Extraction of features in images based on integral transformations in solving problems of classification of fragments of photographs

A. A. Klyachin

Izvestiya of Saratov University. Mathematics. Mechanics. Informatics, Tome 24 (2024) no. 3, pp. 432-441.

Résumé

The article proposes a method for calculating features in an image to form a training data set for solving various problems of video image classification. This method involves the use of well-known integral transformations  — the Radon transform and the Steklov function. The proposed method is compared with convolutional neural networks both in terms of the percentage of correct prediction and in terms of its execution time. As a test task, the problem of finding a fragment of a photograph containing an image of a car is considered.

Bibliographie

[1] R. Gonsales, R. Vuds, Tsifrovaya obrabotka izobrazhenii, Tekhnosfera, M., 2005, 1072 pp.

[2] S. M. Borzov, A. O. Potaturkin, O. I. Potaturkin, A. M. Fedotov, “Issledovanie effektivnosti klassifikatsii giperspektralnykh sputnikovykh izobrazhenii prirodnykh i antropogennykh territorii”, Avtometriya, 52:1 (2016), 3–14 | DOI | MR

[3] A. Yu. Dorogov, R. G. Kurbanov, V. V. Razin, “Bystraya klassifikatsiya JPEG-izobrazhenii”, Internet-matematika 2005. Avtomaticheskaya obrabotka veb-dannykh (Moskva, 2005), 147–172

[4] V. V. Osokin, T. D. Aipov, Z. A. Niyazova, “O klassifikatsii izobrazhenii i muzykalnykh failov”, Intellektualnye sistemy. Teoriya i prilozheniya, 19:1 (2015), 49–70, UWLELY, EDN | MR

[5] E. M. Kupenova, A. V. Kashnitskii, “Metod sluchainykh lesov v zadachakh klassifikatsii sputnikovykh snimkov”, Vestnik Tverskogo gosudarstvennogo universiteta. Seriya: Geografiya i geoekologiya, 2018, no. 3, 99–107

[6] E. F. Goncharova, A. V. Gaidel, “Metody otbora priznakov dlya zadach klassifikatsii izobrazhenii zemnoi poverkhnosti”, Informatsionnye tekhnologii i nanotekhnologii, ITNT-2017, sb. tr. III mezhdunarodnoi konferentsii i molodezhnoi shkoly, Novaya tekhnika, Samara, 2017, 535–540

[7] M. A. Turkova, A. V. Gaidel, “Korrelyatsionnye priznaki dlya klassifikatsii teksturnykh izobrazhenii”, Informatsionnye tekhnologii i nanotekhnologii, sb. tr. IV mezhdunarodnoi konferentsii i molodezhnoi shkoly, Novaya tekhnika, Samara, 2018, 595–599

[8] A. I. Chulichkov, I. V. Morozova, “Klassifikatsiya razmytykh izobrazhenii i otsenka parametrov sistemy registratsii metodami morfologicheskogo analiza”, Intellektualnye sistemy, 9:1-4 (2005), 321–344

[9] S. M. Borzov, O. I. Potaturkin, “Klassifikatsiya tipov rastitelnogo pokrova po giperspek tralnym dannym distantsionnogo zondirovaniya zemli”, Vestnik Novosibirskogo gosudar stvennogo universiteta. Seriya: Informatsionnye tekhnologii, 12:4 (2014), 13–22

[10] I. A. Pestunov, S. A. Rylov, P. V. Melnikov, “Klassifikatsiya giperspektralnykh izobrazhenii vysokogo prostranstvennogo razresheniya”, Journal of Siberian Federal University. Engineering Technologies, 11:1 (2018), 69–76 | DOI | MR

[11] D. N. Kitaev, Sravnenie svertochnoi neironnoi seti i metoda glavnykh komponent v zadache klassifikatsii teksturnykh izobrazhenii, tez. dokl., ed. A. B. Prokofev, Izd-vo Samarskogo un-ta, Samara, 2018, 88–89

[12] A. M. Golubkov, “Binarnaya klassifikatsiya izobrazhenii na primere zadachi raspoznavaniya lits”, Izvestiya SPbGETU «LETI», 2018, no. 7, 26–30

[13] E. S. Nezhevenko, A. S. Feoktistov, O. Yu. Dashevskii, “Neirosetevaya klassifikatsiya giperspektralnykh izobrazhenii na osnove preobrazovaniya Gilberta–Khuanga”, Avtometriya, 53:2 (2017), 79–84 | DOI

[14] V. D. Vaskan, “Obzor arkhitektur svertochnykh neironnykh setei dlya zadachi klassifikatsii izobrazhenii”, IT-Standart, 2021, no. 3 (28), 34–39

[15] O. P. Soldatova, A. A. Garshin, “Primenenie svertochnoi neironnoi seti dlya raspoznavaniya rukopisnykh tsifr”, Kompyuternaya optika, 34:2 (2010), 252–260

[16] Le Man Kha, “Svertochnaya neironnaya set dlya resheniya zadachi klassifikatsii”, Trudy Moskov skogo fiziko-tekhnicheskogo instituta (NIU), 8:3 (2016), 91–97

[17] I. S. Azarov, A. S. Prokopenya, “Svertochnye neironnye seti dlya raspoznavaniya izobrazhenii”, Big Data and Advanced Analytics, 2020, no. 6-1, 271–280

[18] O. S. Sikorskii, “Obzor svertochnykh neironnykh setei dlya zadachi klassifikatsii izobrazhenii”, Novye informatsionnye tekhnologii v avtomatizirovannykh sistemakh, 2017, no. 20, 37–42

[19] S. R. Deans, S. Roderick, The Radon Transform and Some of its Applications, John Wiley Sons, New York, 1983, 289 pp. | MR | Zbl

[20] V. V. Zhuk, V. F. Kuzyutin, Approksimatsiya funktsii i chislennoe integrirovanie, Izd-vo Sankt-Peterburgskogo un-ta, Sankt-Peterburg, 1995, 352 pp. | MR

[21] V. V. Voronina, A. V. Mikheev, N. G. Yarushkina, K. V. Svyatov, Teoriya i praktika mashinnogo obucheniya, ucheb. posobie, UlGTU, Ulyanovsk, 2017, 290 pp.

[22] Sholle F., Glubokoe obuchenie na Python, Biblioteka programmista, Piter, Sankt-Peterburg, 2018, 400 pp.