Advertisement

A Novel Genetically Optimized Convolutional Neural Network for Traffic Sign Recognition: A New Benchmark on Belgium and Chinese Traffic Sign Datasets

  • Arpan Jain
  • Apoorva MishraEmail author
  • Anupam Shukla
  • Ritu Tiwari
Article
  • 17 Downloads

Abstract

Traffic signs are a key constituent of the road network and prove to be very useful for warning and guiding the drivers. In intelligent transport systems, traffic sign recognition (TSR) is indispensable for autonomous driving. However, due to the complex outdoor environment, real-time recognition of traffic signs is much more challenging in comparison with many other pattern recognition tasks. Convolutional neural networks (CNNs) have an exceptional capability of recognizing patterns and are one of the most popular deep learning techniques. Finding the optimal configuration of a CNN for a task is a major challenge and is an active area of research. Genetic algorithm (GA) is a meta-heuristic approach well-known for its optimization power. In this paper, we propose a novel deep learning technique based on the concept of domain transfer learning for the recognition of traffic signs. This technique utilizes a newly proposed variant of the GA for finding the optimal values of the number of epochs and the learning rate parameter for each layer of the pre-trained CNN model (VGG-16). To examine the effectiveness of our technique, we apply it to the following two benchmark datasets of TSR: Belgium Traffic Sign Classification (BTSC) dataset and Chinese Traffic Sign Dataset (TT100K). The results indicate that our model outperforms all the existing approaches applied to these datasets and gives a new benchmark of the recognition accuracies of 99.16% for the BTSC and 96.28% for the TT100K datasets, thus establishing the robustness of our model.

Keywords

Traffic sign recognition Convolutional neural network Domain transfer learning Genetic algorithms Ternary crossover 

Notes

Compliance with Ethical Standards

Conflict of interest

The authors declare that they have no conflict of interest.

