<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Publishing DTD v1.3 20210610//EN" "JATS-journalpublishing1-3.dtd">
<article article-type="research-article" dtd-version="1.3" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xml:lang="ru"><front><journal-meta><journal-id journal-id-type="publisher-id">pribor</journal-id><journal-title-group><journal-title xml:lang="ru">Известия высших учебных заведений. Приборостроение</journal-title><trans-title-group xml:lang="en"><trans-title>Journal of Instrument Engineering</trans-title></trans-title-group></journal-title-group><issn pub-type="ppub">0021-3454</issn><issn pub-type="epub">2500-0381</issn><publisher><publisher-name>Национальный исследовательский университет ИТМО</publisher-name></publisher></journal-meta><article-meta><article-id pub-id-type="doi">10.17586/0021-3454-2022-65-3-194-203</article-id><article-id custom-type="elpub" pub-id-type="custom">pribor-233</article-id><article-categories><subj-group subj-group-type="heading"><subject>Research Article</subject></subj-group><subj-group subj-group-type="section-heading" xml:lang="ru"><subject>КОМПЬЮТЕРНОЕ ЗРЕНИЕ И ПЛАНИРОВАНИЕ ДВИЖЕНИЯ РОБОТОВ В ЗАДАЧАХ МАНИПУЛИРОВАНИЯ</subject></subj-group><subj-group subj-group-type="section-heading" xml:lang="en"><subject>COMPUTER VISION AND ROBOT MOVEMENT PLANNING IN MANIPULATION TASKS</subject></subj-group></article-categories><title-group><article-title>Сегментация объектов с функцией дообучения</article-title><trans-title-group xml:lang="en"><trans-title>Objects segmentation with retraining function</trans-title></trans-title-group></title-group><contrib-group><contrib contrib-type="author" corresp="yes"><name-alternatives><name name-style="eastern" xml:lang="ru"><surname>Ненахов</surname><given-names>И. Д.</given-names></name><name name-style="western" xml:lang="en"><surname>Nenakhov</surname><given-names>I. D.</given-names></name></name-alternatives><bio xml:lang="ru"><p>Ненахов Иван Дмитриевич — студент; факультет систем управления и робототехники, лаборатория биомехатроники и энергоэффективной робототехники.</p><p>Санкт-Петербург</p></bio><bio xml:lang="en"><p>Ivan D. Nenakhov — Student; ITMO University, Faculty of Control Systems and Robotics, International Laboratory of Biomechatronics and Energy-Efficient Robotics.</p><p>St. Petersburg</p></bio><email xlink:type="simple">ivdmne@gmail.com</email><xref ref-type="aff" rid="aff-1"/></contrib><contrib contrib-type="author" corresp="yes"><name-alternatives><name name-style="eastern" xml:lang="ru"><surname>Артемов</surname><given-names>К.</given-names></name><name name-style="western" xml:lang="en"><surname>Artemov</surname><given-names>K.</given-names></name></name-alternatives><bio xml:lang="ru"><p>Артемов Кирилл — аспирант; факультет систем управления и робототехники, лаборатория биомехатроники и энергоэффективной робототехники; инженер-исследователь.</p><p>Санкт-Петербург</p></bio><bio xml:lang="en"><p>Kirill Artemov — Post-Graduate Student; ITMO University, Faculty of Control Systems and Robotics, International Laboratory of Biomechatronics and Energy-Efficient Robotics; Engineer-Researcher.</p><p>St. Petersburg</p></bio><email xlink:type="simple">kaartemov@itmo.ru</email><xref ref-type="aff" rid="aff-1"/></contrib><contrib contrib-type="author" corresp="yes"><name-alternatives><name name-style="eastern" xml:lang="ru"><surname>Забихифар</surname><given-names>С.</given-names></name><name name-style="western" xml:lang="en"><surname>Zabihifar</surname><given-names>S.