Lesnoy Vestnik / Forestry BulletinLesnoy Vestnik / Forestry BulletinЛесной вестник / Forestry Bulletin2542-1468Bauman Moscow State Technical University70604110.18698/2542-1468-2026-2-31-42Biological and technological aspects of forestryБиологические и технологические аспекты лесного хозяйстваResearch ArticleDetection of small spruce bark beetle outbreaks using neural network analysis of Siberian fir trees imagesОбнаружение очагов уссурийского полиграфа с помощью нейросетевого анализа изображений деревьев пихтыMarkovNikolay G.МарковНиколай Григорьевич
Russian Federation

Dr. Sci. (Tech.), Professor of the Department of Information Technologies

д-р техн. наук, профессор отделения информационных технологий (ОИТ) Инженерной школы информационных технологий и робототехники (ИШИТР)

markovng@tpu.ruMachucaMendosa Cristian R.МачукаМендоса Кристиан Родриго
Russian Federation

pg.

аспирант

kristianrodrigo1@tpu.ruNational Research Tomsk Polytechnic UniversityФГАОУ ВО «Национальный исследовательский Томский политехнический университет» (ТПУ)1404202630231421204202612042026Copyright ©; 2026, Markov N.G., Machuka M.R.Copyright ©; 2026, Марков Н.Г., Мачука М.Р.2026Markov N.G., Machuka M.R.Марков Н.Г., Мачука М.Р.https://creativecommons.org/licenses/by/4.0https://journals.eco-vector.com/2542-1468/article/view/706041

The main tasks of remote forest pathology monitoring of coniferous forests infested with insect pests are considered. The importance of using multiclass classification results of coniferous trees in high-resolution images obtained during aerial forest monitoring using spacecraft or unmanned aerial vehicles for rapid problem-solving is demonstrated. A method for the rapid detection of small spruce bark beetle outbreaks is proposed. This method involves multiclass classification of infested trees in images using the Res-Mo-U-Net neural network model, identification of tree age based on the assessment of their crown area using segmentation masks obtained through multiclass classification, and decision-making on the presence of small spruce bark beetle outbreaks in the studied fir forest area based on calculations of actual tree mortality. To test the proposed method, a neural network analysis of an image of a fir forest area infested by the small spruce bark beetle in the Tomsk region was carried out. The results of the testing indicate the potential of the method for practical application.

Рассмотрены основные задачи дистанционного лесопатологического мониторинга пораженных насекомыми-вредителями хвойных лесов. Показана важность для их оперативного решения использования результатов мультиклассификации хвойных деревьев на изображениях высокого разрешения, получаемых при мониторинге путем аэрофотосъемки лесов космическими или беспилотными летательными аппаратами. Предложена методика оперативного выявления очагов размножения уссурийского полиграфа, которая предусматривает мультиклассификацию с помощью модели нейронной сети Res-Mo-U-Net пораженных деревьев на изображениях, выявление возраста деревьев по результатам оценки площади их крон на полученных при мультиклассификации масках сегментации и принятие решения о наличии очага размножения уссурийского полиграфа на исследуемом участке пихтового леса по результатам расчетов фактического отпада деревьев. Выполнен нейросетевой анализ изображения пораженного уссурийским полиграфом участка пихтового леса в Томской области в целях апробации предложенной методики. Результаты апробации указывают на перспективность методики для практического применения.

