Informatics and Applications
2023, Volume 17, Issue 1, pp 50-56
DEVELOPMENT OF A NEW MODEL OF STEP CONVOLUTIONAL NEURAL NETWORK FOR CLASSIFICATION OF ANOMALIES ON PANORAMAS
- P. O. Arkhipov
- S. L. Philippskih
- M. V. Tsukanov
Abstract
A new model ofa stepped convolutional neural network for classifying anomalies in panoramas has been developed. Appropriate datasets for classification are selected. The conclusion is made about the incompleteness of the method previously used by the authors to find anomalies in special areas with high color difference in panoramas. The search for these areas by the previously developed method did not set the task of their classification.
For automatic identification of detected objects, it is proposed to apply deep learning models using suitable neural networks. Particular attention is paid to work with data containing unbalanced classes and images of different sizes. The results of image classification of popular architectures of neural networks are compared with the newly developed stepped convolutional neural network.
[+] References (11)
- Arkhipov, P.O., and M.V. Tsukanov. 2019. Algoritmicheskaya model' metoda detektirovaniya anomaliy pri aerofotos"emke ob"ektov zhilishchno-kommunal'nogo khozyaystva [Algorithmic model of the method of detecting anomalies in the aerial photography of objects of
housing and communal services]. Informatsionnye sistemy i tekhnologii [Information Systems and Technologies] 6(116):39-45.
- Arkhipov, P.O., and M.V. Tsukanov. 2020. Informatsionnaya model' tekhnologii korrektsii yarkosti i tsveta pri sozdanii panoramnykh izobrazheniy [Information model of brightness and color correction technology for creating panoramic images]. Sistemy vysokoy dostupnosti [Highly Available Systems] 16(3):46-51. doi: 10.18127/ j20729472-202003-04.
- Arkhipov, P. O., and M. V. Tsukanov. 2021. Algoritmicheskaya model' obnaruzheniya anomaliy na raznovremennykh panoramakh [Algorithmic model of anomaly detection on different panoramas]. Sistemy vysokoy dostupnosti [Highly Available Systems] 17(2):5-10. doi: 10.18127/ j20729472-202102-01.
- Arkhipov, P.O., A.V. Kolesnik, and M.V. Tsukanov.
30.03.2022. Programmnaya sistema obnaruzheniya anomaliy na raznovremennykh panoramakh obsleduemoy mestnosti (CL_22) [Software system for detecting anomalies on multitemporal panoramas of the surveyed area (CL_22)]. Certificate of State Registration No. 2022615139.
- Trofimenkov, A.K., M.V. Tsukanov, and N.Y. Nosova. 2022. Modification of an algorithmic model for detecting anomalies on panoramas of different times obtained from unmanned aerial vehicles. Modern informatization problems in simulation and social technologies proceedings.
Yelm, WA: Science Book Publishing House LLC. 183186.
- Krizhevsky, À., I. Sutskever, and G. Hinton. 2017. ImageNet classification with deep convolutional neural networks. Commun. ACM 60:84-90. doi: 10.1145/ 3065386.
- Chollet, F 2021. Deep learning with Python. 2nded. Shelter Island, NY: Manning Publications Co. 504 p.
- Zhu, P, L. Wen, D. Du, X. Bian, H. Fan, Q. Hu, and
H. Ling. 2021. Detection and tracking meet drones challenge. arXiv.org. 20 p. doi: 10.48550/arXiv.2001.06303.
- Simonyan, K., and A. Zisserman. 2015. Very deep convolutional networks for large-scale image recognition. arXiv.org. 14 p. doi: 10.48550/arXiv.1409.1556.
- Chollet, F. 2017. Xception: Deep learning with depth- wise separable convolutions. arXiv.org. 8 p. doi: 10.48550/ arXiv.1610.02357.
- Kingma, D., andJ. Ba.2017. Adam: A method for stochastic optimization. arXiv.org. 15 p. doi: 10.48550/arXiv. 1412.6980.
[+] About this article
Title
DEVELOPMENT OF A NEW MODEL OF STEP CONVOLUTIONAL NEURAL NETWORK FOR CLASSIFICATION OF ANOMALIES ON PANORAMAS
Journal
Informatics and Applications
2023, Volume 17, Issue 1, pp 50-56
Cover Date
2023-04-10
DOI
10.14357/19922264230107
Print ISSN
1992-2264
Publisher
Institute of Informatics Problems, Russian Academy of Sciences
Additional Links
Key words
panoramic image; data set; multilabel classification; stepwise convolutional neural network; ensemble; transfer learning
Authors
P. O. Arkhipov  , S. L. Philippskih , and M. V. Tsukanov
Author Affiliations
Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation
|