</given-names></name></name-alternatives><bio xml:lang="ru"><p>Забихифар СейедХассан — канд. техн. наук; лаборатория робототехники; инженер-разработчик.</p><p>Москва</p></bio><bio xml:lang="en"><p>Seyedhassan Zabihifar — PhD; Sberbank, Robotics Laboratory; Engineer-Designer.</p><p>Moscow</p></bio><email xlink:type="simple">zabikhifar.s@sberbank.ru</email><xref ref-type="aff" rid="aff-2"/></contrib><contrib contrib-type="author" corresp="yes"><name-alternatives><name name-style="eastern" xml:lang="ru"><surname>Семочкин</surname><given-names>А. Н.</given-names></name><name name-style="western" xml:lang="en"><surname>Semochkin</surname><given-names>A. N.</given-names></name></name-alternatives><bio xml:lang="ru"><p>Семочкин Александр Николаевич — канд. физ.-мат. наук, доцент; лаборатория робототехники; гл. инженер-разработчик.</p><p>Москва</p></bio><bio xml:lang="en"><p>Aleksandr N. Semochkin — PhD, Associate Professor; Sberbank, Robotics Laboratory; Сhief Engineer-Designer.</p><p>Moscow</p></bio><email xlink:type="simple">Semochkin.A.N@sberbank.ru</email><xref ref-type="aff" rid="aff-2"/></contrib><contrib contrib-type="author" corresp="yes"><name-alternatives><name name-style="eastern" xml:lang="ru"><surname>Колюбин</surname><given-names>С. А.</given-names></name><name name-style="western" xml:lang="en"><surname>Kolyubin</surname><given-names>S. A.</given-names></name></name-alternatives><bio xml:lang="ru"><p>Колюбин Сергей Алексеевич — д-р техн. наук, доцент; факультет систем управления и робототехники, лаборатория биомехатроники и энергоэффективной робототехники; вед. научный сотрудник.</p><p>Санкт-Петербург</p></bio><bio xml:lang="en"><p>Sergey A. Kolyubin — Dr. Sci., Associate Professor; ITMO University, Faculty of Control Systems and Robotics, International Laboratory of Biomechatronics and Energy-Efficient Robotics; Leading Researcher.</p><p>St. Petersburg</p></bio><email xlink:type="simple">s.kolyubin@itmo.ru</email><xref ref-type="aff" rid="aff-1"/></contrib></contrib-group><aff-alternatives id="aff-1"><aff xml:lang="ru"><institution>Университет ИТМО</institution></aff><aff xml:lang="en"><institution>ITMO University</institution></aff></aff-alternatives><aff-alternatives id="aff-2"><aff xml:lang="ru"><institution>Сбербанк</institution></aff><aff xml:lang="en"><institution>Sberbank</institution></aff></aff-alternatives><pub-date pub-type="collection"><year>2022</year></pub-date><pub-date pub-type="epub"><day>01</day><month>12</month><year>2024</year></pub-date><volume>65</volume><issue>3</issue><fpage>194</fpage><lpage>203</lpage><permissions><copyright-statement>Copyright &amp;#x00A9; Национальный исследовательский университет ИТМО, 2024</copyright-statement><copyright-year>2024</copyright-year><copyright-holder xml:lang="ru">Национальный исследовательский университет ИТМО</copyright-holder><copyright-holder xml:lang="en">Национальный исследовательский университет ИТМО</copyright-holder><license xlink:href="https://pribor.ifmo.ru/jour/about/submissions#copyrightNotice" xlink:type="simple"><license-p>https://pribor.ifmo.ru/jour/about/submissions#copyrightNotice</license-p></license></permissions><self-uri xlink:href="https://pribor.ifmo.ru/jour/article/view/233">https://pribor.ifmo.ru/jour/article/view/233</self-uri><abstract><p>Рассматриваются способы расширения набора распознаваемых классов объектов для задачи их сегментирования, где необходимо построить маску объекта, а также узнать его