AUTHOR=Zheng Ye , Li Xiaoming , Xu LiuChang , Wen Nu 

TITLE=A Deep Learning–Based Approach for Moving Vehicle Counting and Short-Term Traffic Prediction From Video Images

JOURNAL=Frontiers in Environmental Science

VOLUME=Volume 10 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/environmental-science/articles/10.3389/fenvs.2022.905443

DOI=10.3389/fenvs.2022.905443

ISSN=2296-665X

ABSTRACT=Intelligent transportation system (ITS) is one of the effective solutions to the problem of urban traffic congestion, and it is also one of the important topics of smart city construction. One particular application is the traffic monitoring and flow prediction. However, there are still challenges regarding both aspects. On one hand, current traffic monitoring relies heavily on single object detection method that cannot achieve accurate statistics of moving-target counting, and meanwhile, have limited speed advantage; On the other, the existing traffic flow prediction models rarely considers different weather conditions. Therefore, the present paper attempts to propose a packaged solution which combines a new target tracking and moving vehicle counting method and an improved long and short-term memory (LSTM) network for traffic flow forecast with weather conditions. More specifically, DCN V2 convolution kernel and MultiNetV3 framework are used to replace Yolo V4's conventional convolution kernel and backbone network to realize multi-target tracking and counting respectively. Subsequently, combined with the temporal characteristics of historical traffic flow, this paper introduces weather conditions into the LSTM network and realizes the short-term prediction of traffic flow at the road junction level. This study carries out a series of experiments using the real traffic video data with a two-month timespan at a popular road junction in the downtown of Shenzhen, China. The results suggest the proposed algorithms outperforms the previous methods in terms of the 10 percent higher accuracy of the target detection tracking, and about a half reduction of traffic prediction error when considering weather conditions.