This study is to analyze the influence of visibility in a foggy weather environment on the accuracy of machine vision obstacle detection in assisted driving. We present a foggy day imaging model and analyze the image characteristics, then we set up the faster region convolutional neural network (Faster R-CNN) as the basic network for target detection in the simulation experiment and use Karlsruhe Institute of Technology and Toyota Technological Institute (KITTI) data for network detection and classification training. PreScan software is used to build weather and traffic scenes based on a foggy imaging model, and we study object detection of machine vision in four types of weather condition—clear (no fog), light fog, medium fog, and heavy fog—by simulation experiment. The experimental results show that the detection recall is 91.55%, 85.21%, 72.54~64.79%, and 57.75% respectively in no fog, light fog, medium fog, and heavy fog environments. Then we used real scenes in medium fog and heavy fog environment to verify the simulation experiment. Through this study, we can determine the influence of bad weather on the detection results of machine vision, and hence we can improve the safety of assisted driving through further research.
Urban emergencies are hard to avoid. Traffic emergency response after an incident plays an important role in reducing losses and is a key link in urban emergency management. However, the traditional traffic management methods have been difficult to face the complicated conditions and requirements in such problems. The introduction of Internet of Things and data mining technology to establish a traffic emergency response system under urban emergencies can significantly improve the level of urban emergency response and realize efficient intensive management. The system mainly includes sub-systems, such as personnel evacuation data collection system, vehicle operation data collection system, rescue material distribution data collection system, personnel settlement place data collection system, traffic bayonet intelligent identification system, etc. It also devises the working programs for command management, personnel evacuation and disaster disposal in case of emergency, and improves the urban emergency support management system. With the support of Internet of Things and the data mining technology, the traffic emergency response system can timely and accurately control the flowing information of personnel and vehicles, quickly and conveniently resettle personnel and vehicles, effectively carry out follow-up rescue work, effectively improve rescue efficiency and improve the level of urban management.
Internet of vehicles is a specific application of Internet of things technology in the field of intelligent transportation. The rapid development of 5G communication technology promotes the development of Internet of vehicles. Car for cellular network communication node random distribution and complex multi-source interference and mobile terminal security calculation ability is limited, this article in view of the actual scene, was proposed based on random geometry contains eavesdropper (Eve) honeycomb - V2V heterogeneous physical layer security system model, the introduction of automatic (PB) as artificial floating vehicle noise, the analysis of cellular network users in the system (CU), V2V users (VU) and interference of the eavesdropper, each user letter simulation with dry to noise ratio (SINR) of the cumulative distribution function, and by using the random geometry tools related safety expression deduction, Then, data mining was carried out on the distance between PB and VU receiver, PB transmitting power and other related variables through genetic algorithm, and the value process was visualized to extract valuable information, providing a mathematical analysis framework and theoretical guidance for the future design, deployment and operation of cellular vehicle network. The results show that the proposed system model can significantly improve the security of vehicle-network communication.
At present, in order to protect the personal safety of riders, electric vehicle riders wearing helmets has become a mandatory requirement of the traffic control department. In order to improve the supervision of cyclists and protect the safety of cyclists. This paper proposes a road helmet detection algorithm based on improved YOLOv5s. Firstly, the SE attention module is embedded in the backbone network of YOLOv5s to solve the problem that small target detection is easy to miss, and secondly, the C3 residual module of the backbone network is replaced by the lightweight ShuffleNetV2 module to reduce the complexity of the network and improve the speed of target detection and identification. Finally, the network is trained and tested on self-made dataset and compared with the original model detection effect. The experimental results show that the detection precision, recall, and mAP0.5 are increased by 4.5%, 3.6, and 4.1%, respectively, after the improvement, and the improved YOLOv5s model can effectively improve the efficiency and timeliness of EV helmet recognition.