Research on Optimization of Railway Obstacle Detection Model Based on Neural Architecture Search

The automatic perception of train operating environments leveraging neural networks has emerged as a new approach critical for ensuring the safe operation of trains. However, traditional neural network models primarily rely on a trial-and-error process conducted by human experts, along with accumula...

Full description

Saved in:
Bibliographic Details
Main Authors: YAO Weiwei, LYU Yu, ZHANG Huiyuan, XIONG Minjun, DONG Wenbo, LI Chen
Format: Article
Language:zho
Published: Editorial Office of Control and Information Technology 2024-08-01
Series:Kongzhi Yu Xinxi Jishu
Subjects:
Online Access:http://ctet.csrzic.com/thesisDetails#10.13889/j.issn.2096-5427.2024.04.012
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849224694758113280
author YAO Weiwei
LYU Yu
ZHANG Huiyuan
XIONG Minjun
DONG Wenbo
LI Chen
author_facet YAO Weiwei
LYU Yu
ZHANG Huiyuan
XIONG Minjun
DONG Wenbo
LI Chen
author_sort YAO Weiwei
collection DOAJ
description The automatic perception of train operating environments leveraging neural networks has emerged as a new approach critical for ensuring the safe operation of trains. However, traditional neural network models primarily rely on a trial-and-error process conducted by human experts, along with accumulated experience, which is not only time-consuming and tedious, but also hardly guarantees optimum performance for models. To address this issue, this paper proposes a method for optimizing railway obstacle detection models based on zero-cost neural architecture search. This method begins with the construction of an comprehensive space of potential model architectures. On this basis, the search scope is effectively constrained according to computational workloads required in real-world applications, ensuring that the selected models meet requirements both in accuracy and operational efficiency. The subsequent utilization of a zero-cost neural architecture search algorithm allows for quickly predicting the practical performance of various architectures, without the need for tedious and time-consuming actual training. Furthermore, a comparison of expected performance scores across different models leads to the selection of the optimal option as the ultimate solution. Experimental results demonstrated that this method achieved an average accuracy of 0.711 and an average inference time of 6.12 ms per frame, significantly outperforming baseline models with equivalent computational loads.
format Article
id doaj-art-ff145d980a7a4375886d8d7fb2077b19
institution Kabale University
issn 2096-5427
language zho
publishDate 2024-08-01
publisher Editorial Office of Control and Information Technology
record_format Article
series Kongzhi Yu Xinxi Jishu
spelling doaj-art-ff145d980a7a4375886d8d7fb2077b192025-08-25T06:57:11ZzhoEditorial Office of Control and Information TechnologyKongzhi Yu Xinxi Jishu2096-54272024-08-01909568496270Research on Optimization of Railway Obstacle Detection Model Based on Neural Architecture SearchYAO WeiweiLYU YuZHANG HuiyuanXIONG MinjunDONG WenboLI ChenThe automatic perception of train operating environments leveraging neural networks has emerged as a new approach critical for ensuring the safe operation of trains. However, traditional neural network models primarily rely on a trial-and-error process conducted by human experts, along with accumulated experience, which is not only time-consuming and tedious, but also hardly guarantees optimum performance for models. To address this issue, this paper proposes a method for optimizing railway obstacle detection models based on zero-cost neural architecture search. This method begins with the construction of an comprehensive space of potential model architectures. On this basis, the search scope is effectively constrained according to computational workloads required in real-world applications, ensuring that the selected models meet requirements both in accuracy and operational efficiency. The subsequent utilization of a zero-cost neural architecture search algorithm allows for quickly predicting the practical performance of various architectures, without the need for tedious and time-consuming actual training. Furthermore, a comparison of expected performance scores across different models leads to the selection of the optimal option as the ultimate solution. Experimental results demonstrated that this method achieved an average accuracy of 0.711 and an average inference time of 6.12 ms per frame, significantly outperforming baseline models with equivalent computational loads.http://ctet.csrzic.com/thesisDetails#10.13889/j.issn.2096-5427.2024.04.012railway obstacledeep learningautomatic perceptionobject detectionneural architecture search
spellingShingle YAO Weiwei
LYU Yu
ZHANG Huiyuan
XIONG Minjun
DONG Wenbo
LI Chen
Research on Optimization of Railway Obstacle Detection Model Based on Neural Architecture Search
Kongzhi Yu Xinxi Jishu
railway obstacle
deep learning
automatic perception
object detection
neural architecture search
title Research on Optimization of Railway Obstacle Detection Model Based on Neural Architecture Search
title_full Research on Optimization of Railway Obstacle Detection Model Based on Neural Architecture Search
title_fullStr Research on Optimization of Railway Obstacle Detection Model Based on Neural Architecture Search
title_full_unstemmed Research on Optimization of Railway Obstacle Detection Model Based on Neural Architecture Search
title_short Research on Optimization of Railway Obstacle Detection Model Based on Neural Architecture Search
title_sort research on optimization of railway obstacle detection model based on neural architecture search
topic railway obstacle
deep learning
automatic perception
object detection
neural architecture search
url http://ctet.csrzic.com/thesisDetails#10.13889/j.issn.2096-5427.2024.04.012
work_keys_str_mv AT yaoweiwei researchonoptimizationofrailwayobstacledetectionmodelbasedonneuralarchitecturesearch
AT lyuyu researchonoptimizationofrailwayobstacledetectionmodelbasedonneuralarchitecturesearch
AT zhanghuiyuan researchonoptimizationofrailwayobstacledetectionmodelbasedonneuralarchitecturesearch
AT xiongminjun researchonoptimizationofrailwayobstacledetectionmodelbasedonneuralarchitecturesearch
AT dongwenbo researchonoptimizationofrailwayobstacledetectionmodelbasedonneuralarchitecturesearch
AT lichen researchonoptimizationofrailwayobstacledetectionmodelbasedonneuralarchitecturesearch