intechnguВестник НГУ. Серия: Информационные технологииVestnik NSU. Series: Information Technologies1818-79002410-0420НГУ10.25205/1818-7900-2022-20-1-47-56intechngu-189Research ArticleСтатьиБиологически-подобные модели сверточных нейронов в задаче распознавания иллюзорного контураBio-Inspired Models of Convolution Neurons in the Problem of Illusory Contour Recognitionhttps://orcid.org/0000-0002-6676-0518КугаевскихА. В.KugaevskikhA. V.

Александр Владимирович Кугаевских, кандидат технических наук

Новосибирск

Alexander V. Kugaevskikh, Candidate of Technical Sciences

Novosibirsk

a-kugaevskikh@yandex.ruБерьяновМ. С.BeryanovM. S.

Максим Сергеевич Берьянов, магистрант

Санкт-Петербург

Maxim S. Berenov, Master’s Student

St. Petersburg

berjnov.ru@mail.ruНовосибирский государственный университетРоссияNovosibirsk State UniversityRussian FederationУниверситет ИТМОРоссияITMO UniversityRussian Federation2022170520222014756Copyright © Кугаевских А.В., Берьянов М.С., 20222022Кугаевских А.В., Берьянов М.С.Kugaevskikh A.V., Beryanov M.S.Данная работа распространяется под лицензией Creative Commons Attribution 4.0.This work is licensed under a Creative Commons Attribution 4.0 License.https://intechngu.elpub.ru/jour/article/view/189

Представлен результат проектирования архитектуры нейронной сети на биологически-подобных нейронах, в задачу которой входит отработка механизма распознавания иллюзорного контура на примере «фигур Канижа». Нейронная сеть позволила добиться инвариантности к количеству углов фигуры и не теряет в качестве распознавания при изменении размеров иллюзорного контура. Основное применение подхода может быть найдено в задаче разделения «фигура – фон» на изображениях.

This paper presents the result of designing the architecture of a neural network on bio-inspired neurons, whose task is to work out the mechanism for recognizing an illusory contour using the example of “Kanizsa’s figures”. The neural network made it possible to achieve invariance to the number of corners of the figure and does not lose recognition quality when changing the size of the illusory contour. The main application of the approach can be found in the problem of separating “figure-background” in images.

иллюзорный контурнейронная сетьбиологически-подобная модельвыделение краевнейроны конца линийсегментацияillusory contourneural networkbio-inspired modeledge detectionend-stopped neuronsegmentationReferencesHeydt R.von der, Peterhans E., Baumgartner G. Illusory Contours and Cortical Neuron Responses. Science, 1984, vol. 224, no. 4654, pp. 1260–1262.Heydt R.von der, Peterhans E., Baumgartner G. Illusory Contours and Cortical Neuron Responses. Science, 1984, vol. 224, no. 4654, pp. 1260–1262.Finkel L., Edelman G. Integration of distributed cortical systems by reentry: a computer simulation of interactive functionally segregated visual areas. J. Neurosci., 1989, vol. 9, no. 9, pp. 3188–3208.Finkel L., Edelman G. Integration of distributed cortical systems by reentry: a computer simulation of interactive functionally segregated visual areas. J. Neurosci., 1989, vol. 9, no. 9, pp. 3188–3208.Guy G., Medioni G. Inferring global perceptual contours from local features. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. New York, NY, IEEE Comput. Soc. Press, 1993, pp. 786–787.Guy G., Medioni G. Inferring global perceptual contours from local features. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. New York, NY, IEEE Comput. Soc. Press, 1993, pp. 786–787.Williams, Hanson. Perceptual completion of occluded surfaces. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition CVPR-94. Seattle, WA, IEEE Comput. Soc. Press, 1994, pp. 104–112.Williams, Hanson. Perceptual completion of occluded surfaces. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition CVPR-94. Seattle, WA, IEEE Comput. Soc. Press, 1994, pp. 104–112.Gove A., Grossberg S., Mingolla E. Brightness perception, illusory contours, and corticogeniculate feedback. Vis Neurosci., 1995, vol. 12, no. 6, pp. 1027–1052.Gove A., Grossberg S., Mingolla E. Brightness perception, illusory contours, and corticogeniculate feedback. Vis Neurosci., 1995, vol. 12, no. 6, pp. 1027–1052.Williams L. R., Jacobs D. W. Stochastic completion fields: a neural model of illusory contour shape and salience. In: Proceedings of IEEE International Conference on Computer Vision. Cambridge, MA, IEEE Comput. Soc. Press, 1995, pp. 408–415.Williams L. R., Jacobs D. W. Stochastic completion fields: a neural model of illusory contour shape and salience. In: Proceedings of IEEE International Conference on Computer Vision. Cambridge, MA, IEEE Comput. Soc. Press, 1995, pp. 408–415.Yankelovich A., Spitzer H. Predicting Illusory Contours Without Extracting Special Image Features. Front. Comput. Neurosci., 2019, vol. 12, p. 106.Yankelovich A., Spitzer H. Predicting Illusory Contours Without Extracting Special Image Features. Front. Comput. Neurosci., 2019, vol. 12, p. 106.Kugaevskikh A., Sogreshilin A. Analyzing the Efficiency of Segment Boundary Detection Using Neural Networks. Optoelectronics Instrumentation and Data Processing, 2019, vol. 55, no. 4, pp. 414–422. DOI 10.3103/S8756699019040137Kugaevskikh A., Sogreshilin A. Analyzing the Efficiency of Segment Boundary Detection Using Neural Networks. Optoelectronics Instrumentation and Data Processing, 2019, vol. 55, no. 4, pp. 414–422. DOI 10.3103/S8756699019040137Kugaevskikh A. Bio-Inspired End-Stopped Neuron Model for the Curves Segmentation. In: Proceedings of 2020 International Russian Automation Conference (RusAutoCon). Sochi, IEEE, 2020, pp. 719–724. DOI 10.1109/RusAutoCon49822.2020.9208069Kugaevskikh A. Bio-Inspired End-Stopped Neuron Model for the Curves Segmentation. In: Proceedings of 2020 International Russian Automation Conference (RusAutoCon). Sochi, IEEE, 2020, pp. 719–724. DOI 10.1109/RusAutoCon49822.2020.9208069Heitger F. et al. Simulation of neural contour mechanisms: from simple to end-stopped cells: 5. Vision Research, 1992, vol. 32, no. 5, pp. 963–981.Heitger F. et al. Simulation of neural contour mechanisms: from simple to end-stopped cells: 5. Vision Research, 1992, vol. 32, no. 5, pp. 963–981.

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