Proceedings of International Conference on Applied Innovation in IT
2021/04/28, Volume 9, Issue 1, pp.55-60

Influence of Synthetic Image Datasets on the Result of Neural Networks for Object Detection


Aleksandr Kniazev, Pavel Slivnitsin, Leonid Mylnikov, Stefan Schlechtweg, Andrey Kokoulin


Abstract: The goal of the article is research of ways to improve the quality of neural networks object detection. To achieve this goal we suggest to use synthetic image datasets. The algorithm of generating synthetic images, which uses the environment of the detected object, is described in the article. That algorithm could be applied in the control algorithm of the robotic system for luminaire replacement that is based on target object detection. 3D models and 3D camera images of detected objects, backgrounds, noise objects and different effects are used to create realistic images that will increase the quality of predictions. Quality tests were made with synthetic and real datasets. Results show that quality could be increased up to 16%. Ratio of real and synthetic data is 1:4.

Keywords: Image Recognition, Object Detection, Neural Network, Synthetic Dataset, Data Generation

DOI: 10.25673/36584

Download: PDF

References:

  1. L. Mylnikov, “Statistical methods of intelligent data analysis,” St. Petersburg: BHV-Petersburg, 2021, 240 p.
  2. Z. Zou, Z. Shi, Y. Guo, and J. Ye, “Object Detection in 20 Years: A Survey,” arXiv, pp. 1-39, May 2019.
  3. P. Slivnitsin, A. Bachurin, and L. Mylnikov, “Robotic system position control algorithm based on target object recognition, ” in Proceedings of International Conference on Applied Innovation in IT, vol. 8, no. 1, pp. 87-94, 2020.
  4. P. A. Slivnitsin and A. A. Bachurin, “A modern way of outdoor lighting maintenance, ” in Journal of Physics: Conference Series, vol. 1415, no. 1, 2019.
  5. T. Anwar, “Training a Custom Object Detector with DLIB & Making Gesture Controlled Applications,” 2020 [Online]. Available: https://www.learnopencv.com/training-a-custom- object-detector-with-dlib-making-gesture-controlled- applications/ [Accessed: 07-Dec-2020].
  6. J. Li, P. L. Götvall, J. Provost, and K. Åkesson, “Training Convolutional Neural Networks with Synthesized Data for Object Recognition in Industrial Manufacturing,” IEEE Int. Conf. Emerg. Technol. Fact. Autom. ETFA, vol. 2019-Septe, pp. 1544-1547, 2019.
  7. M. Andulkar, J. Hodapp, T. Reichling, M. Reichenbach, and U. Berger, “Training CNNs from Synthetic Data for Part Handling in Industrial Environments,” IEEE Int. Conf. Autom. Sci. Eng., vol. 2018-Augus, pp. 624–629, 2018.
  8. D. Mas Montserrat, Q. Lin, J. P. Allebach, and E. J. Delp, “Scalable Logo Detection and Recognition with Minimal Labeling,” Proc. - IEEE 1st Conf. Multimed. Inf. Process. Retrieval, MIPR 2018, pp. 152-157, 2018.
  9. G. Volk, S. Muller, A. Von Bernuth, D. Hospach, and O. Bringmann, “Towards Robust CNN-based Object Detection through Augmentation with Synthetic Rain Variations,” 2019 IEEE Intell. Transp. Syst. Conf. ITSC 2019, pp. 285-292, 2019.
  10. G. Georgakis, A. Mousavian, A. C. Berg, and J. Košecká, “Synthesizing training data for object detection in indoor scenes,” Robot. Sci. Syst., vol. 13, 2017.
  11. H. Rezatofighi, N. Tsoi, J. Gwak, A. Sadeghian, I. Reid, and S. Savarese, “Generalized intersection over union: A metric and a loss for bounding box regression,” in Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2019, vol. 2019-June, pp. 658-666.
  12. G. T. U. A. Colleges, et al., “Microsoft COCO,” Eccv, no. June, pp. 740-755, 2014.
  13. M. Everingham, L. Van Gool, C. K. I. Williams, J. Winn, and A. Zisserman, “The pascal visual object classes (VOC) challenge,” Int. J. Comput. Vis., vol. 88, no. 2, pp. 303-338, 2010.
  14. J. Redmon and A. Farhadi, “YOLO v.3,” Tech Rep., pp. 1-6, 2018.
  15. W. He, Z. Huang, Z. Wei, C. Li, and B. Guo, “TF- YOLO: An improved incremental network for real- time object detection,” Appl. Sci., vol. 9, no. 16, 2019.
  16. P. G. Doyle, “Grinstead and Snell's Introduction to Probability,” 2006, American Mathematical Society. 518 p
  17. J. Redmon and A. Farhadi, “YOLO: Real-Time Object Detection,” 2018 [Online]. Available: https://pjreddie.com/darknet/yolo/. [Accessed: 25-Nov-2020].

    Home

    PARTICIPATION

       - Call for Papers
       - Paper Submission
       - Important Dates
       - Committee
       - Guest registration


    PROCEEDINGS

       - Volume 9, Issue 1 (ICAIIT 2021)
       - Volume 8, Issue 1 (ICAIIT 2020)
       - Volume 7, Issue 1 (ICAIIT 2019)
       - Volume 7, Issue 2 (ICAIIT 2019)
       - Volume 6, Issue 1 (ICAIIT 2018)
       - Volume 5, Issue 1 (ICAIIT 2017)
       - Volume 4, Issue 1 (ICAIIT 2016)
       - Volume 3, Issue 1 (ICAIIT 2015)
       - Volume 2, Issue 1 (ICAIIT 2014)
       - Volume 1, Issue 1 (ICAIIT 2013)


    PAST CONFERENCES

       ICAIIT 2021
         - Photos
         - Reports

       ICAIIT 2020
         - Photos
         - Reports

       ICAIIT 2019
         - Photos
         - Reports

       ICAIIT 2018
         - Photos
         - Reports

    ETHICS IN PUBLICATIONS

    ACCOMODATION

    CONTACT US

 


           ISSN 2199-8876
           Copyright © 2013-2021 Leonid Mylnikov. All rights reserved.