sergeogrИзвестия Российской академии наук. Серия географическаяIzvestiya Rossiiskoi Akademii Nauk. Seriya Geograficheskaya2587-55662658-697510.31857/S2587556623010089sergeogr-1734Research ArticleГеоинформационные системы и картографированиеGeoinformation Systems and MappingsРаспознавание образов в задачах картографирования рельефа сушиPattern Recognition in the Tasks of Landform MappingХарченкоС. В.KharchenkoS. V.

Москва

Moscow

xar4enkkoff@yandex.ruМосковский государственный университет имени М.В. Ломоносова; Институт географии РАНРоссияMoscow State University; Institute of Geography, Russian Academy of SciencesRussian Federation202316042023871192206Copyright © Харченко С.В., 20232023Харченко С.В.Kharchenko S.V.Данная работа распространяется под лицензией Creative Commons Attribution 4.0.This work is licensed under a Creative Commons Attribution 4.0 License.https://izvestia.igras.ru/jour/article/view/1734

В статье рассмотрено современное состояние методов распознавания образов для автоматического и полуавтоматического составления геоморфологических карт. В настоящее время среди специалистов в области морфометрии и математического моделирования рельефа широко распространено мнение, что экспертные знания и правила, используемые для создания таких карт, могут быть описаны количественно и представлены алгоритмически. Основные нерешенные пока удовлетворительно проблемы автоматизации картографирования рельефа: распознавание морфологически идентичных форм разного генезиса, выработка критериев перехода от морфологических к возрастным и генетическим характеристикам поверхностей, превентивный подбор оптимальной детальности данных дистанционного зондирования (не всегда наиболее детальные данные оказываются эффективнее всего в модели), выбор и обоснование весовых коэффициентов количественных переменных. Приведены примеры использования нескольких методов распознавания образов в геоморфологии вообще и геоморфологическом картографировании: обобщенных линейных моделей, классификаций деревьев решений, искусственных нейронных сетей и некоторых других методов, включая “компьютерное зрение”. Показано, что точность различных моделей в отношении распознавания форм рельефа (равно как и геологических структур) составляет порядка 50–70%, реже больше. В то же время распознавание конкретных форм, в зависимости от исходных данных и степени выраженности на них признаков данных форм рельефа, может иногда быть даже абсолютной (100%), но чаще всего точность при тестировании находится в пределах 90%.

The article aims to show the modern state of pattern recognition techniques for automatic and semi-automatic geomorphological mapping. There is opinion among the geomorphometrists about the expert rules for traditional landform mapping can be quantitated. The general unsolved tasks of automatic landform mapping are: recognition of origin for morphologically similar Earth’s surface forms; criteria development for transfer from morphological to genetic and age landform’s characteristics; preventive choosing the optimal resolution of the remote sensing data; the choosing and rationale of predictor’s weights in statistical modeling procedures. Some cases of the pattern recognition techniques using in geomorphology and landform mapping are given: generalized linear models; classification trees; random forest; artificial neural networks; and computer vision methods. The overall accuracy of the different models according to planar continuous landform recognition (and recognition of lithology types too) is about 50–70% and more. At the same time, specific landform type’s (craters, volcanic cones and others) recognition can reach 90–100%.

