IMAGE RECOGNITION SYSTEM BASED ON A NEURAL NETWORK WITH DEEP LEARNING
Keywords:
neural network, deep learning network, feature classifier, support vector, graphic frame, machine visionAbstract
An object recognition system based on deep learning neural networks has been developed. Such an image recognition system is capable of accurately and quickly recognizing images known to it and similar to them in video content obtained from IP surveillance cameras. Depending on the video shooting conditions and the viewing angle of the IP camera, the algorithm of this system achieves a recognition accuracy of 96.38%. This image recognition rate is practically constant for 11 classes of identification and recognition objects. Such high results are achieved through the use of the CamVid video database as a training sample for the neural network. This database is based on 421 training and 280 test video images. The recognition system model provides for the optimization of training and identification function parameters, as well as changes in the method of measuring the distance between feature vectors (point distance measurement metrics).
References
1. Moskalenko, V. V., & Korobov, A. H. (2019). Optimization parameters of intellectual identification system of objects on the terrain. Radioelectronic and Computer Systems, (2), 32–39. https://doi.org/10.32620/reks.2016.2.05
2. Lakhno, V., Akhmetov, B., Chubaievskyi, V., Desiatko, A., Palaguta, K., Blozva, A., & Chasnovskyi, Y. (2021). Information security audit method based on the use of a neuro-fuzzy system. Lecture Notes in Networks and Systems, 232, 171–184.
3. Herasina, O. V., & Korniienko, V. I. (2010). Alhorytmy hlobalnoi i lokalnoi optymizatsii v zadachi identyfikatsii skladnykh dynamichnykh system [Algorithms of global and local optimization in the problem of identification of complex dynamic systems]. Systemy obrobky informatsii, (6), 73–77.
4. Lakhno, V., Kryvoruchko, O., Desiatko, A., Blozva, A., & Semidotska, V. (2020). Development strategy model of the informational management logistic system of a commercial enterprise by neural network apparatus. CEUR Workshop Proceedings, 2746, 87–98.
5. Schuster, H. G. (1992). Deterministic chaos: Introduction and recent results. In H. Thomas (Ed.), Nonlinear dynamics in solids (pp. 11–19). Springer. https://doi.org/10.1007/978-3-642-95650-8_2
6. Ljung, L., Andersson, C., Tiels, K., & Schön, T. B. (2020). Deep learning and system identification. IFAC-PapersOnLine, 53(2), 1175–1181. https://doi.org/10.1016/j.ifacol.2020.12.1329
7. Nelles, O. (2001). Nonlinear system identification: From classical approaches to neural and fuzzy models. Springer.
8. Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep learning. MIT Press.
9. Lee, J.-H. (2019). Minimum Euclidean distance evaluation using deep neural networks. AEU – International Journal of Electronics and Communications, 112, Article 152964. https://doi.org/10.1016/j.aeue.2019.152964
10. Sahun, A., Khaidurov, V., & Bobkov, V. (2024). Model of graphic object identification in a video surveillance system based on a neural network. In Cybersecurity Providing in Information and Telecommunication Systems (CPITS-2024). CEUR Workshop Proceedings. https://ceur-ws.org/Vol-3654/short2.pdf
11. Sturgess, P. (2016). CamVid database. Oxford Brookes University.
12. Apostolopoulos, I. D., & Tzani, M. (2022). Industrial object, machine part and defect recognition towards fully automated industrial monitoring employing deep learning: The case of multilevel VGG19. Journal of Ambient Intelligence and Humanized Computing. https://doi.org/10.1007/s12652-021-03688-7
Downloads
Published
Issue
Section
License
Copyright (c) 2024 Інформаційні технології в економіці та природокористуванні
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
