Инд. авторы:	Синявский Ю.Н., Пестунов И.А., Дубровская О.А., Рылов С.А., Мельников П.В., Ермаков Н.Б., Полякова М.А.
Заглавие:	Методы и технология сегментации мультиспектральных изображений высокого разрешения для исследования природных и антропогенных объектов
Библ. ссылка:	Синявский Ю.Н., Пестунов И.А., Дубровская О.А., Рылов С.А., Мельников П.В., Ермаков Н.Б., Полякова М.А. Методы и технология сегментации мультиспектральных изображений высокого разрешения для исследования природных и антропогенных объектов // Вычислительные технологии. - 2016. - Т.21. - № 1. - С.127-140. - ISSN 1560-7534. - EISSN 2313-691X.
Внешние системы:	РИНЦ: 25644534;
Реферат:	eng: A logical scheme for uniform representation and technology for joint processing of heterogeneous spatial data is proposed. A set of raster layers (data layers) is generated by all available geo-referenced data (satellite images, maps, digital elevation models, field data etc.). Data layers are used to generate a set of binary masks (thematic layers) built with available a priori information and expert knowledge. Thematic layers are designed to highlight specific types of objects (water bodies, shadows, vegetation, manmade areas, etc). All of the new raster layers are interpreted as additional features during further processing. Thematic layers allow using the most appropriate method of processing for each type of object. Methods necessary for solving practical problems with the proposed technology being developed at the laboratory of data processing of the Institute of Computational Technologies SB RAS. These methods are implemented as standardized web services (WPS processes). Nine web services are created based on original algorithms of image segmentation and highlighting of different object types. This approach allows using the proposed technology to solve practical problems on the client side using both freeware GIS packages (QGIS, uDig, openJUMP et al) and commercial geographic information system ArcGIS. The technology and methods been used successfully to solve three practical problems: 1) discovery and mapping of pine tree stands damage by the Pleiades-1A satellite images; 2) identification of the fundamental laws of formation for steppe vegetation biome by the WorldView-2 satellite data; 3) rapid assessment of the flood situation and flooded areas identification by images from Russian satellites (Canopus-B, Resurs-P, and Meteor-M). rus: Предложена логическая схема единообразного представления разнородных пространственных данных. На ее основе разработана технология сегментации спутниковых изображений высокого пространственного разрешения, которая позволяет учесть всю имеющуюся информацию (спектральные и пространственные признаки, данные полевых наблюдений, тематические карты, базы данных, априорные и экспертные сведения и т. п.). Представлены методы, предназначенные для тематической обработки спутниковых снимков высокого пространственного разрешения. Описана реализация методов и алгоритмов в виде стандартизованных веб-сервисов, которая позволяет использовать технологию на стороне пользователя. Приведены примеры решения практических задач.
Ключевые слова:	texture and context features; high spatial resolution; multispectral satellite images segmentation; веб-сервисы; обработка разнородных данных; текстурные и контекстные признаки; высокое пространственное разрешение; сегментация мультиспектральных спутниковых изображений; web services; heterogeneous data processing;
Издано:	2016
Физ. характеристика:	с.127-140
Цитирование:	1. Dvorkin, B.A., Dudkin, S.A. Newest and perspective satellites for Earth remote sensing. Geomatics. 2013; (2):16-21. (In Russ.). 2. Dey, V., Zhang, Y., Zhong, M. A review on image segmentation techniques with remote sensing perspective. ISPRS TC VII Symp. - 100 Years ISPRS. Vienna, Austria, July 5-7, 2010. IAPRS. 2010; XXXVIII(7A):31-42. 3. Wang, A., Wang, S., Lucieer, A. Segmentation of multispectral high-resolution imagery based on integrated feature distribution. International Journal of Remote Sensing. 2010; 31(6):1471-1483. 4. Pestunov, I.A., Sinyavskiy, Yu.N. Clustering algorithms in satellite images segmentatioin tasks. Bulletin of KemSU. 2012; 52(4/2):110-125. (In Russ.). 5. Pestunov, I.A., Sinyavsky, Yu.N. Nonparametric grid-based clustering algorithm for remote sensing data. Optoelectronics, Instrumentation and Data Processing. 2006; 42(2):78-85. 6. Pestunov, I.A., Berikov, V.B., Sinyavskiy, Yu.N. Algorithm for multispectral image segmentation based оn ensemble of nonparametric clustering algorithms. Vestnik SibGAU. 2010; 5 (31):56-64. (In Russ.). 