распознавание образовгеоморфологическое картографированиеклассификация с обучениеманализ изображенийpattern recognitiongeomorphological mappingsupervised classificationimage analysisИсследование выполнено за счет гранта Российского научного фонда (проект № 19-77-10036)The study was supported by the Russian Science Foundation (project no. 19-77-10036)ReferencesГаврилов А.А. О природе явлений геоморфологической конвергенции и гомологии // Вестн. Моск. ун-та. Сер. 5: География. 2016. № 4. С. 3–12.Anders N.S., Seijmonsbergen A.C., Bouten W. Segmentation optimization and stratified object-based analysis for semi-automated geomorphological mapping. Remote Sensing Environ., 2011, vol. 115, no. 12, pp.2976– 2985.Геоморфологическая карта Мурманской области / сост. М.К. Граве, Л.М. Граве / Атлас Мурманской области. М.: ГУГК, 1971. С. 8.Atkinson J., Jiskoot H., Massari R., and Murray T. Generalized linear modelling in geomorphology. Earth Surface Processes and Landforms, 1998, no. 23, pp. 1185– 1195. https://doi.org/10.1002/(SICI)1096-9837(199812)23:13%3C1185::AID-ESP928%3E3.0.CO;2-WКрамбейн У., Грейбилл Ф. Статистические модели в геологии. М.: Мир, 1969. 400 с.Ayalew L., Yamagishi H. The application of GIS-based logistic regression for landslide susceptibility mapping in the Kakuda-Yahiko Mountains, Central Japan. Geomorphology, 2005, vol. 65, no. 1–2, pp. 15–31. https://doi.org/10.1016/j.geomorph.2004.06.010Ласточкин А.Н. Морфодинамический анализ. Ленинград: Недра, 1987. 256 с.Brandmeier M., Chen Y. Lithological classification using multi-sensor data and convolutional neural networks, International Archives of the Photogrammetry, Remote Sensing & Spatial Information Sciences, 2019. https://doi.org/10.5194/isprs-archives-XLII-2-W1655-2019Лобанов В.В. Еще раз об “элементарной морфологической единице”, ее содержании и методике установления // Геоморфология. 1988. № 4. С. 29–34.Breiman L. Random Forests. Machine Learning, 2001, no.45,pp.5–32. https://doi.org/10.1023/A:1010933404324Лопатин Д.В. О полиморфизме в геоморфологии // Геоморфология. 2007. № 3. С. 22–23.Brown D. G., Lusch D. P., Duda K. A. Supervised classification of types of glaciated landscapes using digital elevation data. Geomorphology, 1998, vol. 21, no. 3–4, pp.233–250. https://doi.org/10.1016/S0169-555X(97)00063-9Серебрянный Л.Р., Чукленкова И.Н. Густота озер как показатель возраста гляцигенной морфоскульптуры (опыт применения морфометрического анализа в северо-западных районах Русской равнины) // Геоморфология. 1973. № 4. С. 79–85.Cavazzi S., Corstanje R., Mayr T., Hannam J., Fealy R. Are fine resolution digital elevation models always the best choice in digital soil mapping? Geoderma, 2013, vol. 195, pp. 111–121. https://doi.org/10.1016/j.geoderma.2012.11.020Солоненко В.П. Сейсмогенные деформации и палеосейсмогеологический метод // Сейсмическое районирование Восточной Сибири и его геологогеофизические основы. Новосибирск: Наука, 1977. С. 5–47.Clark C.D., Hughes A.L., Greenwood S.L., Spagnolo M., Ng F.S. Size and shape characteristics of drumlins, derived from a large sample, and associated scaling laws.Quaternary Sci. Rev., 2009, vol. 28, no. 7–8, pp.677–692. https://doi.org/10.1016/j.quascirev.2008.08.035Спиридонов А.И. Геоморфологическое картографирование. М.: Недра, 1974. 184 с.Deng Y., Wilson J., and Sheng J. Effects of variable attribute weights on landform classification. Earth Surface Processes and Landforms, 2006, no. 31, pp. 1452–1462. https://doi.org/10.1002/esp.1401Тимофеев Д.А. О полиморфизме как общем свойстве земной поверхности // Геоморфология. 