Developing plastic recycling classifier by deep learning and directed acyclic graph residual network

Authors

DOI:

https://doi.org/10.15587/1729-4061.2022.254285

Keywords:

Directed Acyclic Graph (DAG), deep learning, Recycling, classification, Convolutional Neural Network (CNN)

Abstract

Recycling is one of the most important approaches to safeguard the environment since it aims to reduce waste in landfills while conserving natural resources. Using deep Learning networks, this group of wastes may be automatically classified on the belts of a waste sorting plant. However, a basic set of connected layers may not be adequate to give satisfactory accuracy for such multi output classifier tasks. To optimize the gradient flow and enable deeper training for network design with multi label classifier, this study suggests a residual-based deep learning convolutional neural network. For network training, ten classes have been explored. The Directed Acyclic Graph (DAG) is a structure with hidden layers that have inputs, outputs, and other layers. The DAG network's residual-based architecture features shortcut connections that bypass some levels of the network, allowing gradients of network parameters to travel freely among the network output layers for deeper training. The methodology includes:

1) preparing the data and creating an augmented image data store;

2) defining the main serially-connected branches of the network architecture;

3) defining the residual interconnections that bypass the main branch layers;

4) defining layers, and finally;

5) creating a residual-based deeper layer graph.

The concept is to split down the multiclass classification problem into minor binary states, where every classifier performs as an expert by concentrating on discriminating between only two labels, improving total accuracy. The results achieve (2.861 %) training error and (9.76 %) a validation error. The training results of this classifier are evaluated by finding the training error, validation error, and showing the confusion matrix of validation data

Author Biographies

Ahmed Burhan Mohammed, University of Kirkuk

Lecturer

Department of Media

Ahmad Abdullah Mohammed Al-Mafrji, University of Kirkuk

Department of Mathematics

Moumena Salah Yassen, University of Kirkuk

Assistant Lecturer

Department of Software

Ahmad H. Sabry, Universiti Tenaga Nasional

Doctor of Control and Automation Engineering

Department ofInstitute of Sustainable Energy

References

  1. Wong, K. C. L., Moradi, M. (2019). SegNAS3D: Network Architecture Search with Derivative-Free Global Optimization for 3D Image Segmentation. Medical Image Computing and Computer Assisted Intervention – MICCAI 2019, 393–401. doi: https://doi.org/10.1007/978-3-030-32248-9_44
  2. Taheri, S., Toygar, Ö. (2019). On the use of DAG-CNN architecture for age estimation with multi-stage features fusion. Neurocomputing, 329, 300–310. doi: https://doi.org/10.1016/j.neucom.2018.10.071
  3. Foahom Gouabou, A. C., Damoiseaux, J.-L., Monnier, J., Iguernaissi, R., Moudafi, A., Merad, D. (2021). Ensemble Method of Convolutional Neural Networks with Directed Acyclic Graph Using Dermoscopic Images: Melanoma Detection Application. Sensors, 21 (12), 3999. doi: https://doi.org/10.3390/s21123999
  4. Golrizkhatami, Z., Taheri, S., Acan, A. (2018). Multi-scale features for heartbeat classification using directed acyclic graph CNN. Applied Artificial Intelligence, 32 (7-8), 613–628. doi: https://doi.org/10.1080/08839514.2018.1501910
  5. Agarwal, N., Balasubramanian, V. N., Jawahar, C. V. (2018). Improving multiclass classification by deep networks using DAGSVM and Triplet Loss. Pattern Recognition Letters, 112, 184–190. doi: https://doi.org/10.1016/j.patrec.2018.06.034
  6. Austin, A. E., Desrosiers, T. A., Shanahan, M. E. (2019). Directed acyclic graphs: An under-utilized tool for child maltreatment research. Child Abuse & Neglect, 91, 78–87. doi: https://doi.org/10.1016/j.chiabu.2019.02.011
  7. Anilkumar, K. K., Manoj, V. J., Sagi, T. M. (2021). Automated detection of leukemia by pretrained deep neural networks and transfer learning: A comparison. Medical Engineering & Physics, 98, 8–19. doi: https://doi.org/10.1016/j.medengphy.2021.10.006
  8. Zhu, T., Cao, C., Wang, Z., Xu, G., Qiao, J. (2020). Anatomical Landmarks and DAG Network Learning for Alzheimer’s Disease Diagnosis. IEEE Access, 8, 206063–206073. doi: https://doi.org/10.1109/access.2020.3037107
  9. Oyewola, D. O., Augustine, A. F. (2021). Predicting Impact of COVID-19 on Crude Oil Price Image with Directed Acyclic Graph Deep Convolution Neural Network. Journal of Robotics and Control (JRC), 2 (2). doi: https://doi.org/10.18196/jrc.2261
  10. Pachón-Suescún, C. G., Pinzón-Arenas, J. O., Jiménez-Moreno, R. (2020). Fruit Identification and Quality Detection by Means of DAG-CNN. International Journal on Advanced Science, Engineering and Information Technology, 10 (5), 2183. doi: https://doi.org/10.18517/ijaseit.10.5.8684
  11. Plastic Waste DataBase of Images – WaDaBa. Available at: http://wadaba.pcz.pl/
  12. Jwaid, W. M., Al-Husseini, Z. S. M., Sabry, A. H. (2021). Development of brain tumor segmentation of magnetic resonance imaging (MRI) using U-Net deep learning. Eastern-European Journal of Enterprise Technologies, 4 (9 (112)), 23–31. doi: https://doi.org/10.15587/1729-4061.2021.238957

Downloads

Published

2022-04-30

How to Cite

Mohammed, A. B., Al-Mafrji, A. A. M., Yassen, M. S., & Sabry, A. H. (2022). Developing plastic recycling classifier by deep learning and directed acyclic graph residual network . Eastern-European Journal of Enterprise Technologies, 2(10 (116), 42–49. https://doi.org/10.15587/1729-4061.2022.254285