Abstract: This paper presents a novel approach to solving the problem of poor learning data in complex object classification task. It efficiently combines the Visual Content Analysis technique known as the Scalable Vocabulary Tree (SVT) and contour-based descriptors to recommend new training samples. The SVT technique uses the SIFT features to identify and accurately localize objects of interest within the visual content of the processed query images. Despite the small learning data set its classification accuracy is pretty good and matches the accuracy of a dedicated CNN network trained under the same conditions. However, due to the ability of fast and effective incremental learning, it overcomes the convnet type networks. Contour-based classification based on Point Distance Histogram (PDH) is utilized then to increase the classification certainty. During this stage, the PDH descriptors representing a given object of interest are matched against descriptors stored in the pattern database, where each object is represented by a collection of 360 pattern outlines extracted from its 3D model. As finally reported, such an exact pattern representation allows for achieving a high classification accuracy of the entire approach.
B I B L I O G R A F I A1. http://www.live-counter.com/how-big-is-the-internet/. Accessed 26 Feb 2020
2. Taigman, Y., Yang, M., Ranzato, M., Wolf, L.: DeepFace: closing the gap to human-level performance in face verification. In: 2014 IEEE Conference on Computer Vision and Pattern Recognition, Columbus, OH, pp. 1701–1708 (2014). https://doi.org/10.1109/cvpr.2014.220
3. Chollet, F.: Deep Learning with Python. Manning Publications Co., New York (2018)
4. Worring, M., Snoek, C.: Visual content analysis. In: Liu, L., Özsu, M.T. (eds.) Encyclopedia of Database Systems, pp 3360–3365. Springer, Boston (2009). https://doi.org/10.1007/978-0-387-39940-9_1019
5. Baran, R., Zeja, A.: The IMCOP system for data enrichment and content discovery and delivery. In: Proceedings of the 2015 International Conference on Computational Science and Computational Intelligence (CSCI 2015), pp 143–146, Las Vegas, USA (2015)
6. Wolff, E.: Microservices: Flexible Software Architectures. Addison-Wesley, Boston (2016)
7. Li, Z., Hoiem, D.: Learning without forgetting. PAMI 40, 2935–2947 (2018)
8. Tao, Y., Tu Y., Shyu, M..: Efficient incremental training for deep convolutional neural networks. In: 2019 IEEE Conference on Multimedia Information Processing and Retrieval (MIPR), San Jose, CA, USA, pp. 286–291 (2019)
9. Cheng, Y., Wang, D., Zhou, P., Zhang, T.: A survey of model compression and acceleration for deep neural networks, CoRR, vol. abs/1710.09282 (2017)
10. Roy, D., Panda, P., Roy, K.: Tree-CNN: a hierarchical deep convolutional neural network for incremental learning. Neural Netw. 121, 148–160 (2018)
11. Nister, D., Stewenius, H.: Scalable recognition with a vocabulary tree. In: Conference on Computer Vision and Pattern Recognition, New York, NY, USA, pp. 2161–2168 (2006)
12. Baran, R.: Efficiency investigation of BoF, SVT and pyramid match algorithms in practical recognition applications. In: Proceedings of the 2017 IEEE International Conference on Mathematics and Computers in Sciences and in Industry (MCSI), pp 171–178, Corfu Island, Greece (2017)
13. Lowe, D.G.: Object recognition from local scale-invariant features. In: Proceedings of the ICCV 1999, vol 2, pp 1150–1157. IEEE Computer Society (1999)
14. Rublee, E., Rabaud, V., Konolige, K., Bradski, G. R.: ORB: an efficient alternative to SIFT or SURF. In: ICCV 2011, pp. 2564–2571 (2011)
15. Baran, R., Rudziński, F., Zeja, A.: Face recognition for movie character and actor discrimination based on similarity scores. In: Proceedings of the 2016 IEEE International Conference on Computational Science and Computational Intelligence (CSCI), pp 1333–1338, Las Vegas, USA (2016)
16. Rother, C., Kolmogorov, V., Blake, A.: GrabCut - interactive foreground extraction using iterated graph cuts. ACM Trans. Graph. 23(3), 309–314 (2004)
17. Otsu, N.: A threshold selection method from gray-level histograms. IEEE Trans. Syst. Man Cybern. 9(1), 62–66 (1979)
18. Baran, R., Kleszcz, A.: The efficient spatial methods of contour approximation. In: Proceedings of the 2014 IEEE International Conference on Signal Processing: Algorithms, Architectures, Arrangements, and Applications (SPA 2014), pp. 116–121, Poznań, Poland (2014)
19. Frejlichowski, D.: Shape representation using point distance histogram. Polish J. Environ. Stud. 16(4A), 90–93 (2007)
20. Frejlichowski, D.: application of the point distance histogram to the automatic identification of people by means of digital dental radiographic images. In: Chmielewski, L.J., Datta, A., Kozera, R., Wojciechowski, K. (eds.) ICCVG 2016. LNCS, vol. 9972, pp. 387–394. Springer, Cham (2016). https://doi.org/10.1007/978-3-319-46418-3_34 
Pełny tekst 
DOI