Image-to-Image Learning to Predict Traffic Speeds by Considering Area-Wide Spatio-Temporal Dependencies

Jo, Dohyoung; Yu, Byeonghyeop; Jeon, Hyunjeong; Sohn, Keemin

doi:10.1109/tvt.2018.2885366

Cited by 41 publications

(15 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Initial studies mostly focused on image detection pertaining to the detection of traffic signs (CireşAn et al 2012), vehicles (Chen et al 2014), and pedestrians (Ouyang and Wang 2013). However, lately deep learning methods have been applied in many studies analysing complex variables such as traffic state prediction (Bai and Chen 2019;Jo et al 2019), travel demand estimation (Tang et al 2019a, b), mode choice and activity choice prediction (Zhao et al 2019a, b, c, d, e, f). LeCun et al (2015) categorised the deep learning methods into three major groups, (a) multilayer architecture using backpropagation (Bengio et al 2007), (b) convolutional neural networks (CNN) (Simonyan and Zisserman 2014), and c) recurrent neural networks (RNN) (Graves et al 2013).…”

Section: A Brief Overview Of Existing Studiesmentioning

confidence: 99%

Deep Learning in Transport Studies: A Meta-analysis on the Prediction Accuracy

Varghese

Chikaraishi

Urata

2020

J. Big Data Anal. Transp.

View full text Add to dashboard Cite

Deep learning methods are being increasingly applied in transport studies, while the methods require modellers to go through a try-and-error model tuning process particularly on choosing neural network structure. Moreover, the accuracy level also depends on other factors such as the type of data, sample size, region of data collection, and time of prediction. To efficiently facilitate such a model tuning process, this study attempts to summarize the relationship between the prediction accuracy of deep learning models and the factors which influence it. We conducted a comprehensive review of the literature by adopting a detailed search strategy, followed by a meta-analysis on prediction accuracy. Four separate linear mixed effects models, taking into account unobserved heterogeneities in prediction accuracy across studies, were developed to statistically test the impacts of influential factors on prediction accuracy for (a) all observations (136 studies; 2314 cases), (b) studies with MAPE, MRE, and average accuracy indicators (86 studies; 1,878 cases), (c) classification-based studies with accuracy indicator (29 studies; 220 cases), and (d) traffic forecasting studies with MAPE, MRE, and average accuracy indicators (36 studies, 991 cases). The final model includes additional factors to test the influence of sample size and time horizon of prediction variables. The findings showed that, as expected, deep learning models, particularly ones that consider spatiotemporal dependencies of transport phenomena, show better prediction accuracies compared to conventional machine learning models. We also found that, on average, the prediction accuracy is improved by 5.90% with 100 million additional data, while the accuracy is reduced by 5.28% with 100 min increase in time horizon of prediction in traffic forecasting studies. We concluded this paper with a comprehensive summary of the existing findings on the applications of deep learning to transport studies.

show abstract

Section: A Brief Overview Of Existing Studiesmentioning

confidence: 99%

Deep Learning in Transport Studies: A Meta-analysis on the Prediction Accuracy

Varghese

Chikaraishi

Urata

2020

J. Big Data Anal. Transp.

View full text Add to dashboard Cite

show abstract

“…4b), allowing convolution layers to extract successive spatial extent of the input data. The examples of CNN-based traffic prediction include traffic volume prediction (Yao et al 2019;Deng et al 2019) and traffic speed prediction (Ma et al 2017;Jo et al 2018).…”

Section: Deep-learning Models For Itsmentioning

confidence: 99%

“…By converting networkwide traffic to the image-like data format, they constructed a time-space matrix with temporal and spatial traffic data and further employed CNNs to process the traffic images for feature extraction and network-wide traffic speed prediction. Similarly, Jo et al (2018) proposed image-to-image learning to predict traffic speed with a novel CNN model that consists of convolutional and deconvolutional filters.…”

Section: Deep Learning For Traffic Speed Predictionmentioning

confidence: 99%

Deep learning for intelligent traffic sensing and prediction: recent advances and future challenges

Fan

Xiang

Gong

et al. 2020

CCF Trans. Pervasive Comp. Interact.

