Object detection in high resolution images based on multiscale and block processing
https://doi.org/10.37661/1816-0301-2020-17-2-7-16
Abstract
In the paper the algorithm for object detection in high resolution images is proposed. The approach uses multiscale image representation followed by block processing with the overlapping value. For each block the object detection with convolutional neural network was performed. Number of pyramid layers is limited by the Convolutional Neural Network layer size and input image resolution. Overlapping blocks splitting to improve the classification and detection accuracy is performed on each layer of pyramid except the highest one. Detected areas are merged into one if they have high overlapping value and the same class. Experimental results for the algorithm are presented in the paper.
About the Authors
R. P. Bohush
Polotsk State University
Belarus
Belarus
Rykhard P. Bohush, Cand. Sci. (Eng.), Associate Professor, Head of the Department of Computer Systems and Networks
Novopolotsk
I. Yu. Zakharava
Polotsk State University
Belarus
Belarus
Iryna Yu. Zakharava, M. Sci. (Eng.), Postgraduate Student at the Department of Computer Systems and Networks
Novopolotsk
S. V. Ablameyko
Belarusian State University; The United Institute of Informatics Problems of the National Academy of Sciences of Belarus
Belarus
Belarus
Sergey V. Ablameyko, Academician of the National Academy of Sciences of Belarus, Dr. Sci. (Eng.), Professor, Professor of the Faculty of Mechanics and Mathematics
Minsk
References
1. Dvorkovich A. V., Dvorkovich V. P. Metrologicheskoe obespechenie videoinformatsionnykh system. Metrological Support of Video Information Systems, Moscow, Technosphera, 2015, 784 p. (in Russian).
2. Goulekas K. Visual Effects in a Digital World: a Comprehensive Glossary of over 7,000 Visual Effects Terms. San Francisco, Morgan Kaufmann, 2001, 600 p.
3. Vorobjov D., Zakharova I., Bohush R., Ablameyko S. An effective object detection algorithm for high resolution video by using convolutional neural network. Advances in Neural Networks-ISNN2018. Lecture Notes in Computer Science, 2018, vol. 10878, pp. 503–510. https://doi.org/10.1007/978-3-319-92537-0_58
4. Yongxi L., Javidi T. Efficient object detection for high resolution images. Proceedings of 53 rd Annual Allerton Conference on Communication, Control, and Computing (Allerton), Monticello, IL, USA, 30 September – 2 October 2015. Monticello, 2015, pp. 1091–1098. https://doi.org/10.1109/ALLERTON.2015.7447130
5. Xiao J., Hays J., Ehinger K., Oliva A., Torralba A. Sun database: large-scale scene recognition from abbey to zoo. Proceedings of IEEE Conference on Computer Vision and Pattern Recognition (CVPR), San Francisco, CA, USA, 13–18 June 2010. San Francisco, 2010, pp. 3485–3492. https://doi.org/10.1109/CVPR.2010.5539970
6. Ruzicka V., Franchetti F. Fast and accurate object detection in high resolution 4K and 8K video using GPUs. Proceedings of 2018 IEEE High Performance Extreme Computing Conference (HPEC), Waltham, MA, USA, 25–27 September 2018. Waltham, 2018, pp. 1–7. https://doi.org/10.1109/HPEC.2018.8547574
7. Korshunov P., Ebrahimi T. UHD video dataset for evaluation of privacy. Proceedings of Sixth International Workshop on Quality of Multimedia Experience (QoMEX), Singapore, 18–20 September 2014. Singapore, 2014, pp. 232–237. https://doi.org/10.1109/QoMEX.2014.6982324
8. Unel F. O., Ozkalayci B., Çigla C. The power of tiling for small object detection. CVPR Workshops, 2019. Available at: http://openaccess.thecvf.com/content_CVPRW_2019/papers/UAVision/Unel_The_Power_of_Tiling_for_Small_Object_Detection_CVPRW_2019_paper.pdf. (accessed 18.01.2020).
9. Girshick R., Donahue J., Darrell T., Malik J. Region-based convolutional networks for accurate object detection and segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2016, vol. 38, pp. 142–158. https://doi.org/10.1109/TPAMI.2015.2437384
10. He K., Zhang X., Ren S., Sun J. Deep residual learning for image recognition. Proceedings of IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2016), Las Vegas, NV, USA, 27–30 June 2016. Las Vegas, 2016, pp. 770–778. https://doi.org/10.1109/CVPR.2016.90
11. Redmon J., Divvala S. K., Girshick R. B., Farhadi A. You only look once: unified, real-time object detection. Proceedings of IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2016), Las Vegas, NV, USA, 27–30 June 2016. Las Vegas, 2016, pp.779–788. https://doi.org/10.1109/CVPR.2016.91
12. Girshick R. Fast R-CNN. Proceedings of IEEE Intern. Conf. on Computer Vision (ICCV), Santiago, Chile, 11–18 December 2015. Santiago, 2015, pp. 1440–1448. https://doi.org/10.1109/ICCV.2015.169
13. Ren S., He K., Girshick R., Sun J. Faster R-CNN: towards real-time object detection with region proposal networks. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015, vol. 39, no. 6, pp. 1137–1149. https://doi.org/10.1109/TPAMI.2016.2577031
14. Kroshchenko A., Golovko V., Bezobrazov S., Mikhno E., Khatskevich M, …, Brich A. Glubokoe obuchenie dlia detektirovaniia obieektov na izobrazheniiakh dokumentov [Deep training for detecting of objects at images of documents]. Vestnik Brestskogo gosudarstvennogo tekhnicheskogo universiteta. Fizika, matematika, informatika [Bulletin of the Brest State Technical University. Physics, mathematics, Computer Science], 2017, vol. 5 (107), pp. 2–9 (in Russian).
15. Szegedy C., Ioffe S., Vanhoucke V. Inception-v4, inception-ResNet and the impact of residual connections on learning. Proceedings of the Thirty-First AAAI Conference on Artificial Intelligence (AAAI-17), San Francisco, California, USA, 4–9 February 2017. San Francisco, 2017, pp. 4278–4284.
16. Everingham M., Van Gool L., Williams C., Winn J., Zisserman A. The pascal Visual Object Classes (VOC) challenge. International Journal of Computer Vision, 2010, vol. 88, pp. 303–338. https://doi.org/10.1007/s11263-009-0275-4
Review
For citations:
Bohush R.P., Zakharava I.Yu., Ablameyko S.V. Object detection in high resolution images based on multiscale and block processing. Informatics. 2020;17(2):7-16. (In Russ.) https://doi.org/10.37661/1816-0301-2020-17-2-7-16
Views: 1344
Email this article 