350 rub
Journal Neurocomputers №4 for 2018 г.
Article in number:
Evaluation of technical means of rehabilitation and selection of the architecture of the construction neural network for the establishment of the support system for people in the urban environment for people with disabled vision opportunities
Type of article: scientific article
UDC: 004.89
Keywords: people with visual disabilities accessible environment universal design technical means of rehabilitation analy- tic hierarchy process decision support methods convolutional neural network objects recognition in real time.
Authors:

V.I. Terekhov – Ph.D (Eng.), Associate Professor, Department of Computer Systems and Networks, Bauman Moscow State Technical University

E-mail: terekchow@bmstu.ru

V.S. Ilin – Student, Department of Computer Systems and Networks, Bauman Moscow State Technical University

E-mail: ivs@bmstu.ru

I.A. Mogilnikov – Student, Department of Computer Systems and Networks, Bauman Moscow State Technical University

E-mail: mogilnikoff2@gmail.com

M.S. Popova – Student, Department of Computer Systems and Networks, Bauman Moscow State Technical University

E-mail: popovams@bmstu.ru 

Abstract:

In the article, a review and analysis of the existing on the market technical rehabilitation tools that solve the problem of moving people with limited vision health in the urban environment. To determine the prospects for the development of a support system for movement in an urban environment for people with limited vision health, an expert assessment of existing solutions was carried out by the hierarchy analysis method, taking into account the functional scheme of the proposed solution, and the choice of architecture and parameters of the convolutional- neural network for solving the problem of recognition of public terrestrial transport.

Pages: 41-49
References
  1. Informacionnyj byulleten' vsemirnoj organizacii zdorov'ya (Narusheniya zreniya i slepota). Rezhim dostupa: http://www.who.int/mediacentre/ factsheets/fs282/en/ (data obrashcheniya 14.10.2017).
  2. Otchet direktora Departamenta po delam invalidov Mintruda Rossii Grigoriya Lekareva ot 24 dekabrya 2014 (kolichestvo slepyh i slabovidyashchih v Rossii). Rezhim dostupa: http://www.rosmintrud.ru/social/invalid-defence/250 (data obrashcheniya 09.10.2017).
  3. Il'in V.S., Popova M.S. Vyyavlenie potrebnostej slepogluhih lyudej dlya proektirovaniya vspomogatel'nyh tekhnologij i ustrojstv // Novaya nauka: teoreticheskij i prakticheskij vzglyad. Sterlitamak: AMI. 2017. № 1(1). S. 218–221.
  4. Proekt Oriense (informaciya ob ustrojstve). Rezhim dostupa: http://oriense.com/ (data obrashcheniya 29.11.2017).
  5. Nosimoe ustrojstvo Horus (informaciya ob ustrojstve). Rezhim dostupa: http://horus.tech/?l=en_us (data obrashcheniya 29.11.2017).
  6. Proekt «Govoryashchij gorod» (informaciya o sisteme) Rezhim dostupa: https://www.speakingcity.org/ (data obrashcheniya 29.11.2017).
  7. Proekt OrCam (informaciya ob ustrojstve). Rezhim dostupa: http://www.orcam.com/ (data obrashcheniya 29.11.2017).
  8. Prilozhenie Seeing AI (informaciya o prilozhenii). Rezhim dostupa: https://itunes.apple.com/us/app/seeing-ai/id999062298 (data obrashcheniya 29.11.2017).
  9. BrainPort V100 (informaciya ob ustrojstve). Rezhim dostupa: http://evercare.ru/brainport-v100 (data obrashcheniya 29.11.2017).
  10. Kompleks gadzhetov dlya nezryachih (informaciya ob ustrojstve). Rezhim dostupa: http://www.smartaids.ru/articles/blog.php?page=post&blog=Dostupnaya_ sreda&post_id=vunderkind-sozdal-gadzhet-dlya-nezryachikh (data obrashcheniya 29.11.2017).
  11. vOICe (informaciya o versii programmy dlya OS Windows). Rezhim dostupa: https://www.seeingwithsound.com/winvoice.htm (data obrashcheniya 29.11.2017).
  12. Saati T.L. Prinyatie reshenij pri zavisimostyah i obratnyh svyazyah. URSS. 2010.
  13. Postnikov V.M., CHernen'kij V.M. Metody prinyatiya reshenij v sistemah organizacionnogo upravleniya. M.: Izd-vo MGTU im. N.EH. Baumana. 2014. 205 s.
  14. Shaoqing Ren, Kaiming He, Ross Girshick, Jian Sun. Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks. Rezhim dostupa: https://arxiv.org/pdf/1506.01497v1.pdf (data obrashcheniya 07.04.2018).
  15. Wei Liu, Dragomir Anguelov, Dumitru Erhan, Christian Szegedy, Scott Reed4, Cheng-Yang Fu1, Alexander C. Berg. SSD: Single Shot MultiBox Detector. Rezhim dostupa: https://arxiv.org/pdf/1512.02325.pdf (data obrashcheniya 07.04.2018).
  16. Christian Szegedy, Vincent Vanhoucke, Sergey Ioffe, Jonathon Shlens, Zbigniew Wojna. Rethinking the Inception Architecture for Computer Vision. Rezhim dostupa: https://arxiv.org/pdf/1512.00567.pdf (data obrashcheniya 07.04.2018).
  17. Andrew G. Howard, Menglong Zhu, Bo Chen, Dmitry Kalenichenko, Weijun Wang, Tobias Weyand, Marco Andreetto, Hartwig Adam. MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications. Rezhim dostupa: https://arxiv.org/pdf/1704.04861.pdf (data obrashcheniya 07.04.2018).
  18. Joseph Redmon, Ali Farhadi. YOLO9000: Better, Faster, Stronger. Rezhim dostupa: https://arxiv.org/pdf/1612.08242.pdf (data obrashcheniya 07.04.2018).
  19. YOLO: Real-Time Object Detection. Rezhim dostupa: https://pjreddie.com/darknet/yolo/ (data obrashcheniya 07.04.2018).
Date of receipt: 23 апреля 2018 г.