Convolutional neural networks of the YOLO class in computer vision systems for mobile robotic complexes

Ivan V. Zoev, Alexey P. Beresnev, Nikolay G. Markov

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

An important scientific direction is the development and study of computer vision systems (CVS) for mobile robotic complexes. Today, developers of CVS are most often using convolutional neural networks (CNN). For increasing the speed detection of objects on images in CVS, there has been a trend of using CNN, which are hardware-implemented on field-programmable gate array (FPGAs).This article shows that the perspective for hardware implementation on the FPGA is the tiny-YOLO CNN from the YOLO class. For reduce required FPGA computing resources in this CNN, was proposed to use Inception-ResNet modules. We was found that with high detection accuracy of objects in images with minimum resources requirements provide by the tiny-YOLO-Inception-ResNet2 network architecture. It is obtained from replacing the fifth tiny-YOLO convolutional layer of the tiny-YOLO CNN with two sequential processing Inception-ResNet modules. Also results of the study of the detection accuracy of objects using the CNN for this architecture with the lack of resource-intensive operations: batch normalization and bias from calculations were given. These studies were performed for different formats of representation numbers in the FPGA.

Original languageEnglish
Title of host publication2019 International Siberian Conference on Control and Communications, SIBCON 2019 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781538651414
DOIs
Publication statusPublished - 1 Apr 2019
Event2019 International Siberian Conference on Control and Communications, SIBCON 2019 - Tomsk, Russian Federation
Duration: 18 Apr 201920 Apr 2019

Publication series

Name2019 International Siberian Conference on Control and Communications, SIBCON 2019 - Proceedings

Conference

Conference2019 International Siberian Conference on Control and Communications, SIBCON 2019
CountryRussian Federation
CityTomsk
Period18.4.1920.4.19

Fingerprint

Mobile Robotics
Vision System
Computer Vision
Computer vision
Robotics
Neural Networks
Neural networks
Field Programmable Gate Array
Field programmable gate arrays (FPGA)
Resources
Hardware
Module
Hardware Implementation
Network Architecture
Network architecture
Batch
Normalization
Class
Computing
Requirements

Keywords

  • Computer vision systems
  • Convolutional neural networks
  • Field-programmable gate array
  • Mobile robotic systems
  • Object detection on images

ASJC Scopus subject areas

  • Control and Optimization
  • Computer Networks and Communications
  • Signal Processing

Cite this

Zoev, I. V., Beresnev, A. P., & Markov, N. G. (2019). Convolutional neural networks of the YOLO class in computer vision systems for mobile robotic complexes. In 2019 International Siberian Conference on Control and Communications, SIBCON 2019 - Proceedings [8729605] (2019 International Siberian Conference on Control and Communications, SIBCON 2019 - Proceedings). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/SIBCON.2019.8729605

Convolutional neural networks of the YOLO class in computer vision systems for mobile robotic complexes. / Zoev, Ivan V.; Beresnev, Alexey P.; Markov, Nikolay G.