класс. Для первой задачи использованы методы, не зависящие от классов предметов и являющиеся наиболее устойчивыми к изменениям формы; для второй задачи проанализированы методы, основанные на итеративном обучении (iterative learning), и методы метрического обучения (metric learning). Второй подход выбран в качестве основного, и для него протестированы различные архитектуры нейронных сетей. Проведена классификация объектов с использованием алгоритма k ближайших соседей. В качестве набора данных для обучения нейронной сети использован набор COIL-100, а после обученная модель протестирована на собственном наборе данных. Проведенные эксперименты показывают, что используемый метод позволяет обрабатывать 7-8 изображений в секунду на видеокарте GTX 1050 ti с 4 Гбайт видеопамяти с точностью классификации в 99 %.</p></abstract><trans-abstract xml:lang="en"><p>Ways to expand the set of recognized object classes for the task of segmenting them, where it is necessary to build an object mask, as well as to find out its class, are considered. For the first task, methods that do not depend on the classes of subjects and are the most resistant to shape changes were used; for the second task, methods based on iterative learning and methods of metric learning are analyzed. The second approach is chosen as the main one, and various neural network architectures are tested for it. The classification of objects using the k nearest neighbors algorithm is carried out. The COIL-100 set is used as a data set for training a neural network, and after that the trained model was tested on its own data set. The experiments show that the method used allows processing 7-8 images per second on a GTX 1050 ti graphics card with 4 GB of video memory with a classification accuracy of 99%.</p></trans-abstract><kwd-group xml:lang="ru"><kwd>метрическое обучение</kwd><kwd>итеративное обучение</kwd><kwd>сегментация</kwd><kwd>классификация</kwd><kwd>сверточные нейронные сети</kwd><kwd>робототехника</kwd><kwd>распознавание изображений</kwd></kwd-group><kwd-group xml:lang="en"><kwd>metric learning</kwd><kwd>iterative learning</kwd><kwd>segmentation</kwd><kwd>classification</kwd><kwd>convolutional neural networks</kwd><kwd>robotics</kwd><kwd>image recognition</kwd></kwd-group></article-meta></front><back><ref-list><title>References</title><ref id="cit1"><label>1</label><citation-alternatives><mixed-citation xml:lang="ru">Krizhevsky A., Sutskever I., Hinton G. E. ImageNet Classification with Deep Convolutional Neural Networks // Advances in Neural Information Processing Systems / Ed.: F. Pereira, C. J. C. Burges, L. Bottou, K. Q. Weinberger. Curran Associates Inc., 2012. Vol. 25. P. 1097—1105.</mixed-citation><mixed-citation xml:lang="en">Krizhevsky A., Sutskever I., Hinton G.E. Advances in Neural Information Processing Systems, 2012, vol. 25, pp. 1097–1105, https://proceedings.neurips.cc/paper/2012/file/c399862d3b9d6b76c8436e924a68c45b-Paper.pdf.</mixed-citation></citation-alternatives></ref><ref id="cit2"><label>2</label><citation-alternatives><mixed-citation xml:lang="ru">Deep Residual Learning for Image Recognition / Kaiming He, Xi-Angyu Zhang, Shaoqing Ren, Jian Sun // CoRR. 2015. Vol. ab-s/1512.03385. [Электронный ресурс]: http://arxiv.org/abs/1512.03385.</mixed-citation><mixed-citation xml:lang="en">He K., Zhang X., Ren Sh., Sun J. CoRR, 2015, vol. abs/1512.03385. http://arxiv.org/abs/1512.03385.