2006. № 2. С. 3–6.d’Oleire-Oltmanns S., Eisank C., Drăgut L., Blaschke T. An object-based workflow to extract landforms at multiple scales from two distinct data types. IEEE Geoscience Remote Sensing Lett., 2013, vol. 10, no. 4, pp.947–951. https://doi.org/0.1109/LGRS.2013.2254465Тимофеев Д.А. Элементарные морфологические единицы как объект геоморфологического анализа // Геоморфология. 1984. № 1. С. 19–29.Ermini L., Catani F., and Casagli N. Artificial natural networks applied to landslide susceptibility assessment. Geomorphology, 2005, vol. 66, pp. 327–343. https://doi.org/10.1016/j.geomorph.2004.09.025Харченко С.В. К вопросу о применении гармонического анализа при количественной характеристике рельефа // Геоморфология. 2017. № 2. С. 14–24.Evans I.S. Geomorphometry and landform mapping: Whatis a landform? Geomorphology, 2012, vol. 137, pp.94–106. https://doi.org/0.1016/j.geomorph.2010.09.029Харченко С.В. Новые задачи морфометрии рельефа и автоматизированные морфологические классификации в геоморфологии // Геоморфология. 2020. № 1. С. 3–21.Florinsky I.V., Eilers R.G., Manning G.R., Fuller L.G. Prediction of soil properties by digital terrain modeling. Environ. Modelling & Software, 2002, no. 17 (3), pp.295–311. https://doi.org/0.1016/S1364-8152(01)00067-6Anders N.S., Seijmonsbergen A.C., Bouten W. Segmentation optimization and stratified object-based analysis for semiautomated geomorphological mapping // Rem. Sens. ofFlorinsky I.V. Digital terrain analysis in soil science and geology. Academic Press, 2016. 506 p.Environ. 2011. Vol. 115. № 12. P. 2976–2985. https://doi.org/10.1016/j.rse.2011.05.007Gavrilov A.A. About the nature of geomorphologic convergence and homology phenomena. Vestn. Mosk. Univ. Ser. 5: Geogr., 2016, no. 4, pp. 3–12. (In Russ.).Atkinson J., Jiskoot H., Massari R., Murray T. Generalized linear modelling in geomorphology // Earth Surface Proc. and Landforms. 1998. № 23. P. 1185–1195. https://doi.org/10.1002/(SICI)1096-9837(199812)23:13%3C1185::AID-ESP928%3E3.0.CO;2-WGeomorfologicheskaya karta Murmanskoi oblasti [Geomorphological Map of the Murmansk oblast (Russia)], by M.K.Grave and L.M. Grave, In Murmansk region atlas. Moscow: GUGK Publ., 1971. 8 p.Ayalew L., Yamagishi H. The application of GIS-based logistic regression for landslide susceptibility mapping in the Kakuda-Yahiko Mountains, Central Japan // Geomorphology. 2005. Vol. 65. № 1–2. P. 15–31. https://doi.org/10.1016/j.geomorph.2004.06.010Kharchenko S.V. Application of harmonic analysis for the quantitative description of Earth surface topography. Geomorfologiya, 2017, no. 2, pp. 14–24. (In Russ.).Brandmeier M., Chen Y. Lithological classification using multi-sensor data and convolutional neural networks // Int. Archives of the Photogrammetry, Remote Sens. & Spatial Inform. Sci. 2019. https://doi.org/10.5194/isprs-archives-XLII-2-W1655-2019Kharchenko S.V. New challenges of geomorphometry and automatic morphological classification in geomorphology. Geomorfologiya, 2020, no. 1, pp. 3–21. (In Russ.).Breiman L. Random Forests // Machine Learning. 2001. №45. P. 5–32. https://doi.org/10.1023/A:1010933404324Kharchenko S. Automated recognition of the landforms origin for the Kola Peninsula based on morphometric variables. EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-15564. https://doi.org/10.5194/egusphere-egu21-15564Brown D.G., Lusch D.P., Duda K.A. Supervised classification of types of glaciated landscapes using digital elevation data // Geomorphology. 1998. Vol. 21. № 3–4. P.233–250. https://doi.org/10.1016/S0169-555X(97)00063-9Krumbein W.C., Graybill F.A. Statisticheskie modeli v geologii [Statistical Models in Geology]. Moscow: Mir Publ., 1969. 400 p.Cavazzi S., Corstanje R., Mayr T., Hannam J., Fealy R. Are fine resolution digital elevation models always the best choice in digital soil mapping? // Geoderma. 