2019 International Siberian Conference on Control and Communications, SIBCON 2019 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2019. 8729605 (2019 International Siberian Conference on Control and Communications, SIBCON 2019 - Proceedings).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Zoev, IV, Beresnev, AP & Markov, NG 2019, Convolutional neural networks of the YOLO class in computer vision systems for mobile robotic complexes. in 2019 International Siberian Conference on Control and Communications, SIBCON 2019 - Proceedings., 8729605, 2019 International Siberian Conference on Control and Communications, SIBCON 2019 - Proceedings, Institute of Electrical and Electronics Engineers Inc., 2019 International Siberian Conference on Control and Communications, SIBCON 2019, Tomsk, Russian Federation, 18.4.19. https://doi.org/10.1109/SIBCON.2019.8729605
Zoev IV, Beresnev AP, Markov NG. Convolutional neural networks of the YOLO class in computer vision systems for mobile robotic complexes. In 2019 International Siberian Conference on Control and Communications, SIBCON 2019 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2019. 8729605. (2019 International Siberian Conference on Control and Communications, SIBCON 2019 - Proceedings). https://doi.org/10.1109/SIBCON.2019.8729605
Zoev, Ivan V. ; Beresnev, Alexey P. ; Markov, Nikolay G. / Convolutional neural networks of the YOLO class in computer vision systems for mobile robotic complexes. 2019 International Siberian Conference on Control and Communications, SIBCON 2019 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2019. (2019 International Siberian Conference on Control and Communications, SIBCON 2019 - Proceedings).
@inproceedings{263d93127f964488990dbdc2f80f10e4,
title = "Convolutional neural networks of the YOLO class in computer vision systems for mobile robotic complexes",
abstract = "An important scientific direction is the development and study of computer vision systems (CVS) for mobile robotic complexes. Today, developers of CVS are most often using convolutional neural networks (CNN). For increasing the speed detection of objects on images in CVS, there has been a trend of using CNN, which are hardware-implemented on field-programmable gate array (FPGAs).This article shows that the perspective for hardware implementation on the FPGA is the tiny-YOLO CNN from the YOLO class. For reduce required FPGA computing resources in this CNN, was proposed to use Inception-ResNet modules. We was found that with high detection accuracy of objects in images with minimum resources requirements provide by the tiny-YOLO-Inception-ResNet2 network architecture. It is obtained from replacing the fifth tiny-YOLO convolutional layer of the tiny-YOLO CNN with two sequential processing Inception-ResNet modules. Also results of the study of the detection accuracy of objects using the CNN for this architecture with the lack of resource-intensive operations: batch normalization and bias from calculations were given. These studies were performed for different formats of representation numbers in the FPGA.",
keywords = "Computer vision systems, Convolutional neural networks, Field-programmable gate array, Mobile robotic systems, Object detection on images",
author = "Zoev, {Ivan V.} and Beresnev, {Alexey P.} and Markov, {Nikolay G.}",
year = "2019",
month = "4",
day = "1",
doi = "10.1109/SIBCON.2019.8729605",
language = "English",
series = "2019 International Siberian Conference on Control and Communications, SIBCON 2019 - Proceedings",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
booktitle = "2019 International Siberian Conference on Control and Communications, SIBCON 2019 - Proceedings",
address = "United States",

}

TY - GEN

T1 - Convolutional neural networks of the YOLO class in computer vision systems for mobile robotic complexes

AU - Zoev, Ivan V.

AU - Beresnev, Alexey P.

AU - Markov, Nikolay G.

PY - 2019/4/1

Y1 - 2019/4/1

N2 - An important scientific direction is the development and study of computer vision systems (CVS) for mobile robotic complexes. Today, developers of CVS are most often using convolutional neural networks (CNN). For increasing the speed detection of objects on images in CVS, there has been a trend of using CNN, which are hardware-implemented on field-programmable gate array (FPGAs).This article shows that the perspective for hardware implementation on the FPGA is the tiny-YOLO CNN from the YOLO class. For reduce required FPGA computing resources in this CNN, was proposed to use Inception-ResNet modules. We was found that with high detection accuracy of objects in images with minimum resources requirements provide by the tiny-YOLO-Inception-ResNet2 network architecture. It is obtained from replacing the fifth tiny-YOLO convolutional layer of the tiny-YOLO CNN with two sequential processing Inception-ResNet modules. Also results of the study of the detection accuracy of objects using the CNN for this architecture with the lack of resource-intensive operations: batch normalization and bias from calculations were given. These studies were performed for different formats of representation numbers in the FPGA.

AB - An important scientific direction is the development and study of computer vision systems (CVS) for mobile robotic complexes. Today, developers of CVS are most often using convolutional neural networks (CNN). For increasing the speed detection of objects on images in CVS, there has been a trend of using CNN, which are hardware-implemented on field-programmable gate array (FPGAs).This article shows that the perspective for hardware implementation on the FPGA is the tiny-YOLO CNN from the YOLO class. For reduce required FPGA computing resources in this CNN, was proposed to use Inception-ResNet modules. We was found that with high detection accuracy of objects in images with minimum resources requirements provide by the tiny-YOLO-Inception-ResNet2 network architecture. It is obtained from replacing the fifth tiny-YOLO convolutional layer of the tiny-YOLO CNN with two sequential processing Inception-ResNet modules. Also results of the study of the detection accuracy of objects using the CNN for this architecture with the lack of resource-intensive operations: batch normalization and bias from calculations were given. These studies were performed for different formats of representation numbers in the FPGA.

KW - Computer vision systems

KW - Convolutional neural networks

KW - Field-programmable gate array

KW - Mobile robotic systems

KW - Object detection on images

UR - http://www.scopus.com/inward/record.url?scp=85068347615&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85068347615&partnerID=8YFLogxK

U2 - 10.1109/SIBCON.2019.8729605

DO - 10.1109/SIBCON.2019.8729605

M3 - Conference contribution

AN - SCOPUS:85068347615

T3 - 2019 International Siberian Conference on Control and Communications, SIBCON 2019 - Proceedings

BT - 2019 International Siberian Conference on Control and Communications, SIBCON 2019 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

ER -