Advertisement

An Efficient ATM Surveillance Framework Using Optical Flow with CNN

  • Ankit Bisht
  • Himanshu Singh Bisht
  • Vikas TripathiEmail author
Conference paper

Abstract

For our daily need, we often visit nearby automatic teller machine (ATM) to withdraw our deposited cash in bank. With increase in bank robberies, money snatching and attack on the customer visiting bank or ATM, our money is no longer safe. To prevent these types of abnormal activities, there is a need of better accuracy security surveillance system which can detect abnormal activities instantly. Our paper tends to work in this direction where we have used normal video, dense optical flow and Lucas–Kanade optical flow for detecting the abnormal motion of customer coming for withdrawal of money at ATM booth, with the help of convolution neural network we trained the model for single normal, multiple normal and multiple abnormal classes with different number of iteration to form 15 different training models. While testing those models, the average accuracy for Lucas–Kanade methods for 1000 iteration is 92.3984%.

Keywords

Optical flow Lucas–Kanade algorithm Dense optical flow Convolutional neural network ATM 

References

  1. 1.
  2. 2.
  3. 3.
    Hassan, H., Yaacob, S., Radman, A., Suandi, S.: Eye state detection for driver inattention based on Lucas Kanade optical flow algorithm. Intelligent and Advanced Systems (ICIAS). 6th International Conference on, pp. 1–6. IEEE, Kuala Lumpur, Malaysia (2016)Google Scholar
  4. 4.
    Vu, Q.D., Chung, S.T.: Real-time robust human tracking based on lucas-kanade optical flow and deep detection for embedded surveillance. In: 8th International Conference of Information and Communication Technology for Embedded Systems (IC-ICTES), pp. 1–6. IEEE, Chonburi, Thailand (2017)Google Scholar
  5. 5.
    Wang, Z., Yang, X.: Moving Target Detection and Tracking Based on Pyramid Lucas-Kanade Optical Flow. IEEE 3rd International Conference on Image. Vision and Computing (ICIVC), pp. 66–69. IEEE, Chongqing, China (2018)Google Scholar
  6. 6.
    Guo, Z., Wu, F., Chen, H., Yuan, J., Cai, C.: Pedestrian violence detection based on optical flow energy characteristics. Systems and Informatics (ICSAI). 4th International Conference on, pp. 1261–1265. IEEE, Hangzhou, China (2017)Google Scholar
  7. 7.
    Sakurada, K., Wang, W., Kawaguchi, N., Nakamura, R.: Dense Optical Flow based Change Detection Network Robust to Difference of Camera Viewpoints. arXiv:1712.02941 (2017)
  8. 8.
    Ghimire, D., Park, S., Kim, M.: Facial Expression Recognition Using Local Region Specific Dense Optical Flow and LBP Features. In: International Conference on Advanced Concepts for Intelligent Vision Systems, pp. 28–38. Springer, Cham (2017)CrossRefGoogle Scholar
  9. 9.
    Hu, B.: Convolutional neural network architectures for matching natural language sentences. Advances in neural information processing systems, pp. 2042–2050. (2014)Google Scholar
  10. 10.
    Schwarz, M., Schulz, H., Behnke, S.: RGB-D object recognition and pose estimation based on pre-trained convolutional neural network features. Robotics and Automation (ICRA). IEEE International Conference on, pp. 1329–1335. IEEE, Seattle, WA, USA (2015)Google Scholar
  11. 11.
    Cho, N., Kim, Y., Park, U., Park, S., Lee, S.: Group activity recognition with group interaction zone based on relative distance between human objects. Int. J. Pattern Recogn. Artificial Intell. 29(05), 1555007-(1–15) (2015)MathSciNetCrossRefGoogle Scholar
  12. 12.
    Zhuang, N., Yusufu, T., Ye, J., Hua, K.: Group activity recognition with differential recurrent convolutional neural networks. Automatic Face & Gesture Recognition (FG 2017). 12th IEEE International Conference on, pp. 526–531. IEEE, Washington DC, USA (2017)Google Scholar
  13. 13.
    Deng, Z., Zhai, M., Chen, L., Liu, Y., Muralidharan, S., Roshtkhari, M., Mori, G.: Deep structured models for group activity recognition. arXiv:1506.04191 (2015)
  14. 14.
    Lu, Y.: Towards unsupervised physical activity recognition using smartphone accelerometers. Multimedia Tools Appl 76(8), 10701–10719, Springer, US (2017)Google Scholar
  15. 15.
    Gupta, P., Dallas, T.: Feature selection and activity recognition system using a single triaxial accelerometer. In: IEEE Transactions on Biomedical Engineering, vol. 61(6), pp. 1780–1786. IEEE (2014)CrossRefGoogle Scholar
  16. 16.
    Amir, M., Kotef, H.: In-secure identities: on the securitization of abnormality. Environ. Planning D Soc. Space 36(2), 236–254 (2018)CrossRefGoogle Scholar
  17. 17.
    George, A.: Open CV pattern based smart bank security system with theft & identification using Android. Applied and Theoretical Computing and Communication Technology (iCATccT). 2nd International Conference on, pp. 254–257. IEEE, Bangalore, India (2016)Google Scholar
  18. 18.
    Tripathi, V., Gangodkar, D., Latta, V., Mittal, A.: Robust abnormal event recognition via motion and shape analysis at ATM installations. J. Electr. Comput. Eng. (2015)Google Scholar
  19. 19.
    Tripathi, V., Mittal, A., Gangodkar, A., Kanth, V.: Real time security framework for detecting abnormal events at ATM installations. J. Real-Time Image Process. 1–11 (2016)Google Scholar
  20. 20.
    Ashokan, V., Murthy, O.: Comparative evaluation of classifiers for abnormal event detection in ATMs. Intelligent Computing. In: Instrumentation and Control Technologies (ICICICT), 2017 International Conference on, pp. 1330–1333. IEEE, Kannur, India (2017)Google Scholar
  21. 21.
    Sanserwal, V., Tripathi, V., Chen, Z., Pandey, M.: Comparative analysis of various feature descriptors for efficient ATM surveillance framework. Computing. Communication and Automation (ICCCA), 2017 International Conference on, pp. 539–544. IEEE, Greater Noida, India (2017)Google Scholar
  22. 22.
    Thakur, B., Verma, B.: Design of an ATM security through smart-vision. In: International Conference on Intelligent Circuits and Systems (ICICS), pp. 133–137. IEEE, Phagwara, India (2018)Google Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2020

Authors and Affiliations

  • Ankit Bisht
    • 1
  • Himanshu Singh Bisht
    • 1
  • Vikas Tripathi
    • 1
    Email author
  1. 1.Graphic Era Deemed to Be UniversityDehradunIndia

Personalised recommendations