coniferous forests pathology monitoringremote sensingsmall spruce bark beetle outbreaksconvolutional neural networkлесопатологический мониторинг хвойных лесовдистанционное зондирование Землиочаг размножения уссурийского полиграфасверточная нейронная сетьKrivets S.A., Bisirova E.M., Kerchev I.A., Pats E.N., Chernova N.A. Transformatsiya taezhnykh ekosistem v ochage invazii poligrafa ussuriyskogo Polygraphus proximus Blandf. (Coleoptera: Curculionidae, Scolytinae) v Zapadnoy Sibiri [Transformation of taiga ecosystems in western Siberia in invasion focus of foureyed fir bark beetle Polygraphus proximus Blandford (coleoptera: curculionidae, scolytinae)]. Rossiyskiy Zhurnal Biologicheskikh Invaziy [Russian J. of Biological Invasions], 2015, pp. 41–63.Кривец С.А., Бисирова Э.М., Керчев И.А., Пац Е.Н., Чернова Н.А. Трансформация таежных экосистем в очаге инвазии полиграфа уссурийского Polygraphus proximus Blandf. (Coleoptera: Curculionidae, Scolytinae) в Западной Сибири // Российский Журнал Биологических Инвазий, 2015. № 1. С. 41–63.Debkov N.M. Prirodnyy potentsial vozobnovleniya v pikhtovykh lesakh, povrezhdennykh v khode invazii ussuriyskogo poligrafa [The natural potential of regeneration in fir forests, damaged during the invasion of Polygraphus proximus]. Lesotekhnicheskiy zhurnal [Forestry Journal], 2017, v. 7, no. 1. pp. 58–68.Дебков Н.М. Природный потенциал возобновления в пихтовых лесах, поврежденных в ходе инвазии уссурийского полиграфа // Лесотехнический журнал, 2017. Т. 7. № 1. С. 58–68.Gosudarstvennyy doklad «O sostoyanii i ob okhrane okruzhayushchey sredy 2023» [State report «On the state and protection of the environment 2023»]. Available at: https://ecostr.ru/novosti/gosdoklad-o-sostoyanii-i-ob-ohrane-okruzhayushhej-sredy-2023/ (accessed 28.03.2025).Государственный доклад «О состоянии и об охране окружающей среды 2023». URL: https://ecostr.ru/novosti/gosdoklad-o-sostoyanii-i-ob-ohrane-okruzhayushhej-sredy-2023/ (дата обращения 28.03.2025).Bystrov S.O., Antonov I. A. First record of the four-eyed fir bark beetle Polygraphus proximus Blandford, 1894 (Coleoptera, Curculionidae: Scolytinae) from Irkutsk province, Russia. Entomological Review, 2019, v. 99, pp. 54–55.Bystrov S.O., Antonov I.A. First record of the four-eyed fir bark beetle Polygraphus proximus Blandford, 1894 (Coleoptera, Curculionidae: Scolytinae) from Irkutsk province, Russia // Entomological Review, 2019, v. 99, pp. 54–55.Dedyukhin S.V., Titova V.V. Finding of the bark beetle polygraphus proximus Blandford, 1894 (Coleoptera, Curculionidae: Scolytinae) in Udmurtia. Russian Journal of Biological Invasions, 2021, v. 12, pp. 258–263.Dedyukhin S.V., Titova V.V. Finding of the bark beetle polygraphus proximus Blandford, 1894 (Coleoptera, Curculionidae: Scolytinae) in Udmurtia // Russian J. of Biological Invasions, 2021, v. 12, pp. 258–263.Kerchev I.A., Mandelshtam M.Y., Krivets S.A., Ilinsky Y.Y. Small spruce bark beetle Ips amitinus (Eichhoff, 1872) (Coleoptera, Curculionidae: Scolytinae): a new alien species in West Siberia. Entomological review, 2019, v. 99, pp. 639–644.Kerchev I.A., Mandelshtam M.Y., Krivets S.A., Ilinsky Y.Y. Small spruce bark beetle Ips amitinus (Eichhoff, 1872) (Coleoptera, Curculionidae: Scolytinae): a new alien species in West Siberia // Entomological review, 2019, v. 99, pp. 639–644.Kerchev I.A., Krivets S.A., Bisirova E.M., Smirnov N.A. Distribution of the small spruce bark beetle Ips amitinus (Eichhoff, 1872) in Western Siberia. Russian J. of Biological Invasions, 2022, v. 13, pp. 58–63.Kerchev I.A., Krivets S.A., Bisirova E.M., Smirnov N.A. Distribution of the small spruce bark beetle Ips amitinus (Eichhoff, 1872) in