position detection; CNN; classification.

Abstract

As a part of the intelligent video surveillance, off-position detection,which needs a real-time and precise algorithm, is used to detect whether the person on duty is absent from working position.This work is necessary for improving efficiency and reducing human resource consumption.Considering the excellent performance of convolutional neural network in image classification, we first propose a method for off-position detection using CNN in this paper and get good results.Furthermore,we introduce a new dataset for working position by generating crops from video frames.Then we randomly generate 224x224 crops from training images to fine-tune our deep neural network.

Copyright

© 2016, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Tianbing Zhang
AU  - Wang Luo
AU  - Qiwei Peng
AU  - Gongyi Hong
AU  - Min Feng
AU  - Yuan Xia
AU  - Lei Yu
AU  - Xu Wang
AU  - Yang Li
PY  - 2016/11
DA  - 2016/11
TI  - Off-position detection based on convolutional neural network
BT  - Proceedings of the 2016 International Conference on Advanced Electronic Science and Technology (AEST 2016)
PB  - Atlantis Press
SP  - 718
EP  - 724
SN  - 1951-6851
UR  - https://doi.org/10.2991/aest-16.2016.96
DO  - 10.2991/aest-16.2016.96
ID  - Zhang2016/11
ER  -