An Efficient and Adaptive Method for Collision Probability of Ships, Icebergs Using CNN and DBSCAN Clustering Algorithm

  • Syed Zishan AliEmail author
  • Monica Makhija
  • Daljeet Choudhary
  • Hitesh Singh
Conference paper
Part of the Communications in Computer and Information Science book series (CCIS, volume 985)


Collision between ships and icebergs is a major problem in glacial area, where large to small icebergs becomes a threat to cargo ships, tankers, fishing ships etc. In this paper, we have devised a new approach for the detection of icebergs and movement of ships to predict their probability of collision. In this proposed work, an adaptive method is used to detect the presence of icebergs and the velocity of ships, followed by integrating the obtained data and applying the Bayesian algorithm we have successfully computed the collision probability. This work exhibits effective results against reduced visibility due to fog. Besides, we have acquired all the foreground authentic data from valid resources. So, the results will help in marking the safe and unsafe zones in the form of clusters by using DBSCAN algorithm.


Ships Icebergs Convolution neural network Collision probability Cluster 


  1. 1.
    Khan, A.A.: Why would sea-level rise for global warming and polar ice-melt, China University of Geosciences (Beijing and Peking University), Elsevier B.V.
  2. 2.
    Li, Y., Han, J., Yang. J.: Clustering moving objects, Department of Computer Science University of Illinois UrbanaChampaign. ACM (2004). 1-58113-888-1/04/0008Google Scholar
  3. 3.
    Wesche, C., Dierking, W.: From ice shelves to icebergs: classification of calving fronts, iceberg monitoring and drift simulation. In: 2014 IEEE Geoscience and Remote Sensing Symposium (2014).
  4. 4.
    Tiago, A.M., Silva, G.R.B.: Computer-based identification and tracking of Antarctic icebergs in SAR images, Department Street of Geography, Sheffield S10 2TN UK. Elsevier Scholar
  5. 5.
  6. 6.
    Phung, S.L., Bouzerdoum, A.: Matlab library for convolutional neural networks (2009)Google Scholar
  7. 7.
    Zhang, M.-L., Pena, J.M., Robles, V.: Feature selection for multi-label naive Bayes classification. Inf. Sci. 179(19), 3218–3229 (2009)CrossRefGoogle Scholar
  8. 8.
    Soman, K.P., Diwakar, S., Ajay, V.: Insight into data mining. PHI Publication (2009)Google Scholar
  9. 9.
  10. 10.
  11. 11.
  12. 12.
    Mittal, M., Goyal, L.M., Hemanth, D.J., Sethi, J.K.: Clustering approaches for high-dimensional databases: a review. WIREs Data Min. Knowl. Discov. (2019). John Wiley & Sons. Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2019

Authors and Affiliations

  • Syed Zishan Ali
    • 1
    Email author
  • Monica Makhija
    • 1
  • Daljeet Choudhary
    • 1
  • Hitesh Singh
    • 1
  1. 1.Bhilai Institute of Technology RaipurRaipurIndia

Personalised recommendations