References

  1. 1.
    Lu Y, Yi S, Zeng N, Liu Y, Zhang Y (2017) Identification of rice diseases using deep convolutional neural networks. Neurocomputing 267:378–384.  https://doi.org/10.1016/j.neucom.2017.06.023 CrossRefGoogle Scholar
  2. 2.
    Zhang P, Niu X, Dou Y, Xia F (2017) Airport detection on optical satellite images using deep convolutional neural networks. IEEE Geosci Remote Sens Lett 14(8):1183–1187.  https://doi.org/10.1109/LGRS.2017.2673118 CrossRefGoogle Scholar
  3. 3.
    Roth HR, Lee CT, Shin HC, Seff A, Kim L, Yao J, Lu L, Summers RM (2015) Anatomy-specific classification of medical images using deep convolutional nets. In: IEEE 12th ISBI, pp 101–104. http://doi.org/10.1109/ISBI.2015.7163826
  4. 4.
    Li J, Zhang Z, He H (2017) Hierarchical convolutional neural networks for EEG-based emotion recognition. Cogn Comput.  https://doi.org/10.1007/s12559-017-9533-x Google Scholar
  5. 5.
    Lopes UK, Valiati JF (2017) Pre-trained convolutional neural networks as feature extractors for tuberculosis detection. Comput Biol Med.  https://doi.org/10.1016/j.compbiomed.2017.08.001 Google Scholar
  6. 6.
    Phan HTH, Kumar A, Kim J, Feng D (2016) Transfer learning of a convolutional neural network for HEp-2 cell image classification. In: IEEE 13th ISBI, pp 1208–1211.  https://doi.org/10.1109/ISBI.2016.7493483
  7. 7.
    Liu W, Wang Z, Liu X, Zeng N, Liu Y, Alsaadi FE (2017) A survey of deep neural network architectures and their applications. Neurocomputing 234:11–26.  https://doi.org/10.1016/j.neucom.2016.12.038 CrossRefGoogle Scholar
  8. 8.
    Basagoiti MBR, Rodriguez IRV (2016) A modified genetic algorithm applied to the elevator dispatching problem. Soft Comput.  https://doi.org/10.1007/s00500-015-1718-1 Google Scholar
  9. 9.
    Hsu C, Cho H (2015) A genetic algorithm for the maximum edge-disjoint paths problem. Neurocomputing.  https://doi.org/10.1016/j.neucom.2012.10.046 Google Scholar
  10. 10.
    Shih CC, Horng MF, Pan TS, Pan JS, Chen CY (2016) A genetic-based effective approach to path-planning of autonomous underwater glider with upstream-current avoidance in variable oceans. Soft Computing.  https://doi.org/10.1007/s00500-016-2122-1 Google Scholar
  11. 11.
    Goldberg D (1989) Genetic algorithms in search, optimization, and machine learning. Addison-Wesley, ReadingzbMATHGoogle Scholar
  12. 12.
    Mishra A, Shukla A (2017) Mathematical analysis of the cumulative effect of novel ternary crossover operator and mutation on probability of survival of a schema. Theoret Comput Sci 666:1–11.  https://doi.org/10.1016/j.tcs.2016.07.035 MathSciNetCrossRefzbMATHGoogle Scholar
  13. 13.
    Uzor CJ, Gongora M, Coupland S, Passow BN (2016) Adaptive-mutation compact genetic algorithm for dynamic environments. Soft Comput 8:3097–3115.  https://doi.org/10.1007/s00500-016-2195-x CrossRefGoogle Scholar
  14. 14.
    Qiongbing Z, Lixin D (2016) A new crossover mechanism for genetic algorithms with variable-length chromosomes for path optimization problems. Expert Syst Appl 60:183–189.  https://doi.org/10.1016/j.eswa.2016.04.005 CrossRefGoogle Scholar
  15. 15.
    Banerjee A (2013) A novel probabilistically-guided context-sensitive crossover operator for clustering. Swarm Evol Comput 13:47–62.  https://doi.org/10.1016/j.swevo.2013.05.007 CrossRefGoogle Scholar
  16. 16.
    Mishra A, Shukla A (2018) Mathematical analysis of schema survival for genetic algorithms having dual mutation. Soft Comput 22(6):1763–1771.  https://doi.org/10.1007/s00500-017-2650-3 CrossRefzbMATHGoogle Scholar
  17. 17.
    Hu W, Zhuo Q, Zhang C, Li J (2017) Fast branch convolutional neural network for traffic sign recognition. IEEE Intell Transp Syst Mag 9(3):114–126.  https://doi.org/10.1109/MITS.2017.2709780 CrossRefGoogle Scholar
  18. 18.
    Li Y, Møgelmose A, Trivedi MM (2016) Pushing the “Speed Limit”: high-accuracy US traffic sign recognition with convolutional neural networks. IEEE Trans Intell Veh 1(2):167–176.  https://doi.org/10.1109/TIV.2016.2615523 CrossRefGoogle Scholar
  19. 19.
    Arcos-García Á, Álvarez-García JA, Soria-Morillo LM (2018) Deep neural network for traffic sign recognition systems: an analysis of spatial transformers and stochastic optimisation methods. Neural Netw 99:158–165.  https://doi.org/10.1016/j.neunet.2018.01.005 CrossRefGoogle Scholar
  20. 20.
