The process of traffic information has become an important aspect of automated driving nowadays. Thus, in this paper a small mobile system is proposed, in which MobileNetV2+SSD model is deployed and combined with binocular stereo-vision ranging to realize the recognition and ranging of traffic lights. Experiment results with real images show that this system can effectively process the information of traffic lights under real road conditions.
Convolution neural network (CNN), Feature Extraction
• Hardware: Processor: i3 ,i5 RAM: 4GB Hard disk: 16 GB • Software: operating System : Windws2000/XP/7/8/10 Anaconda,jupyter,spyder,flask,hadoop Frontend :-python Backend:- MYSQL
₹10000 (INR)
2019