2013. Vol.195. P. 111–121. https://doi.org/10.1016/j.geoderma.2012.11.020Lastochkin A.N., Zhirov A.I., Boltramovich S.F. Systemmorphological approach: Another look at morphology research and geomorphological mapping. Geomorphology, 2018, vol. 303, pp. 486–503. https://doi.org/10.1016/J.GEOMORPH.2017.10.022Clark C.D., Hughes A.L., Greenwood S.L., Spagnolo M., Ng F.S. Size and shape characteristics of drumlins, derived from a large sample, and associated scaling laws // Quat. Sci. Rev. 2009. Vol. 28. № 7–8. P. 677–692. https://doi.org/10.1016/j.quascirev.2008.08.035Lastochkin A.N. Morfodinamicheskii analiz [Morphodynamical Analysis]. Leningrad: Nedra Publ., 1987. 256 p.Deng Y., Wilson J., Sheng J. Effects of variable attribute weights on landform classification // Earth Surface Proc. and Landforms. 2006. № 31. P. 1452–1462. https://doi.org/10.1002/esp.1401Leverington D., Duguay C. A neural network method to determine the presence or absence of permafrost near Mayo, Yukon Territory, Canada. Permafrost and Periglacial Processes, 1997, no. 8, pp. 205–215. https://doi.org/10.1002/(SICI)1099-1530(199732)8:23.0.CO;2-5d’Oleire-Oltmanns S., Eisank C., Drăgut L., Blaschke T. An object-based workflow to extract landforms at multiple scales from two distinct data types // IEEE Geosci. and Rem. Sens. Lett. 2013. Vol. 10. № 4. P. 947–951. https://doi.org/10.1109/LGRS.2013.2254465Li W., Hsu C.Y. Automated terrain feature identification from remote sensing imagery: a deep learning approach. Int. J. Geogr. Information Sci., 2020, no.34(4), pp. 637–660. https://doi.org/10.1080/13658816.2018.1542697Ermini L., Catani F., Casagli N. Artificial natural networks applied to landslide susceptibility assessment // Geomorphology. 2005. Vol. 66. P. 327–343. https://doi.org/10.1016/j.geomorph.2004.09.025Link A.J. A physic-chemical and textural study of carbonate sedimentation in a lagoonal environment. Evanston, IL: Northwestern University, 1964.Evans I.S. Geomorphometry and landform mapping: What is a landform? // Geomorphology. 2012. Vol. 137. P.94–106. https://doi.org/10.1016/j.geomorph.2010.09.029Lobanov V.V. Once more on the “elementary morphological unit,” its content and methods of its identification. Geomorfologiya, 1988, no. 4, pp. 29–34. (In Russ.).Florinsky I.V., Eilers R.G., Manning G.R., Fuller L.G. Prediction of soil properties by digital terrain modeling // Environ. Modelling & Software. 2002. № 17 (3). P.295–311. https://doi.org/10.1016/S1364-8152(01)00067-6Lopatin D.V. Polymorphism in geomorphology. Geomorfologiya, 2007, no. 3, pp. 22–23. (In Russ.).Florinsky I.V. Digital terrain analysis in soil science and geology. Academic Press, 2016. 506 p. Kharchenko S. Automated recognition of the landforms origin for the Kola Peninsula based on morphometric variables / EGU General Assembly 2021 (19–30 April 2021) EGU21-15564. https://doi.org/10.5194/egusphere-egu21-15564Luoto M., Hjort J. Evaluation of current statistical approaches for predictive geomorphological mapping. Geomorphology, 2005, no. 67, pp. 299–315. https://doi.org/10.1016/j.geomorph.2004.10.006Lastochkin A.N., Zhirov A.I., Boltramovich S.F. Systemmorphological approach: Another look at morphology research and geomorphological mapping // Geomorphology. 