Context-Aware Object Region Proposals for Efficient Vehicle Detection from Traffic Surveillance Videos Using Deep Neural Networks

DOI

10.2991/isaeece-17.2017.60How to use a DOI?

Keywords

Region Propose, Vehicle Detection, Image Segmentation, Traffic Surveillance, Deep Convolutional Neural Network (DCNN)

Abstract

Recently, many methods based on deep neural networks have been developed for object recognition, which dominate various performance competitions on public datasets such as ImageNet and Pascal VOC. Existing methods suffer from high computational complexity and/or insufficient recognition accuracy for practical use. In this paper, we demonstrate that, in specific application domains, such as traffic video surveillance, the priori knowledge or environmental context information can be utilized to dramatically reduce the computational complexity and improve the object detection performance. Specifically, our method models the traffic scene background, using the model as a context to guide the generation of a much smaller number of high quality object region proposals that maintain 100% coverage. We then train a deep convolutional neural network (DCNN) to classify these proposal regions and have achieved 99% accuracy on a large test dataset, which outperforms existing methods DCNN-based methods, such as YOLO.

Copyright

© 2017, 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  - Jianhe Yuan
AU  - Wenming Cao
AU  - Fangfang Lv
PY  - 2017/03
DA  - 2017/03
TI  - Context-Aware Object Region Proposals for Efficient Vehicle Detection from Traffic Surveillance Videos Using Deep Neural Networks
BT  - Proceedings of the 2017 2nd International Symposium on Advances in Electrical, Electronics and Computer Engineering (ISAEECE 2017)
PB  - Atlantis Press
SP  - 316
EP  - 320
SN  - 2352-5401
UR  - https://doi.org/10.2991/isaeece-17.2017.60
DO  - 10.2991/isaeece-17.2017.60
ID  - Yuan2017/03
ER  -