Western Siberia // Russian J. of Biological Invasions, 2022, v. 13, pp. 58–63.Smirnov N.A, Kerchev I.A. Experimental rearing of the small spruce bark beetle Ips amitinus (Eichhoff, 1872) (Coleoptera: Curculionidae, Scolytinae) on aboriginal Siberian coniferous species. Acta Biologica Sibirica, 2024, v. 10, pp. 1699–1710. DOI: 10.5281/zenodo.14468138Smirnov N.A, Kerchev I.A. Experimental rearing of the small spruce bark beetle Ips amitinus (Eichhoff, 1872) (Coleoptera: Curculionidae, Scolytinae) on aboriginal Siberian coniferous species // Acta Biologica Sibirica, 2024, v. 10, pp. 1699–1710. DOI: 10.5281/zenodo.14468138Chang W.Y., Lantz V.A., Hennigar C R., MacLean D.A. Economic impacts of forest pests: a case study of spruce budworm outbreaks and control in New Brunswick, Canada. Canadian J. of Forest Research, 2012, v. 42, pp. 490–505.Chang W.Y., Lantz V.A., Hennigar C R., MacLean D.A. Economic impacts of forest pests: a case study of spruce budworm outbreaks and control in New Brunswick, Canada // Canadian J. of Forest Research, 2012, v. 42, pp. 490–505.Van Lierop P., Lindquist E., Sathyapala S., Franceschini G. Global Forest area disturbance from fire, insect pests, diseases and severe weather events. Forest Ecology and Management, 2015, v. 352, pp. 78–88.Van Lierop P., Lindquist E., Sathyapala S., Franceschini G. Global Forest area disturbance from fire, insect pests, diseases and severe weather events // Forest Ecology and Management, 2015, v. 352, pp. 78–88.Seidl, R., Schelhaas, M.J., Rammer, W., Verkerk, P.J. Increasing forest disturbances in Europe and their impact on carbon storage. Nature Climate Change, 2014, v. 4, pp. 806–810. DOI: 10.1038/nclimate2318Seidl, R., Schelhaas, M.J., Rammer, W., Verkerk, P.J. Increasing forest disturbances in Europe and their impact on carbon storage // Nature Climate Change, 2014, v. 4, pp. 806–810. DOI: 10.1038/nclimate2318Alekseev N.A., Shcherbakov A.N. Stvolovye vrediteli krupnomernogo posadochnogo materiala khvoynykh porod [Stem pests of large-sized planting material of coniferous trees]. Available аt: https://www.greeninfo.ru/protection_plants/pests_and_diseases.html/Article/_/aID/5095 (accessed 28.03.2025).Алексеев Н.А., Щербаков А.Н. Стволовые вредители крупномерного посадочного материала хвойных пород. URL: https://www.greeninfo.ru/protection_plants/pests_and_diseases.html/Article/_/aID/5095 (дата обращения 28.03.2025).Kobzar’ V.F., Kolesova N.I., Petrik A.A. Ussuriyskiy poligraf Polygraphus proximus Blandford, 1894 v pikhtarnikakh ekoparka «Ozera na Snezhnoy» (Irkutskaya oblast’) [Polygraphus proximus Blandford, 1894 in the fir forests of the «Lakes on Snezhnaya» Ecopark (Irkutsk Oblast)]. Fitosanitariya. Karantin rasteniy [Plant Health and Quarantine], 2023, v. 1, pp. 59-71. DOI: 10.69536/a4850-3897-3753-xКобзарь В.Ф., Колесова Н.И., Петрик А.А. Уссурийский полиграф Polygraphus proximus Blandford, 1894 в пихтарниках экопарка «Озера на Снежной» (Иркутская область) // Фитосанитария. Карантин растений, 2023. Т. 1. С. 59–71. DOI: 10.69536/a4850-3897-3753-xSenf C., Pflugmacher D., Zhiqiang Y., Sebald J., Knorn J., Neumann M., Hostert P., Seidl R. Canopy mortality has doubled in Europe’s temperate forests over the last three decades. Nature Communications, 2018, v. 9. DOI: 10.1038/s41467-018-07539-6Senf C., Pflugmacher D., Zhiqiang Y., Sebald J., Knorn J., Neumann M., Hostert P., Seidl R. Canopy mortality has doubled in Europe’s temperate forests over the last three decades // Nature Communications, 2018, v. 9. DOI: 10.1038/s41467-018-07539-6Krivets S.A., Bisirova E.M., Volkova E.S., Debkov N.M., Kerchev I.A., Mel’nik M.A., Nikiforov A.N., Chernova N.A. Tekhnologiya monitoringa pikhtovykh lesov v zone invazii ussuriyskogo poligrafa v Sibiri [Technology for monitoring fir forests in the Ussuri polygraph invasion zone in Siberia]. Tomsk: UMIUM, 2018, 74 p.