Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2023 Mar 14:2023:1456971.
doi: 10.1155/2023/1456971. eCollection 2023.

Road-Type Classification with Deep AutoEncoder

Mohale E Molefe  1 , Jules R Tapamo  1
Affiliations

Road-Type Classification with Deep AutoEncoder

Mohale E Molefe et al. Comput Intell Neurosci. .

Abstract

Machine learning algorithms are among the driving forces towards the success of intelligent road network systems design. Such algorithms allow for the design of systems that provide safe road usage, efficient infrastructure, and traffic flow management. One such application of machine learning in intelligent road networks is classifying different road network types that provide useful traffic information to road users. We propose a deep autoencoder model for representation learning to classify road network types. Each road segment node is represented as a feature vector. Unlike existing graph embedding methods that perform road segment embedding using the neighbouring road segments, the proposed method performs embedding directly on the road segment vectors. The proposed method performs embedding directly on the road segment vectors. Comparison with state-of-the-art graph embedding methods show that the proposed method outperforms graph convolution networks, GraphSAGE-MEAN, graph attention networks, and graph isomorphism network methods, and it achieves similar performance to GraphSAGE-MAXPOOL.

PubMed Disclaimer

Conflict of interest statement

The authors declare that they have no conflicts of interest.

Figures

Figure 1
Figure 1
System diagram of the proposed method.
Figure 2
Figure 2
Line graph transformation: edges in the original graph are nodes in the transformed graph, and edges that share a node in the original graph become an edge in the transformed graph.
Figure 3
Figure 3
Illustration of the basic deep autoencoder model.
Figure 4
Figure 4
Application of the SVM classifier on two road-type classes.
Figure 5
Figure 5
Performance of DAE model at 5 hidden layers of size {49, 40, 31, 22, 13} on the encoder and decoder; and embedding space of size 4: (a) reconstruction error at increasing learning rates and (b) validation accuracy at increasing learning rates.
Figure 6
Figure 6
Performance of DAE model at 4 hidden layers of size {48, 38, 28, 18} on the encoder and decoder; and embedding space of size 8: (a) reconstruction error at increasing learning rates and (b) validation accuracy at increasing learning rates.
Figure 7
Figure 7
Performance of DAE model at 3 hidden layers of size {46, 34, 22} on the encoder and decoder; and embedding space of size 10: (a) reconstruction error at increasing learning rates and (b) validation accuracy at increasing learning rates.
Algorithm 1
Algorithm 1
Line graph transformation.
Algorithm 2
Algorithm 2
Feature generation for each road segment.
Algorithm 3
Algorithm 3
Feature embedding with deep autoencoder.
Algorithm 4
Algorithm 4
Road segment classification using DNN.
Algorithm 5
Algorithm 5
Road segment classification using SVM.
Algorithm 6
Algorithm 6
Road segment classification using K-NN.
See this image and copyright information in PMC

References

    1. Sahoo J., Rath M. Study and analysis of smart applications in smart city context. Proceedings of the 2017 International Conference on Information Technology (ICIT); December 2017; Bhubaneswar, India. pp. 225–228. - DOI
    1. Gharaee Z., Kowshik S., Stromann O., Felsberg M. Graph representation learning for road type classification. Pattern Recognition . 2021;120 doi: 10.1016/j.patcog.2021.108174.108174 - DOI
    1. Deekshetha H. R, Madhav A. V. S., Amit K. T. Evolutionary Computing and Mobile Sustainable Networks . Berlin, Germany: Springer; 2022. Traffic prediction using machine learning; pp. 969–983.
    1. Yin X., Wu G., Wei J., Shen Y., Qi H., Yin B. Deep learning on traffic prediction: methods, analysis, and future directions. IEEE Transactions on Intelligent Transportation Systems . jun 2022;23(6):4927–4943. doi: 10.1109/tits.2021.3054840. - DOI
    1. Vázquez J. J., Arjona J., Linares MaP., Casanovas-Garcia J. A comparison of deep learning methods for urban traffic forecasting using floating car data. Transportation Research Procedia . 2020;47:195–202. doi: 10.1016/j.trpro.2020.03.079. - DOI

LinkOut - more resources