2018. Vol. 303. P. 486–503. https://doi.org/10.1016/J.GEOMORPH.2017.10.022MacMillan R.A., Pettapiece W.W., Nolan S.C., and Goddard T.W. A generic procedure for automatically segmenting landforms into landform elements using DEMs, heuristic rules and fuzzy logic. Fuzzy Sets.Systems, 2000, no. 113, pp. 81–109. https://doi.org/10.1016/S0165-0114(99)00014-7Leverington D., Duguay C. A neural network method to determine the presence or absence of permafrost near Mayo, Yukon Territory, Canada // Permafrost and Periglacial Proc. 1997. № 8. P. 205–215. https://doi.org/10.1002/(SICI)10991530(199732)8:23.0.CO;2-5MacMillan R.A., Shary P.A. Landforms and landform elements in geomorphometry. In Geomorphometry: Concepts, Software, Applications. Developments in soil science – 33. Hengl T., Reuter H.I., Eds. 2009, pp. 227–254. https://doi.org/10.1016/S0166-2481(08)00009-3Li W., Hsu C.Y. Automated terrain feature identification from remote sensing imagery: a deep learning approach // Int. J. Geographical Inform. Sci. 2020. № 34 (4). P. 637– 660. https://doi.org/10.1080/13658816.2018.1542697MacMillan R.A., Pettapiece W.W. Alberta Landforms: Quantitative morphometric descriptions and classification of typical Alberta landforms. Technical Bulletin No. 2000-2E. Research Branch, Agriculture and Agri-Food Canada, Semiarid Prairie. Agricultural Research Centre, Swift Current, SK, 2000.Link A.J. A physic-chemical and textural study of carbonate sedimentation in a lagoonal environment. Evanston, IL: Northwestern Univ., 1964.Marmion M., Hjort J., Thuiller W., Luoto M. A comparison of predictive methods in modelling the distribution of periglacial landforms in Finnish Lapland. Earth Surface Processes, Landforms, 2008, no. 33, pp. 2241– 2254. https://doi.org/0.1002/esp.1695Luoto M., Hjort J. Evaluation of current statistical approaches for predictive geomorphological mapping // Geomorphology. 2005. № 67. P. 299–315. https://doi.org/10.1016/j.geomorph.2004.10.006Miska L. and Jan H. Evaluation of current statistical approaches for predictive geomorphological mapping. Geomorphology, 2005, vol. 67, no. 3–4, pp. 299–315. https://doi.org/10.1016/j.geomorph.2004.10.006MacMillan R.A., Pettapiece W.W., Nolan S.C., Goddard T.W. A generic procedure for automatically segmenting landforms into landform elements using DEMs, heuristic rules and fuzzy logic // Fuzzy Sets and Systems. 2000. № 113. P. 81–109. https://doi.org/10.1016/S0165-0114(99)00014-7Pennock D.J., Zebarth B.J., De Jong E. Landform classification and soil distribution in hummocky terrain, Saskatchewan, Canada. Geoderma, 1987, vol. 40, no. 3–4, pp. 297–315. https://doi.org/10.1016/0016-7061(87)90040-1MacMillan R.A., Shary P.A. Landforms and landform elements in geomorphometry // Geomorphometry: Concepts, Software, Applications. Developments in soil science – 33 / T. Hengl, H.I. Reuter (Eds.). 2009. Vol.33. P. 227–254. https://doi.org/10.1016/S0166-2481(08)00009-3Rashid I., Romshoo S.A., Hajam J.A., Abdullah T. A semiautomated approach for mapping geomorphology in mountainous terrain, Ferozpora watershed (Kashmir Himalaya). J. Geological Society India, 2016, vol. 88, no. 2, pp. 206–212. https://doi.org/10.1007/s12594-016-0479-5MacMillan R.A., Pettapiece W.W. Alberta Landforms: Quantitative morphometric descriptions and classification of typical Alberta landforms / Technical Bul. №2000-2E. Research Branch, Agriculture and AgriFood Canada, Semiarid Prairie. Agricultural Research Centre, Swift Current, SK, 2000. 118 p.Shary P.A., Sharaya L.S., and Mitusov A.V. Fundamental quantitative methods of land surface analysis. Geoderma, 2002, vol. 107, no. 1–2, pp. 1–32. https://doi.org/10.1016/S0016-7061(01)00136-7Marmion M., Hjort J., Thuiller W., Luoto M. A comparison of predictive methods in modelling the distribution of periglacial landforms in Finnish Lapland // Earth Surface Proc. and Landforms. 2008. № 33. P. 2241–2254. https://doi.org/10.1002/esp.1695Smith M.J. and Clark C.D. Methods for the visualization of digital elevation models for landform mapping. Earth Surface Processes, Landforms, 2005, vol. 30, no. 7, pp.885–900. https://doi.org/10.1002/esp.1210Miska L., Jan H. Evaluation of current statistical approaches for predictive geomorphological mapping // Geomorphology. 