7. Pestunov, I.A., Berikov, V.B., Kulikova, E.A., Rylov, S.A. Ensemble of clustering algorithm for large datasets. Optoelectronics, Instrumentation and Data Processing. 2011; 47(3):245-252. 8. Rylov, S.A., Pestunov, I.A. NVIDIA GPU for multispectral data clustering with grid-based algorithm CCA. Interexpo Geo-Siberia. 2015; 4(2):51-56. (In Russ.). 9. Pestunov, I.A., Rylov, S.A., Berikov, V.B. Hierarchical clustering algorithms for segmentation of multispectral images. Optoelectronics, Instrumentation and Data Processing. 2015; 51(4):329-338. 10. Pestunov, I., Rylov, S., Berikov, V. Hierarchical ensemble clustering algorithm for multispectral image segmentation. Proceedings 9th Open German-Russian Worokshop on Pattern Recognition and Image Understanding (OGRW-2014). December 1-5, 2014, Koblenz, Germany. Koblenz: University of Koblenz-Landau; 2015:123-127. Available at: http://kola.opus.hbz- nrw.de/volltexte/2015/1136/pdf/OGRW_2014_Proceedings.pdf#130 11. Pestunov, I.A. Algorithms for processing polizonal video information for detection and classification of forests infested with insects. Pattern Recognition and Image Analysis. 2001; 11(2):368-371. 12. Pestunov, I.A., Rylov, S.A. Spectral-textural segmentation algorithms for satellite images with high spatial resolution. Bulletin of KemSU. 2012; 4/2 (52):104-110. (In Russ.). 13. Pestunov, I.A., Melnikov, P.V., Dubrovskaya, O.A., Sinyavskiy, Yu.N., Kharuk, V.I. Detection and mapping of nut pine stands damage on pleiades satellite images. Interexpo Geo-Siberia. 2014; (2):359-366. (In Russ.). 14. Ermakov, N.B., Pestunov, I.A., Polyakova, M.A., Dubrovskaya, O.A., Rylov, S.A., Sinyavskiy, Yu.N. Large-scale mapping of steppe vegetation structure and identification of communities which contain rare and unique plants in South Siberia using high spatial resolution satellite images. Regional Issues of Earth Remote Sensing: Proceedings of the International Conference. Krasnoyarsk: SFU; 2014: 224-229. (In Russ.). 15. Kulikova, E.A., Pestunov, I.A. Semisupervised classification in multispectral images processing. Vestnik KazNU. 2008; 3(58): 284-291. Common Is. Pt II. (In Russ.). 16. Pestunov, I.A., Lazarev, D.V., Sinyavskiy, Yu.N. Detecting man-made objects on high-spatial resolution satellite images with Canny filter. Spatial Data Processing in Natural and Antropogenous Processes Monitoring: Proceedings Russian Conference. Novosibirsk: ICT SB RAS; 2015:115-119. Available at: http://conf.nsc.ru/files/conferences/SDM-2015/294652/SDM-2015%20Thesis.pdf. (In Russ.). 17. Canny J. A computational approach to edge Detection. IEEE Transactions on Pattern Analysis and Machine Intelligence. 1986; PAMI-8(6):679-698. 18. Borzov, S.M., Pestunov, I.A. High-spatial resolution images segmentation based on spectral, texture and structure features for nature and man-made emergency analysis. Safety and Survivability of Technical Systems: Proceedings Russian Conference Krasnoyarsk; 2012: 209-212. (In Russ.). 19. Pestunov, I.A., Rylov, S.A. A method for shadow detection on high-spatial resolution multispectral satellite images. Proceedings Youth Seminar «Earth remote sensing from space: algorithms, technologies, data». Barnaul: AZBUKA; 2013: 60-73. (In Russ.). 20. Rylov, S.A., Novgorodtseva, O.G., Pestunov, I.A. Flood hazard monitoring that uses high-resolution satellite images and takes shadows into account. AGROINFO-2015: Proceedings International Conference. Pt 1. Novosibirsk; 2015: 434-439. (In Russ.). 21. Pestunov, I.A., Rylov, S.A., Melnikov, P.V., Sinyavskiy, Yu.N. Technology and software toolkit for segmentation of satellite high spatial resolution images. Interexpo Geo-Siberia. 2013; (1): 202-208. (In Russ.). 22. Dobrotvorskiy, D.I, Kulikova, E.A., Pestunov, I.A., Sinyavskiy, Yu.N. Web-services for nonparametric classification of satellite data. 2010; 1(2):171-175. (In Russ.). 23. Zhizhimov, O.L., Molorodov, Yu.I., Pestunov, I.A., Smirnov, V.V., Fedotov, A.M. Heterogenous data integration for nature ecosystems research. NSU Journal of Information Technologies. 2011; 9(1):67-74. (In Russ.). 24. Cherepanov, A.S. Vegetation indices. Geomatics. 2011; (2):98-102. (In Russ.). 25. Ermakov, N., Larionov, A., Polyakova. M., Pestunov, I., Didukh, Y. Diversity and spatial structure of cryophytic steppes of the Minusinskaya intermountain basin in Southern Siberia. Tuexenia. 2014; (34):431-446. 26. Rylov, S.A., Novgorodtseva, O.G., Pestunov, I.A., Dubrovskaya, O.A., Sinyavskiy, Yu.N. Technology of «Kanopus-V», «Resurs-P» and «Meteor-M» data processing for flood hazard monitoring and mapping. Regional Issues of Earth Remote Sensing: Proceedings International Conference Krasnoyarsk: SFU; 2015: 207-212. (In Russ.).