Enhancing real‐time traffic volume prediction: A two‐step approach of object detection and time series modelling

Abstract A two‐step framework that integrates real‐time data collection with time series forecasting models for predicting traffic volume is proposed. In the first step, the framework utilizes live highway surveillance video data and YOLO‐v7 object detector to construct accurate traffic volume data....

Full description

Saved in:
Bibliographic Details
Main Authors: Junwoo Lim, Juyeob Lee, Chaehee An, Eunil Park
Format: Article
Language:English
Published: Wiley 2024-12-01
Series:IET Intelligent Transport Systems
Subjects:
Online Access:https://doi.org/10.1049/itr2.12576
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846139788770410496
author Junwoo Lim
Juyeob Lee
Chaehee An
Eunil Park
author_facet Junwoo Lim
Juyeob Lee
Chaehee An
Eunil Park
author_sort Junwoo Lim
collection DOAJ
description Abstract A two‐step framework that integrates real‐time data collection with time series forecasting models for predicting traffic volume is proposed. In the first step, the framework utilizes live highway surveillance video data and YOLO‐v7 object detector to construct accurate traffic volume data. In the second step, an ARIMA–LSTM time series model is applied to forecast future traffic volumes. Experimental results show that YOLO‐v7 achieved a vehicle detection accuracy of over 93.30%, ensuring high precision in traffic volume data construction. The ARIMA–LSTM model demonstrated superior performance in traffic volume prediction, with a mean squared error of 87.97, root mean squared error of 10,388.57, and mean absolute error of 101.39. YOLO‐v7's detection speed of 7.8 ms per frame further validates the feasibility of real‐time data construction. The findings indicate that the combination of YOLO‐v7 for vehicle detection and ARIMA–LSTM for traffic prediction is highly effective, offering a significant reduction in training time compared to more complex deep learning models while maintaining high prediction accuracy. This research presents a unified solution for traffic data collection and prediction, enhancing transportation infrastructure planning and optimizing traffic flow. Future work will focus on extending the prediction intervals and further refining the models to improve performance.
format Article
id doaj-art-9ab1cb28f52c4956aad044fa87abb59a
institution Kabale University
issn 1751-956X
1751-9578
language English
publishDate 2024-12-01
publisher Wiley
record_format Article
series IET Intelligent Transport Systems
spelling doaj-art-9ab1cb28f52c4956aad044fa87abb59a2024-12-06T05:51:13ZengWileyIET Intelligent Transport Systems1751-956X1751-95782024-12-0118122744275810.1049/itr2.12576Enhancing real‐time traffic volume prediction: A two‐step approach of object detection and time series modellingJunwoo Lim0Juyeob Lee1Chaehee An2Eunil Park3Department of Applied Artificial IntelligenceSungkyunkwan UniversitySeoulSouth KoreaDepartment of Applied Artificial IntelligenceSungkyunkwan UniversitySeoulSouth KoreaDepartment of Applied Artificial IntelligenceSungkyunkwan UniversitySeoulSouth KoreaDepartment of Applied Artificial IntelligenceSungkyunkwan UniversitySeoulSouth KoreaAbstract A two‐step framework that integrates real‐time data collection with time series forecasting models for predicting traffic volume is proposed. In the first step, the framework utilizes live highway surveillance video data and YOLO‐v7 object detector to construct accurate traffic volume data. In the second step, an ARIMA–LSTM time series model is applied to forecast future traffic volumes. Experimental results show that YOLO‐v7 achieved a vehicle detection accuracy of over 93.30%, ensuring high precision in traffic volume data construction. The ARIMA–LSTM model demonstrated superior performance in traffic volume prediction, with a mean squared error of 87.97, root mean squared error of 10,388.57, and mean absolute error of 101.39. YOLO‐v7's detection speed of 7.8 ms per frame further validates the feasibility of real‐time data construction. The findings indicate that the combination of YOLO‐v7 for vehicle detection and ARIMA–LSTM for traffic prediction is highly effective, offering a significant reduction in training time compared to more complex deep learning models while maintaining high prediction accuracy. This research presents a unified solution for traffic data collection and prediction, enhancing transportation infrastructure planning and optimizing traffic flow. Future work will focus on extending the prediction intervals and further refining the models to improve performance.https://doi.org/10.1049/itr2.12576artificial intelligenceobject detectionreal‐time systemsroad traffictime seriestraffic management and control
spellingShingle Junwoo Lim
Juyeob Lee
Chaehee An
Eunil Park
Enhancing real‐time traffic volume prediction: A two‐step approach of object detection and time series modelling
IET Intelligent Transport Systems
artificial intelligence
object detection
real‐time systems
road traffic
time series
traffic management and control
title Enhancing real‐time traffic volume prediction: A two‐step approach of object detection and time series modelling
title_full Enhancing real‐time traffic volume prediction: A two‐step approach of object detection and time series modelling
title_fullStr Enhancing real‐time traffic volume prediction: A two‐step approach of object detection and time series modelling
title_full_unstemmed Enhancing real‐time traffic volume prediction: A two‐step approach of object detection and time series modelling
title_short Enhancing real‐time traffic volume prediction: A two‐step approach of object detection and time series modelling
title_sort enhancing real time traffic volume prediction a two step approach of object detection and time series modelling
topic artificial intelligence
object detection
real‐time systems
road traffic
time series
traffic management and control
url https://doi.org/10.1049/itr2.12576
work_keys_str_mv AT junwoolim enhancingrealtimetrafficvolumepredictionatwostepapproachofobjectdetectionandtimeseriesmodelling
AT juyeoblee enhancingrealtimetrafficvolumepredictionatwostepapproachofobjectdetectionandtimeseriesmodelling
AT chaeheean enhancingrealtimetrafficvolumepredictionatwostepapproachofobjectdetectionandtimeseriesmodelling
AT eunilpark enhancingrealtimetrafficvolumepredictionatwostepapproachofobjectdetectionandtimeseriesmodelling