Кривец С.А., Бисирова Э.М., Волкова Е.С., Дебков Н.М., Керчев И.А., Мельник М.А., Никифоров А.Н., Чернова Н.А. Технология мониторинга пихтовых лесов в зоне инвазии уссурийского полиграфа в Сибири. Томск: Изд-во УМИУМ, 2018. 74 с.Yarmishko V.T., Ignat’eva O.V., Evdokimov A. Nekotorye aspekty monitoringa sosnovykh lesov v ekstremal’nykh usloviyakh Kol’skogo Severa [Some aspects of scots pine forests monitoring in extreme conditions of the kola north]. Samarskiy nauchnyy vestnik [Samara Scientific Bulletin], 2019, v. 8, no 2, pp. 81-86. DOI:10.24411/2309-4370-2019-12115.Ярмишко В.Т., Игнатьева О.В., Евдокимов А. Некоторые аспекты мониторинга сосновых лесов в экстремальных условиях Кольского Севера // Самарский научный вестник, 2019. Т. 8. № 2. С. 81–86. DOI:10.24411/2309-4370-2019-12115Musolin D.L., Kirichenko N.I., Karpun N.N., Aksenenko E.V., Golub V.B., Kerchev I.A., Mandelshtam M.Y., Vasaitis R., Volkovitsh M.G., Zhuravleva E.N., Selikhovkin A.V. Invasive insect pests of forests and urban trees in Russia: Origin, pathways, damage, and management. Forests, 2022, v. 13, no. 4, pp. 521.Musolin D.L., Kirichenko N.I., Karpun N.N., Aksenenko E.V., Golub V.B., Kerchev I.A., Mandelshtam M.Y., Vasaitis R., Volkovitsh M.G., Zhuravleva E.N., Selikhovkin A.V. Invasive insect pests of forests and urban trees in Russia: Origin, pathways, damage, and management // Forests, 2022, v. 13, no. 4, pp. 521.Chenari A., Erfanifard Y., Dehghani M., Pourghasemi H.R. Woodland mapping at single-tree levels using object-oriented classification of unmanned aerial vehicle (UAV) images. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2017, v. XLII-4/W4, pp. 43–49.Chenari A., Erfanifard Y., Dehghani M., Pourghasemi H.R. Woodland mapping at single-tree levels using object-oriented classification of unmanned aerial vehicle (UAV) images // The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2017, v. XLII-4/W4, pp. 43–49.Safonova A., Tabik S., Alcaraz-Segura D., Rubtsov A., Maglinets Y., Herrera F. Detection of fir trees (Abies sibirica) damaged by the bark beetle in unmanned aerial vehicle images with deep learning. Remote Sensing, 2019, v. 11, no. 6, pp. 643.Safonova A., Tabik S., Alcaraz-Segura D., Rubtsov A., Maglinets Y., Herrera F. Detection of fir trees (Abies sibirica) damaged by the bark beetle in unmanned aerial vehicle images with deep learning // Remote Sensing, 2019, v. 11, no. 6, pp. 643.Kerchev I.A., Maslov K.A., Markov N.G., Tokareva O.S. Semanticheskaya segmentatsiya povrezhdennykh derev’ev pikhty na snimkakh s bespilotnykh letatel’nykh apparatov [Semantic segmentation of damaged fir trees in unmanned aerial vehicle images]. Sovremennye problemy distantsionnogo zondirovaniya Zemli iz kosmosa [Modern problems of remote sensing of the Earth from space], 2021, vol. 1, no. 1, pp. 116–126. DOI: 10.21046/2070-7401-2021-18-1-116-126.Керчев И.А., Маслов К.А., Марков Н.Г., Токарева О.С. Семантическая сегментация поврежденных деревьев пихты на снимках с беспилотных летательных аппаратов // Современные проблемы дистанционного зондирования Земли из космоса, 2021. Т. 18. № 1. С. 116–126. DOI: 10.21046/2070-7401-2021-18-1-116-126Zhou H., Yuan X., Zhou H., Shen H., Ma L., Sun L., Fang G., Sun H. Surveillance of pine wilt disease by high resolution satellite. J. of Forestry Research, 2022, v. 33, pp. 1401–1408.Zhou H., Yuan X., Zhou H., Shen H., Ma L., Sun L., Fang G., Sun H. Surveillance of pine wilt disease by high resolution satellite // J. of