Adaptive learning based improved performance of activation functions in hidden layer using artificial neural network

  • Authors

    • Tahir Khan Maulana Azad National Institute of Technology, BHOPAL MP INDIA
    • Dr J. S. Yadav Maulana Azad National Institute of Technology, BHOPAL MP INDIA
    2018-12-05
    https://doi.org/10.14419/ijet.v7i4.19905
  • Activation Function, Adaptive Learning, Back Propagation Neural Network, Hidden Layer, Learning Rate, Pattern Recognition, Time Complexity.

  • Abstract

    Advancement in Artificial Neural Network always playing vital role in complex pattern recognition system. In this research work im-proved performance of the activation functions for the hidden layer in artificial neural network supervised by backpropagation algorithm with adaptive learning feature has been recorded for pattern recognition. Complexity of large data of pattern such as face recognition, cancer detection, object recognition, number plate surveillance etc is increasing day by day. To resolve complexity, performance of hid-den layer is registered using Log-sigmoid, Tan-sigmoid and purelin activation functions respectively due to their inherent properties. An excellent neural network training model of 1460 Alpha-Numeric data set with 3000 Epoch (iterations) have been trained in neural net-work through activation functions for number plate recognition. Hence the performance efficiency of hidden layer activation functions is recorded for pruning the overall back propagation neural network architecture with improved learning rate along with better time com-plexity for pattern matching.

     

     

  • References

    1. [1] H. Bura, N. Lin, N. Kumar, S. Malekar, S. Nagaraj and K. Liu, "An Edge Based Smart Parking Solution Using Camera Networks and Deep Learning," 2018 IEEE International Conference on Cognitive Computing (ICCC), San Francisco, CA, USA, 2018, pp. 17-24.

      [2] R. Panahi and I. Gholampour, "Accurate Detection and Recognition of Dirty Vehicle Plate Numbers for High-Speed Applications," in IEEE Transactions on Intelligent Transportation Systems, vol. 18, no. 4, pp. 767-779, April 2017. https://doi.org/10.1109/TITS.2016.2586520.

      [3] Yo-Ping Huang, Shi-Yong Lai and Wei-Po Chuang, "A template-based model for license plate recognition," IEEE International Conference on Networking, Sensing and Control, 2004, 2004, pp. 737-742 Vol.2. https://doi.org/10.1109/ICNSC.2004.1297038.

      [4] K. A. Mohamed Junaid, “Classification Using Two Layer Neural Network Back Propagation Algorithm†Circuits and Systems, 2016, 7, 1207-1212. https://doi.org/10.4236/cs.2016.78104.

      [5] N. Wang, X. Zhu and J. Zhang, "License Plate Segmentation and Recognition of Chinese Vehicle Based on BPNN," 2016 12th International Conference on Computational Intelligence and Security (CIS), Wuxi, 2016, pp. 403-406.

      [6] JosephTarigan Nadia Ryanda Diedan and YayaSuryana " Plate Recognition Using Backpropagation Neural Network and Genetic Algorithm ," Science Direct Procedia Computer Science Volume 116, 2017, Pages 365-372 https://doi.org/10.1016/j.procs.2017.10.068.

      [7] Khan, Tahir; Dr J S Yadav, Dr; Dheeraj Agarwal, a Broad Survey on Performance Analysis of Number Plate Recognition from Stationary Images and Video Sequences. International Journal of Engineering & Technology, [S.l.], v. seven, n. 3.10, p. 164-168, july 2018. ISSN2227-524X.-https://www.sciencepubco.com/index.php/ijet /article/view/15652.

      [8] B. Ding, H. Qian and J. Zhou, "Activation functions and their characteristics in deep neural networks," 2018 IEEE Chinese Control and Decision Conference (CCDC), Shenyang, 2018, pp. 1836-1841. https://doi.org/10.1109/CCDC.2018.8407425.

      [9] B. V. Kakani, D. Gandhi and S. Jani, "Improved OCR based automatic vehicle number plate recognition using features trained neural network," IEEE 2017 8th International Conference on Computing, Communication and Networking Technologies (ICCCNT), Delhi, 2017, pp. 1-6 C. https://doi.org/10.1109/ICCCNT.2017.8203916.

      [10] Bircanoğlu and N. Arıca, "A comparison of activation functions in artificial neural networks," 2018 IEEE 26th Signal Processing and Communications Applications Conference (SIU), Izmir, 2018, pp. 1-4.

      [11] Y. Liu, Q. Yang, D. An, Y. Nai and Z. Zhang, "An improved fault diagnosis method based on deep wavelet neural network," 2018 IEEE Chinese Control and Decision Conference (CCDC), Shenyang, 2018, pp. 1048-1053. https://doi.org/10.1109/CCDC.2018.8407284.

      [12] G. Alpaydin, "An Adaptive Deep Neural Network for Detection, Recognition of Objects with Long Range Auto Surveillance," 2018 IEEE 12th International Conference on Semantic Computing (ICSC), Laguna Hills, CA, 2018, pp. 316-317.

      [13] X. Pang, H. Ma, P. Su and G. Tang, "TPPMA: New Adaptive BP Neural Network Based on PSO and PCA Algorithms," 2018 IEEE 27th International Symposium on Industrial Electronics (ISIE), Cairns, QLD, 2018, pp. 637-642.

  • Downloads

  • How to Cite

    Khan, T., & J. S. Yadav, D. (2018). Adaptive learning based improved performance of activation functions in hidden layer using artificial neural network. International Journal of Engineering & Technology, 7(4), 3879-3883. https://doi.org/10.14419/ijet.v7i4.19905

    Received date: 2018-09-19

    Accepted date: 2018-10-04

    Published date: 2018-12-05