2005. Vol. 67. № 3–4. P. 299–315. https://doi.org/10.1016/j.geomorph.2004.10.006Sutfin N.A., Shaw J.R., Wohl E.E., and Cooper D.J. A geomorphic classification of ephemeral channels in a mountainous, arid region, southwestern Arizona, USA. Geomorphology, 2014, no. 221, pp. 164–175. https://doi.org/10.1016/j.geomorph.2014.06.005Pennock D.J., Zebarth B.J., De Jong E. Landform classification and soil distribution in hummocky terrain, Saskatchewan, Canada // Geoderma. 1987. Vol. 40. № 3– 4. P. 297–315. https://doi.org/10.1016/0016-7061(87)90040-1Serebryanny L.R., Chuklenkova I.N. Density of lakes as an age indicator of glacigenetic morphosculpture: an application of morphometrical analysis in the north-west areas of the Russian Plain. Geomorfologiya, 1973, no. 4, pp. 79–85. (In Russ.).Rashid I., Romshoo S.A., Hajam J.A., Abdullah T. A semiautomated approach for mapping geomorphology in mountainous terrain, Ferozpora watershed (Kashmir Himalaya) // J. Geological Soc. of India. 2016. Vol. 88. № 2. P. 206–212. https://doi.org/10.1007/s12594-016-0479-5Solonenko V.P. Seismogenetic deformations and palaeoseismic method. In Seismicheskoe raionirovanie Vostochnoi Sibiri i ego geologo-geofizicheskie osnovy [Seismic Zoning of the East Siberia and its Geological and Geophysical Basics]. Novosibirsk: Nauka Publ., 1977, pp. 5–47. (In Russ.).Shary P.A., Sharaya L.S., Mitusov A.V. Fundamental quantitative methods of land surface analysis // Geoderma. 2002. Vol. 107. № 1–2. P. 1–32. https://doi.org/10.1016/S0016-7061(01)00136-7Spiridonov A.I. Geomorfologicheskoe kartografirovanie [Geomorphological Mapping]. Moscow: Nedra Publ., 1974. 184 p.Smith M.J., Clark C.D. Methods for the visualization of digital elevation models for landform mapping // Earth Surface Proc. and Landforms. 2005. Vol. 30. № 7. P.885–900. https://doi.org/10.1002/esp.1210Timofeyev D.A. Polymorphism as general attribute of land surface. Geomorfologiya, 2006, no. 2, pp. 3–6. (In Russ.).Sutfin N.A., Shaw J.R., Wohl E.E., Cooper D.J. A geomorphic classification of ephemeral channels in a mountainous, arid region, southwestern Arizona, USA // Geomorphology. 2014. № 221. P. 164–175. https://doi.org/10.1016/j.geomorph.2014.06.005Timofeev D.A. Elementary morphological units as an object of geomorphological analysis. Geomorfologiya, 1984, no. 1, pp. 19–29. (In Russ.).Veronesi F., Hurni L. Random Forest with semantic tie points for classifying landforms and creating rigorous shaded relief representations // Geomorphology. 2014. Vol. 224. P. 152–160. https://doi.org/10.1016/j.geomorph.2014.07.020Veronesi F. and Hurni L. Random Forest with semantic tie points for classifying landforms and creating rigorous shaded relief representations. Geomorphology, 2014, vol.224, pp. 152–160. https://doi.org/0.1016/j.geomorph.2014.07.020Weiss A.D. Topographic positions and landforms analysis (Conference Poster). ESRI Int. User Conference. San Diego, CA July 9–13 2001.Weiss A.D. Topographic positions and landforms analysis (Conference Poster). ESRI International User Conference. San Diego, CA July 9–13, 2001.Zhu Y.M., Lu X.X., Zhou Y. Suspended sediment flux modeling with artificial neural network: An example of the Longchuanjiang River in the Upper Yangtze Catchment, China // Geomorphology. 2007. № 84. P. 111– 125. https://doi.org/10.1016/j.geomorph.2006.07.010Zhu Y.M., Lu X.X., Zhou Y. Suspended sediment flux modeling with artificial neural network: An example ofthe Longchuanjiang River in the Upper Yangtze Catchment, China. Geomorphology, 2007, no. 84, pp.111–125. https://doi.org/10.1016/j.geomorph.2006.07.010

The authors declare that there are no conflicts of interest present.