    Yang T, Long X, Sangaiah AK, Zheng Z, Tong C (2018) Deep detection network for real-life traffic sign in vehicular networks. Comput Netw 136:95–104.  https://doi.org/10.1016/j.comnet.2018.02.026 CrossRefGoogle Scholar
  21. 21.
    Kassani PH, Teoh ABJ (2017) A new sparse model for traffic sign classification using soft histogram of oriented gradients. Appl Soft Comput 52:231–246.  https://doi.org/10.1016/j.asoc.2016.12.037 CrossRefGoogle Scholar
  22. 22.
    Huang Z, Yu Y, Gu J, Liu H (2017) An efficient method for traffic sign recognition based on extreme learning machine. IEEE Trans Cybern 47(4):920–933.  https://doi.org/10.1109/TCYB.2016.2533424 CrossRefGoogle Scholar
  23. 23.
    Yu Y, Li J, Wen C, Guan H, Luo H, Wang C (2016) Bag-of-visual-phrases and hierarchical deep models for traffic sign detection and recognition in mobile laser scanning data. ISPRS J Photogr Remote Sens 113:106–123.  https://doi.org/10.1016/j.isprsjprs.2016.01.005 CrossRefGoogle Scholar
  24. 24.
    Jurisic F, Filković I, Kalafatić Z (2015) Multiple-dataset traffic sign classification with OneCNN. In: 2015 3rd IAPR Asian conference on pattern recognition (ACPR). IEEE, pp 614–618.  https://doi.org/10.1109/ACPR.2015.7486576
  25. 25.
    Mathias M, Timofte R, Benenson R, Van Gool L (2013) Traffic sign recognition—how far are we from the solution? In: The 2013 international joint conference on neural networks (IJCNN). IEEE, pp 1–8.  https://doi.org/10.1109/IJCNN.2013.6707049
  26. 26.
    Lu K, Ding Z, Ge S (2012) Sparse-representation-based graph embedding for traffic sign recognition. IEEE Trans Intell Transp Syst 13(4):431, 1515–1524.  https://doi.org/10.1109/TITS.2012.2220965
  27. 27.
    Madani A, Yusof R (2017) Traffic sign recognition based on color, shape, and pictogram classification using support vector machines. Neural Comput Appl 30:2807–2817.  https://doi.org/10.1007/s00521-017-2887-x CrossRefGoogle Scholar
  28. 28.
    Yang Y, Luo H, Xu H, Wu F (2016) Towards real-time traffic sign detection and classification. IEEE Trans Intell Transp Syst 17(7):2022–2031.  https://doi.org/10.1109/TITS.2015.2482461 CrossRefGoogle Scholar
  29. 29.
    Yin S, Deng J, Zhang D, Du J (2017) Traffic sign recognition based on deep convolutional neural network. In: CCF Chinese conference on computer vision. Springer, Singapore, pp 685–695.  https://doi.org/10.1007/978-981-10-7299-4_57
  30. 30.
    Pan SJ, Yang Q (2010) A survey on transfer learning. IEEE Trans Knowl Data Eng 22:1345–1359.  https://doi.org/10.1109/TKDE.2009.191 CrossRefGoogle Scholar
  31. 31.
    Ahn E, Kumar A, Kim J, Li C, Feng D, Fulham M (2006) X-ray image classification using domain transferred convolutional neural networks and local sparse spatial pyramid. In: IEEE 13th ISBI, pp 855–858.  https://doi.org/10.1109/ISBI.2016.7493400
  32. 32.
    Protopapadakis E, Schauer M, Pierri E, Doulamis AD, Stavroulakis GE, Böhrnsen JU, Langer S (2016) A genetically optimized neural classifier applied to numerical pile integrity tests considering concrete piles. Comput Struct 162:68–79.  https://doi.org/10.1016/j.compstruc.2015.08.005 CrossRefGoogle Scholar
  33. 33.
    Xie L, Yuille A (2017) Genetic CNN. In: ICCV, pp 1388–1397Google Scholar
  34. 34.
    Young SR, Rose DC, Karnowsky TP, Lim S-H, Patton RM (2015) Optimizing deep learning hyper-parameters through an evolutionary algorithm. In: Proceedings of the workshop on machine learning in high-performance computing environments.  https://doi.org/10.1145/2834892.2834896
  35. 35.
    Loshchilov I, Hutter F (2016) CMA-ES for hyperparameter optimization of deep neural networks. arXiv preprint arXiv:1604.07269
  36. 36.
    Simonyan K, Zisserman A (2014) Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556
  37. 37.
    Zhu Z, Liang D, Zhang S, Huang X, Li B, Hu S (2016) Traffic-sign detection and classification in the wild. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 2110–2118Google Scholar
  38. 38.
    Timofte R, Van Gool L (2011) Sparse representation based projections. In: Proceedings of the 22nd British machine vision conference-BMVC 2011. BMVA Press, pp 61-1. http://dx.doi.org/10.5244/C.25.61

Copyright information

© Springer Science+Business Media, LLC, part of Springer Nature 2019

Authors and Affiliations

  1. 1.Department of Computer Science and EngineeringOhio State UniversityColumbusUSA
  2. 2.Department of Computer Science and EngineeringBennett UniversityGreater NoidaIndia
  3. 3.Soft Computing and Expert System LaboratoryABV-IIITMGwaliorIndia

Personalised recommendations