|

Analysis of objects detection algorithms for the computer vision system of the fire extinguishing robot

Authors: Popov V.V.
Published in issue: #2(19)/2018
DOI: 10.18698/2541-8009-2018-2-252


Category: Mechanical Engineering and Machine Science | Chapter: Robots, Mechatronics, and Robotic Systems

Keywords: computer vision, image recognition, image processing, loss function, linear classification algorithm, searching the nearest neighbor algorithm, neural networks
Published: 30.01.2018

The article analyses current intelligent image recognition algorithms with the view of selecting the most appropriate one for subsequent implementation in the computer vision system of the fire extinguishing robot. We use the methods of searching the nearest neighbor, linear classification and neural networks. The main principles of each method as well as their benefits and drawbacks are considered. We choose a method which is the most advanced one according to the combination of algorithms performance indices and, consequently, the most suitable one for being applied in the computer vision systems.


References

[1] Vazaev A.V., Noskov V.P., Rubtsov I.V., Tsarichenko S.G. Combined computer vision system in firefighting robot control system. Izvestiya YuFU. Tekhnicheskie nauki [Izvestiya SFedU. Engineering Sciences], 2017, no. 1(126), pp. 121–132.

[2] Zelentsov I.A. Raspoznavanie obrazov [Pattern recognition]. Moscow, Bauman Press, 2008, 17 p.

[3] Forsyth D.A., Ponce J. Computer vision: a modern approach. Pearson, 2011, 792 p.

[4] Simon J.D. Prince. Computer vision: models, learning and inference. Cambridge University Press, 2012, 598 p.

[5] Szeliski R. Computer vision: algorithms and applications. Springer, 2010, 812 p.

[6] Callan R. The essence of neural networks. Prentice Hall, 1998, 248 p. (Russ. ed.: Osnovnye kontseptsii neyronnykh setey. Moscow, Vil’yams publ., 2001, 287 p.)

[7] Menshawy A., Karim Md.R., Zaccone G. Deep learning with TensorFlow: explore neural networks with Python. Packt Publishing, 2017, 320 p.

[8] Zhukov L.A., Reshetnikova N.V. Uchebnoe posobie po distsipline «Prilozheniya neyronnykh setey» [Study guide on “Neural networks application” course]. Krasnoyarsk, Politekhnicheskiy institut SFU publ., 2007, 154 p.

[9] Haykin S.S. Neural networks: a comprehensive foundation. Macmillan, 1994, 696 p. (Russ. ed.: Neyronnye seti. Polnyy kurs. Moscow, Vil’yams publ., 2006, 1104 p.)

[10] Goodfellow I., Bengio Y., Courville A. Deep learning. The MIT Press, 2016, 800 p.