Design and Optimization of a Deep Neural Network Architecture for Traffic Light Detection

Design and Optimization of a Deep Neural Network Architecture for Traffic Light Detection

Autonomous Driving has recently become a research trend and efficient autonomous driving system is difficult to achieve due to safety concerns, Applying traffic light recognition to autonomous driving system is one of the factors to prevent accidents that occur as a result of traffic light violation...

Full description

Bibliographic Details
Main Authors: Fukuchi Tomohide, Ikechukwu Mark Ogbodo, Ben Abdallah Abderazek
Format: Article
Language:English
Published: EDP Sciences 2020-01-01
Series:SHS Web of Conferences
Online Access:https://www.shs-conferences.org/articles/shsconf/pdf/2020/05/shsconf_etltc2020_01002.pdf
id doaj-9055474fce5e497cb1130eaad8516f40
record_format Article
spelling doaj-9055474fce5e497cb1130eaad8516f402021-04-02T14:57:06ZengEDP SciencesSHS Web of Conferences2261-24242020-01-01770100210.1051/shsconf/20207701002shsconf_etltc2020_01002Design and Optimization of a Deep Neural Network Architecture for Traffic Light DetectionFukuchi TomohideIkechukwu Mark OgbodoBen Abdallah AbderazekAutonomous Driving has recently become a research trend and efficient autonomous driving system is difficult to achieve due to safety concerns, Applying traffic light recognition to autonomous driving system is one of the factors to prevent accidents that occur as a result of traffic light violation. To realize safe autonomous driving system, we propose in this work a design and optimization of a traffic light detection system based on deep neural network. We designed a lightweight convolution neural network with parameters less than 10000 and implemented in software. We achieved 98.3% inference accuracy with 2.5 fps response time. Also we optimized the input image pixel values with normalization and optimized convolution layer with pipeline on FPGA with 5% resource consumption.https://www.shs-conferences.org/articles/shsconf/pdf/2020/05/shsconf_etltc2020_01002.pdf
collection DOAJ
language English
format Article
sources DOAJ
author Fukuchi Tomohide
Ikechukwu Mark Ogbodo
Ben Abdallah Abderazek
spellingShingle Fukuchi Tomohide
Ikechukwu Mark Ogbodo
Ben Abdallah Abderazek
Design and Optimization of a Deep Neural Network Architecture for Traffic Light Detection
SHS Web of Conferences
author_facet Fukuchi Tomohide
Ikechukwu Mark Ogbodo
Ben Abdallah Abderazek
author_sort Fukuchi Tomohide
title Design and Optimization of a Deep Neural Network Architecture for Traffic Light Detection
title_short Design and Optimization of a Deep Neural Network Architecture for Traffic Light Detection
title_full Design and Optimization of a Deep Neural Network Architecture for Traffic Light Detection
title_fullStr Design and Optimization of a Deep Neural Network Architecture for Traffic Light Detection
title_full_unstemmed Design and Optimization of a Deep Neural Network Architecture for Traffic Light Detection
title_sort design and optimization of a deep neural network architecture for traffic light detection
publisher EDP Sciences
series SHS Web of Conferences
issn 2261-2424
publishDate 2020-01-01
description Autonomous Driving has recently become a research trend and efficient autonomous driving system is difficult to achieve due to safety concerns, Applying traffic light recognition to autonomous driving system is one of the factors to prevent accidents that occur as a result of traffic light violation. To realize safe autonomous driving system, we propose in this work a design and optimization of a traffic light detection system based on deep neural network. We designed a lightweight convolution neural network with parameters less than 10000 and implemented in software. We achieved 98.3% inference accuracy with 2.5 fps response time. Also we optimized the input image pixel values with normalization and optimized convolution layer with pipeline on FPGA with 5% resource consumption.
url https://www.shs-conferences.org/articles/shsconf/pdf/2020/05/shsconf_etltc2020_01002.pdf
work_keys_str_mv AT fukuchitomohide designandoptimizationofadeepneuralnetworkarchitecturefortrafficlightdetection
AT ikechukwumarkogbodo designandoptimizationofadeepneuralnetworkarchitecturefortrafficlightdetection
AT benabdallahabderazek designandoptimizationofadeepneuralnetworkarchitecturefortrafficlightdetection
_version_ 1721560947091308544

Similar Items