Affiliation:
1. Saint-Petersburg State University of Aerospace Instrumentation
Abstract
The most complete information on the state of different man-made objects, collected with on-site and remote sensing methods, is required to ensure environmental and technospheric safety at the regional scale. The procedures of oil tank detection traditionally utilize high-cost, ultra-high-resolution images. The research is devoted to studying the possibility of using high and mediumresolution data from Landsat‑8 and Sentinel‑2 sensors to handle the task. Operating of the parts of the detection algorithms used in practice was analyzed, and some of them were selected as workable options which lead to satisfactory results being applied to data from mentioned instruments. A new technique of tank identifying which consists of classification, blob detection and filtering steps was developed. Testing of proposed solutions on data on the territory of Chaunsky District of Chukotka Autonomous Okrug showed the possibility of their use for detecting objects of the considered category.
Reference15 articles.
1. Gerus A.V., Savchenko E.V., Savorskiy V.P. Using the orthogonal projection method to identify small objects in multispectral analysis. Sovremennye problemy distantsionnogo zondirovaniya Zemli iz kosmosa, 2018, vol. 15, no. 4, pp. 27–35. (In Russian).
2. Dostovalov M. Yu., Ermakov R.V., Teplov A.A. Detection of aircraft using change detection method on Sentinel 1 synthetic aperture radar images. Sovremennye problemy distantsionnogo zondirovaniya Zemli iz kosmosa, 2018, vol. 15, no. 4, pp. 49–57. (In Russian).
3. Melentyev V.V., Chernook V.I., Melentyev K.V., Startsev A.A., Zakharova T.A. Satellite traceology: experience of using in the interests of mammalogy for protection of ice-associated marine mammals. Izvestiya TINRO, 2014, no. 179, pp. 138–157. (In Russian).
4. Sohn G., Dowman I.J. Extraction of buildings from high resolution satellite data. Automated extraction of man-made objects from aerial and space images (III). Balkema Publishers, 2001, pp. 345–355.
5. Vakalopoulou M., Karantzalos K., Komodakis N., Paragios N. Building detection in very high resolution multispectral data with deep learning features. Geoscience and Remote Sensing Symposium (IGARSS), 2015 IEEE International, pp. 1873–1876.