Identification and classification of objects in images obtained by UAV and orbital base imaging equipmenttion
https://doi.org/10.21122/2309-4923-2022-4-30-37
Abstract
To identify and classify objects on images obtained using UAV imaging and orbital-based imaging, a neural network classification model based on the use of an autoencoder and built on the architecture of an ensemble of multilayer perceptrons is proposed. Additionally, at the stage of highlighting informative features, is added a color information, which is based on the per-channel histograms and is invariant to the scale and rotations of the image. The model is implemented using the Keras library. The use of the proposed model for classification into four classes: “Fire”, “Smoke”, “Vegetation” and “Buildings”, allows to achieve a classification accuracy above 99%.
About the Authors
A. A. Doudkin
United Institute of Informatics Problems of the National Academy of Sciences of Belarus
Belarus
Prof.
Belarus
Prof.
V. V. Ganchenko
United Institute of Informatics Problems of the National Academy of Sciences of Belarus
Belarus
PhD
Belarus
PhD
A. V. Inyutin
United Institute of Informatics Problems of the National Academy of Sciences of Belarus
Belarus
Researcher
Belarus
Researcher
E. E. Marushko
United Institute of Informatics Problems of the National Academy of Sciences of Belarus
Belarus
Researcher
Belarus
Researcher
References
1. Sentinel Hub EO Browser [Electronic resource]. – Mode of access: https://apps.sentinel-hub.com/eo-browser/?zoom=13&lat=50.20921&lng=30.23931&themeId=DEFAULT-THEME&visualizationUrl=https://services.sentinel-hub.com/ogc/wms/e35192fe-33a1-41f3-b798-b755e771c5a5&datasetId=AWS_LOTL1&fromTime=2015-06-09T00:00:00.000Z&toTime=2015-06-09T23:59:59.999Z&layerId=1_TRUE_COLOR/ – Date of access: 28.05.2022.
2. Oreshkina L.V. Object class detection and recognition in multi-zone remote sensing images / L.V. Oreshkina // Informatics. – 2019. – №. 2 (6). – С. 79-85. [In russian]
3. Kuga R. Multi-task learning using multi-modal encoder-decoder networks with shared skip connections / R. Kuga [et al.] // Proceedings of the IEEE International Conference on Computer Vision Workshops. – 2017. – С. 403-411.
4. Chen K. Deep Semantic Segmentation of Aerial Imagery Based on Multi-Modal Data / K. Chen [et al.] // IGARSS 2018-2018 IEEE International Geoscience and Remote Sensing Symposium. – 2018. – С. 6219-6222.
5. Feng D. [et al.] Deep multi-modal object detection and semantic segmentation for autonomous driving: Datasets, methods, and challenges // IEEE Transactions on Intelligent Transportation Systems. – 2020.
6. Kingma, D. P. Adam: A Method for Stochastic Optimization / D. P. Kingma, J. Ba // 3rd Intern. Conf. for Learning Representations, 7–9 May, 2015, San Diego. – San Diego, 2015. – Vol. 1412. – P. 6980.
7. Sokolova, M. Beyond accuracy, F-score and ROC: a family of discriminant measures for performance evaluation / M. Sokolova, N. Japkowicz, S. Szpakowicz // Advances in Artificial Intelligence : 19th Australian Joint Conf. on Artificial Intelligence, Hobart, 4–8 Dec., 2006. – Hobart, 2006. – P. 1015–1021.
Review
For citations:
Doudkin A.A., Ganchenko V.V., Inyutin A.V., Marushko E.E. Identification and classification of objects in images obtained by UAV and orbital base imaging equipmenttion. «System analysis and applied information science». 2022;(4):30-37. (In Russ.) https://doi.org/10.21122/2309-4923-2022-4-30-37
Views:
387
Email this article 