Forestry Research, 2022, v. 33, pp. 1401–1408.Zhou Y., Liu W., Bi H., Chen R., Zong S., Luo Y. A detection method for individual infected pine trees with pine wilt disease based on deep learning. Forests, 2022, v. 13, no. 11, p. 1880.Zhou Y., Liu W., Bi H., Chen R., Zong S., Luo Y. A detection method for individual infected pine trees with pine wilt disease based on deep learning // Forests, 2022, v. 13, no. 11, p. 1880.Markov N.G., Maslov K.A., Kerchev I.A., Tokareva O.S., Machuka K.R., Osipova V.V. Polnosvertochnye neyronnye seti dlya klassifikatsii povrezhdennykh derev’ev sosny sibirskoy kedrovoy na izobrazheniyakh s BPLA [Fully convolutional networks for classification of damaged pinus sibirica trees in uav imagery]. Regional’nye problemy distantsionnogo zondirovaniya Zemli: Materialy IX Mezhdunar. nauch. konf. [Regional Problems of Remote Sensing of the Earth. Proceedings of the IX International Scientific Conf.], Krasnoyarsk, September 13–16, 2022. Krasnoyarsk: Siberian Federal University, 2022, pp. 61–64.Марков Н.Г., Маслов К.А., Керчев И.А., Токарева О.С., Мачука К.Р., Осипова В.В. Полносверточные нейронные сети для классификации поврежденных деревьев сосны сибирской кедровой на изображениях с БПЛА // Региональные проблемы дистанционного зондирования Земли: Материалы IX Междунар. науч. конф. Красноярск, 13–16 сентября 2022. Красноярск: Изд-во СФУ, 2022. С. 61–64.Machuka K.R. Model’ polnosvertochnoy neyronnoy seti dlya semanticheskoy segmentatsii derev’ev pikhty na izobrazheniyakh s BPLA [A Fully Convolutional Neural Network Model for Semantic Segmentation of Fir Trees in UAV Images]. Molodezh’ i sovremennye informatsionnye tekhnologii: Materialy XXI Mezhdunar. nauch.-prakt. konf. studentov, aspirantov i molodykh uchenykh [Youth and modern information technologies. Proceedings of the XXI International scientific and practical conference of students, postgraduates and young scientists], Tomsk, April 15–18, 2024. Tomsk: Tomsk Polytechnic University, 2024, pp. 60–63.Мачука К.Р. Модель полносверточной нейронной сети для семантической сегментации деревьев пихты на изображениях с БПЛА // Молодежь и современные информационные технологии: Материалы XXI Междунар. науч.-практ. конф. студентов, аспирантов и молодых ученых. Томск, 15–18 апреля 2024 г. Томск: ТПУ, 2024. C. 60–63.Machuka K.R., Markov N.G. Modeli neyronnykh setey dlya segmentatsii snimkov derev’ev pikhty, poluchennykh s pomoshch'yu BPLA [Neural networks models for semantic segmentation of siberian fir trees in UAV images]. RRegional’nye problemy distantsionnogo zondirovaniya Zemli: Materialy XI Mezhdunar. nauch. konf. [Regional Problems of Remote Sensing of the Earth. Proceedings of the IX International Scientific Conf.], Krasnoyarsk, September 10–13, 2024. Krasnoyarsk: Siberian Federal University, 2024, pp. 344–347.Мачука К.Р., Марков Н.Г. Модели нейронных сетей для сегментации снимков деревьев пихты, полученных с помощью БПЛА // Региональные проблемы дистанционного зондирования Земли: Материалы XI Междунар. науч. конф. Красноярск, 10–13 сентября 2024 г. Красноярск: СФУ, 2024. С. 344–347.Li X., Wang A. Forest pest monitoring and early warning using UAV remote sensing and computer vision techniques. Sci Rep, 2025, v. 15. DOI: 10.1038/s41598-024-84464-3Li X., Wang A. Forest pest monitoring and early warning using UAV remote sensing and computer vision techniques // Sci Rep, 2025, v. 15. DOI: 10.1038/s41598-024-84464-3Luo Y., Huang H., Roques A. Early Monitoring of Forest Wood-Boring Pests with Remote Sensing. Annual review of entomology, 2023, v. 68, pp. 277–298. DOI: 10.1146/annurev-ento-120220-125410Luo Y., Huang H., Roques A. Early Monitoring of Forest Wood-Boring Pests with Remote Sensing // Annual review of entomology, 2023, v. 68, pp. 277–298. DOI: 10.1146/annurev-ento-120220-125410Maslov A.D. Metodicheskie rekomendatsii po nadzoru, uchetu i prognozu massovykh razmnozheniy stvolovykh vrediteley i sanitarnogo sostoyaniya lesov [Methodological recommendations for monitoring, recording and forecasting mass reproduction of stem pests and the sanitary condition of forests]. Pushkino: VNII lesovodstva i mekhanizatsii lesnogo khozyaystva, 2006, 68 p.