</mixed-citation></citation-alternatives></ref><ref id="cit3"><label>3</label><citation-alternatives><mixed-citation xml:lang="ru">MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications / A. G. Howard, Menglong Zhu, Bo Chen et al. // CoRR. 2017. Vol. abs/1704.04861. [Электронный ресурс]: http://arxiv.org/abs/1704.04861.</mixed-citation><mixed-citation xml:lang="en">Howard A.G., Zhu M., Chen B. et al. CoRR, 2017, vol. abs/1704.04861, http://arxiv.org/abs/1704.04861.</mixed-citation></citation-alternatives></ref><ref id="cit4"><label>4</label><citation-alternatives><mixed-citation xml:lang="ru">Mask R-CNN / Kaiming He, G. Gkioxari, P. Doll ar, R. B. Girshick // CoRR. 2017. Vol. abs/1703.06870. [Электронный ресурс]: http://arxiv.org/abs/1703.06870.</mixed-citation><mixed-citation xml:lang="en">He K., Gkioxari G., Doll ar P., Girshick R.B. CoRR, 2017, vol. abs/1703.06870, http://arxiv.org/abs/1703.06870.</mixed-citation></citation-alternatives></ref><ref id="cit5"><label>5</label><citation-alternatives><mixed-citation xml:lang="ru">Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks / Shaoqing Ren, Kaiming He, R. Girshick, Jian Sun // Advances in Neural Information Processing Systems / Ed.: C. Cortes, N. Lawrence, D. Lee et al. Curran Associates Inc., 2015. Vol. 28. P. 91—99. [Электронный ресурс]: https://proceedings.neurips.cc/paper/2015/file/14bfa6bb14875e45bba028a21ed38046-Paper.pdf.15.</mixed-citation><mixed-citation xml:lang="en">Ren Sh., He K., Girshick R., Sun J. Advances in Neural Information Processing Systems, 2015, vol. 28, pp. 91–99, https://proceedings.neurips.cc/paper/2015/file/14bfa6bb14875e45bba028a21ed38046-Paper.pdf.15.</mixed-citation></citation-alternatives></ref><ref id="cit6"><label>6</label><citation-alternatives><mixed-citation xml:lang="ru">Kirkpatrick J., Pascanu R., Rabinowitz N. et al. Overcoming catastrophic forgetting in neural networks // Proc. of the National Academy of Sciences. 2017. N 114(13). P. 3521—3526.</mixed-citation><mixed-citation xml:lang="en">Kirkpatrick J., Pascanu R., Rabinowitz N., Veness J., Desjardins G., Rusu A.A., Milan K., Quan J., Ramalho T., Grabska-Barwinska A., Hassabis D., Clopath C., Kumaran D., and Hadsell R. Overcoming catastrophic forgetting in neural networks, 2017, no. 13(114), pp. 3521–3526.</mixed-citation></citation-alternatives></ref><ref id="cit7"><label>7</label><citation-alternatives><mixed-citation xml:lang="ru">Zenke F., Poole B., Ganguli S. Continual Learning Through Synaptic Intelligence // Proc. of the 34th Intern. Conf. on Machine Learning. Sydney, Australia. 2017. Vol. 70. P. 3987—3995,</mixed-citation><mixed-citation xml:lang="en">Zenke F., Poole B., and Ganguli S. Proceedings of the 34th International Conference on Machine Learning, Sydney, Australia, 2017, vol. 70, pр. 3987–3995.</mixed-citation></citation-alternatives></ref><ref id="cit8"><label>8</label><citation-alternatives><mixed-citation xml:lang="ru">Lomonaco V., Maltoni D. CORe50: A New Dataset and Benchmark for Continuous Object Recognition // Proc. of the 1st Annual Conf. on Robot Learning. PMLR. 2017. Vol. 78. P. 17—26.</mixed-citation><mixed-citation xml:lang="en">Lomonaco V. and Maltoni D. Proceedings of the 1st Annual Conference on Robot Learning, PMLR, 2017, vol. 78, Proceedings of Machine Learning Research, pр. 17–26.</mixed-citation></citation-alternatives></ref><ref id="cit9"><label>9</label><citation-alternatives><mixed-citation xml:lang="ru">Progressive Neural Networks / A. A. Rusu, N. C. Rabinowitz, G. Desjardins, H. Soyer, J. Kirkpatrick, K. Kavukcuoglu, R. Pascanu, R. Hadsell. // arXiv preprint arXiv:1606.04671, 2016.