Deep Learning Technique for Improving Data Reception in Optical Camera Communication-Based V2I
Authors:
- Dong Nyeok ChoiDCDong Nyeok Choi
- Sung Yooun JinSJSung Yooun Jin
- Juhee LeeJLJuhee Lee
- Byung Wook KimBKByung Wook Kim
- 1 July 2019
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-2
Abstract
Recently, as the existing lighting infrastructures are replaced by LED lighting, optical camera communication (OCC) technology in vehicle-infrastructure communication (V2I) systems have been actively researched. In this paper, we introduce a method to improve data packet reception rate of OCC-based V2I by using deep learning based region-of-interest (ROI) detector. In the V2I environment, traffic lights on the roadside are ROIs that perform as transmitters sending OCC data, and should be detected at the receiving camera. If ROI detectors utilize a deep learning model, data can be extracted stably during actual driving. From the experiment result, it was found that data packet reception rate using deep-learning based ROI detection technique outperforms that of conventional method based on image differentiation.Keywords
This publication has 8 references indexed in Scilit:
- A deep learning approach to traffic lights: Detection, tracking, and classification2017
- A survey of design and implementation for optical camera communicationSignal Processing: Image Communication, 2017
- You Only Look Once: Unified, Real-Time Object Detection2016
- DeepTLR: A single deep convolutional network for detection and classification of traffic lights2016
- Traffic Light Detection: A Learning Algorithm and Evaluations on Challenging Dataset2015
- Fully convolutional networks for semantic segmentation2015
- ImageNet Large Scale Visual Recognition ChallengeInternational Journal of Computer Vision, 2015
- Modulation and coding for dimmable visible light communicationIEEE Communications Magazine, 2015