View full text Add to dashboard Cite

With the emerging concepts of smart cities and intelligent transportation systems, accurate traffic sensing and prediction have become critically important to support urban management and traffic control. In recent years, the rapid uptake of the Internet of Vehicles and the rising pervasiveness of mobile services have produced unprecedented amounts of data to serve traffic sensing and prediction applications. However, it is significantly challenging to fulfill the computation demands by the big traffic data with ever-increasing complexity and diversity. Deep learning, with its powerful capabilities in representation learning and multi-level abstractions, has recently become the most effective approach in many intelligent sensing systems. In this paper, we present an up-to-date literature review on the most advanced research works in deep learning for intelligent traffic sensing and prediction.

show abstract

“…1), we are motivated to reuse building sensing data for nearby traffic sensing for the following reasons. Intuitively, most urban buildings are connected by roads, while residents move among different buildings, mainly via commuting on these roads [9]. According to the report in [10], the American citizens averagely spend over 93% of their daily time in enclosed buildings and vehicles (accounting for 87% and 6%, respectively).…”

Section: Introductionmentioning

confidence: 99%

BuildSenSys: Reusing Building Sensing Data for Traffic Prediction With Cross-Domain Learning

Fan

Xiang

Chen

et al. 2021

IEEE Trans. on Mobile Comput.

View full text Add to dashboard Cite

With the rapid development of smart cities, smart buildings are generating a massive amount of building sensing data by the equipped sensors. Indeed, building sensing data provides a promising way to enrich a series of data-demanding and cost-expensive urban mobile applications. In this paper, as a preliminary exploration, we study how to reuse building sensing data to predict traffic volume on nearby roads. Compared with existing studies, reusing building sensing data has considerable merits of cost-efficiency and high-reliability. Nevertheless, it is non-trivial to achieve accurate prediction on such cross-domain data with two major challenges. First, relationships between building sensing data and traffic data are not unknown as prior, and the spatio-temporal complexities impose more difficulties to uncover the underlying reasons behind the above relationships. Second, it is even more daunting to accurately predict traffic volume with dynamic building-traffic correlations, which are cross-domain, non-linear, and time-varying. To address the above challenges, we design and implement BuildSenSys, a first-of-its-kind system for nearby traffic volume prediction by reusing building sensing data. Our work consists of two parts, i.e., Correlation Analysis and Cross-domain Learning. First, we conduct a comprehensive building-traffic analysis based on multi-source datasets, disclosing how and why building sensing data is correlated with nearby traffic volume. Second, we propose a novel recurrent neural network for traffic volume prediction based on cross-domain learning with two attention mechanisms. Specifically, a cross-domain attention mechanism captures the building-traffic correlations and adaptively extracts the most relevant building sensing data at each predicting step. Then, a temporal attention mechanism is employed to model the temporal dependencies of data across historical time intervals. The extensive experimental studies demonstrate that BuildSenSys outperforms all baseline methods with up to 65.3% accuracy improvement (e.g., 2.2% MAPE) in predicting nearby traffic volume. We believe that this work can open a new gate of reusing building sensing data for urban traffic sensing, thus establishing connections between smart buildings and intelligent transportation.

show abstract

Image-to-Image Learning to Predict Traffic Speeds by Considering Area-Wide Spatio-Temporal Dependencies

Cited by 41 publications

References 32 publications

Deep Learning in Transport Studies: A Meta-analysis on the Prediction Accuracy

Deep Learning in Transport Studies: A Meta-analysis on the Prediction Accuracy

Deep learning for intelligent traffic sensing and prediction: recent advances and future challenges

BuildSenSys: Reusing Building Sensing Data for Traffic Prediction With Cross-Domain Learning

Contact Info

Product

Resources

About