</mixed-citation><mixed-citation xml:lang="en">Rusu A.A., Rabinowitz N.C., Desjardins G., Soyer H., Kirkpatrick J., Kavukcuoglu K., Pascanu R., and Hadsell R. arXiv preprint, arXiv:1606.04671, 2016.</mixed-citation></citation-alternatives></ref><ref id="cit10"><label>10</label><citation-alternatives><mixed-citation xml:lang="ru">Hayes T. L., Cahill N. D., Kanan C. Memory Efficient Experience Replay for Streaming Learning // arXiv preprint arXiv:1809.05922, 2018.</mixed-citation><mixed-citation xml:lang="en">Hayes T.L., Cahill N.D., and Kanan Ch. arXiv preprint, arXiv:1809.05922, 2018.</mixed-citation></citation-alternatives></ref><ref id="cit11"><label>11</label><citation-alternatives><mixed-citation xml:lang="ru">Rebuffi S. A., Kolesnikov A., Sperl G., Lampert C. H. iCaRL: Incremental Classifier and Representation Learning // IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), Honolulu, Hawaii. 2017.</mixed-citation><mixed-citation xml:lang="en">Rebuffi S.-a., Kolesnikov A., Sperl G., and Lampert Ch.H. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, Hawaii, 2017.</mixed-citation></citation-alternatives></ref><ref id="cit12"><label>12</label><citation-alternatives><mixed-citation xml:lang="ru">Hoffer E., Nir A. Deep metric learning using Triplet network // Intern. Workshop on Similarity-Based Pattern Recognition. Springer, Cham, 2015.</mixed-citation><mixed-citation xml:lang="en">Hoffer E. and Nir A. International Workshop on Similarity-Based Pattern Recognition, Springer, Cham, 2015.</mixed-citation></citation-alternatives></ref><ref id="cit13"><label>13</label><citation-alternatives><mixed-citation xml:lang="ru">Columbia Object Image Library (COIL-100) / S. A. Nene, S. K. Nayar, H. Murase. // Tech. Report CUCS-006-96. 1996. February.</mixed-citation><mixed-citation xml:lang="en">Nene S.A., Nayar S.K., and Murase H. Columbia Object Image Library (COIL-100), Technical Report CUCS-006-96, February 1996.</mixed-citation></citation-alternatives></ref><ref id="cit14"><label>14</label><citation-alternatives><mixed-citation xml:lang="ru">Wu Yuxin. Detectron2, 2019. [Электронный ресурс]: https://github.com/facebookresearch/detectron2.</mixed-citation><mixed-citation xml:lang="en">Wu Y. Detectron2, https://github.com/facebookresearch/detectron2, 2019.</mixed-citation></citation-alternatives></ref><ref id="cit15"><label>15</label><citation-alternatives><mixed-citation xml:lang="ru">Pytorch: An Imperative Style, High-Performance Deep Learning Library / A. Paszke, S. Gross, F. Massa et al. // Advancesin Neural Information Processing Systems 32 / Ed.: H. Wallach,H. Larochelle, A. Beygelzimer et al. Curran Associates Inc., 2019. P. 8024—8035.</mixed-citation><mixed-citation xml:lang="en">Paszke A., Gross S., Massa F. et al. Advancesin Neural Information Processing Systems 32, Curran Associates, Inc., 2019, pp. 8024–8035.</mixed-citation></citation-alternatives></ref><ref id="cit16"><label>16</label><citation-alternatives><mixed-citation xml:lang="ru">Musgrave K. ., Belongie S., Lim S.-N. Pytorch metric learning // arXiv preprint, arXiv:2008.09164, 2020.</mixed-citation><mixed-citation xml:lang="en">Musgrave K., Belongie S., and Lim S.-N. arXiv preprint, arXiv:2008.09164, 2020.</mixed-citation></citation-alternatives></ref></ref-list><fn-group><fn fn-type="conflict"><p>The authors declare that there are no conflicts of interest present.</p></fn></fn-group></back></article>
