Abstract:Traffic problems continue to deteriorate because of increasing population in urban areas that rely on many modes of transportation, the transportation infrastructure has achieved considerable strides in the last several decades. This has led to an increase in congestion control difficulties, which directly affect citizens through air pollution, fuel consumption, traffic law breaches, noise pollution, accidents, and loss of time. Traffic prediction is an essential aspect of an intelligent transportation system in smart cities because it helps reduce overall traffic congestion. This article aims to design and enforce a traffic prediction scheme that is efficient and accurate in forecasting traffic flow. Available traffic flow prediction methods are still unsuitable for real-world applications. This fact motivated us to work on a traffic flow forecasting issue using Vision Transformers (VTs). In this work, VTs were used in conjunction with Convolutional neural networks (CNN) to predict traffic congestion in urban spaces on a city-wide scale. In our proposed architecture, a traffic image is fed to a CNN, which generates feature maps. These feature maps are then fed to the VT, which employs the dual techniques of tokenization and projection. Tokenization is used to convert features into tokens containing Vision information, which are then sent to projection, where they are transformed into feature maps and ultimately delivered to LSTM. The experimental results demonstrate that the vision transformer prediction method based on Spatio-temporal characteristics is an excellent way of predicting traffic flow, particularly during anomalous traffic situations. The proposed technology surpasses traditional methods in terms of precision, accuracy and recall and aids in energy conservation. Through rerouting, the proposed work will benefit travellers and reduce fuel use.

Traffic Safety Oriented Multi-Intersection Flow Prediction Based on Transformer and CNN

Short-Term Forecasting of Traffic Flow and Speed: A Deep Learning Approach

Short-Term Speed Forecasting of Large-Scale Urban Road Network Based on Transformer

Traffic flow prediction based on transformer

Spatial linear transformer and temporal convolution network for traffic flow prediction

MF-CNN: Traffic Flow Prediction Using Convolutional Neural Network and Multi-Features Fusion

Daily Traffic Flow Forecasting Through a Contextual Convolutional Recurrent Neural Network Modeling Inter- and Intra-Day Traffic Patterns.

Real-Time Movement-Based Traffic Volume Prediction at Signalized Intersections

A Multi-Layer Model Based on Transformer and Deep Learning for Traffic Flow Prediction

Machine Learning-Based Traffic Flow Prediction and Intelligent Traffic Management

Urban Short - Term Traffic Flow Prediction Algorithm Based on CNN-LSTM Model

A Multi-Channel Spatial-Temporal Transformer Model for Traffic Flow Forecasting

TST-Trans: A Transformer Network for Urban Traffic Flow Prediction

Two-Stream Multi-Channel Convolutional Neural Network (TM-CNN) for Multi-Lane Traffic Speed Prediction Considering Traffic Volume Impact

A Vision Transformer Approach for Traffic Congestion Prediction in Urban Areas

Multi-graph Spatio-temporal Graph Convolutional Network for Traffic Flow Prediction

City-Wide Traffic Congestion Prediction Based on CNN, LSTM and Transpose CNN

Short-Term Traffic Flow Prediction: An Integrated Method of Econometrics and Hybrid Deep Learning

Multi-scale spatial-temporal aware transformer for traffic prediction

STTD: spatial-temporal transformer with double recurrent graph convolutional cooperative network for traffic flow prediction

Traffic Volume Prediction Based on Multi-Sources GPS Trajectory Data by Temporal Convolutional Network