Маслов А.Д. Методические рекомендации по надзору, учету и прогнозу массовых размножений стволовых вредителей и санитарного состояния лесов. Пушкино: Изд-во ВНИИ лесоводства и механизации лесного хозяйства, 2006. 68 с.Goodfellow I., Bengio Y., Courville A. Deep Learning. Cambridge, MA: MIT Press, 2016.Rahman M.A., Wang Y. Optimizing intersection-over-union in deep neural networks for image segmentation. Advances in Visual Computing, 2016, v. 10072, pp. 234–244. DOI: 10.1007/978-3-319-50835-1_22Rahman M.A., Wang Y. Optimizing intersection-over-union in deep neural networks for image segmentation // Advances in Visual Computing, 2016, v. 10072, pp. 234–244. DOI: 10.1007/978-3-319-50835-1_22Bertels J., Eelbode T., Berman M., Vandermeulen D., Maes F., Bisschops R., Blaschko M. Optimizing the dice score and jaccard index for medical image segmentation: theory & practice. Medical Image Computing and Computer Assisted Intervention, 2019, v. 11765. DOI: 10.1007/978-3-030-32245-8_11Bertels J., Eelbode T., Berman M., Vandermeulen D., Maes F., Bisschops R., Blaschko M. Optimizing the dice score and jaccard index for medical image segmentation: theory & practice // Medical Image Computing and Computer Assisted Intervention, 2019, v. 11765. DOI: 10.1007/978-3-030-32245-8_11.Markov N.G., Machuka K. Modeli i metody glubokogo obucheniya dlya resheniya zadach distantsionnogo monitoringa lesnykh resursov [Deep learning models and methods for solving the problems of remote monitoring of forest resources]. Izvestiya Tomskogo politekhnicheskogo universiteta. Inzhiniring georesursov [Bulletin of the Tomsk Polytechnic University. Geo Assets Engineering], 2024, v. 335, no. 6, pp. 55–74. DOI: 10.18799/24131830/2024/6/4600Марков Н.Г., Мачука К. Модели и методы глубокого обучения для решения задач дистанционного мониторинга лесных ресурсов // Известия Томского политехнического университета. Инжиниринг георесурсов, 2024. Т. 335. № 6. С. 55–74. DOI: 10.18799/24131830/2024/6/4600Kerchev I.A., Markov N.G., Machuca C.R., Tokareva O.S. Classification of pest-damaged coniferous trees in unmanned aerial vehicles images using convolutional neural network models. Computer Research and Modeling, 2024, v. 16, no. 5, pp. 1271–1294.Kerchev I.A., Markov N.G., Machuca C.R., Tokareva O.S. Classification of pest-damaged coniferous trees in unmanned aerial vehicles images using convolutional neural network models // Computer Research and Modeling, 2024, v. 16, no. 5, pp. 1271–1294.Ronneberger O., Fischer P., Brox T. U-Net: convolutional networks for biomedical image segmentation. Medical Image Computing and Computer-Assisted Intervention, 2015, v. 9351. DOI: 10.1007/978-3-319-24574-4_28.Ronneberger O., Fischer P., Brox T. U-Net: convolutional networks for biomedical image segmentation // Medical Image Computing and Computer-Assisted Intervention, 2015, v. 9351. DOI: 10.1007/978-3-319-24574-4_28.He K., Zhang X., Ren S., Sun J. Deep residual learning for image recognition. Computer Vision and Pattern Recognition, 2016, pp. 770–778.He K., Zhang X., Ren S., Sun J. Deep residual learning for image recognition // Computer Vision and Pattern Recognition, 2016, pp. 770–778.