Chapter A Comparative Study of Deep Learning Models for Symbol Detection in Technical Drawings

Symbols are a universal way to convey complex information in technical drawings since they can represent a wide range of elements, including components, materials, or relationships, in a concise and space-saving manner. Therefore, to enable a digital and automatic interpretation of pixel-based drawi...

ver descrição completa

Na minha lista:
Detalhes bibliográficos
Main Authors: Gann, Damaris, Faltin, Benedikt, König, Markus
Formato: Online
Idioma:inglês
Publicado em: Firenze University Press 2024
Assuntos:
Acesso em linha:ONIX_20240402_9791221502893_14
Tags: Adicionar Tag
Sem tags, seja o primeiro a adicionar uma tag!
_version_ 1869516754694176768
author Gann, Damaris
Faltin, Benedikt
König, Markus
author_browse Faltin, Benedikt
Gann, Damaris
König, Markus
author_facet Gann, Damaris
Faltin, Benedikt
König, Markus
author_sort Gann, Damaris
collection Directory of Open Access Books
description Symbols are a universal way to convey complex information in technical drawings since they can represent a wide range of elements, including components, materials, or relationships, in a concise and space-saving manner. Therefore, to enable a digital and automatic interpretation of pixel-based drawings, accurate detection of symbols is a crucial step. To enhance the efficiency of the digitization process, current research focuses on automating this symbol detection using deep learning models. However, the ever-increasing repertoire of model architectures poses a challenge for researchers and practitioners alike in retaining an overview of the latest advancements and selecting the most suitable model architecture for their respective use cases. To provide guidance, this contribution conducts a comparative study of prevalent and state-of-the-art model architectures for the task of symbol detection in pixel-based construction drawings. Therefore, this study evaluates six different object detection model architectures, including YOLOv5, YOLOv7, YOLOv8, Swin-Transformer, ConvNeXt, and Faster-RCNN. These models are trained and tested on two distinct datasets from the bridge and residential building domains, both representing substantial sub-sectors of the construction industry. Furthermore, the models are evaluated based on five criteria, i.e., detection accuracy, robustness to data scarcity, training time, inference time, and model size. In summary, our comparative study highlights the performance and capabilities of different deep learning models for symbol detection in construction drawings. Through the comprehensive evaluation and practical insights, this research facilitates the advancement of automated symbol detection by showing the strengths and weaknesses of the model architectures, thus providing users with valuable guidance in choosing the most appropriate model for their real-world applications
format Online
id doab-20.500.12854ir-137218
institution Directory of Open Access Books
language eng
publishDate 2024
publishDateRange 2024
publishDateSort 2024
publisher Firenze University Press
publisherStr Firenze University Press
record_format ojs
spelling doab-20.500.12854ir-1372182024-05-13T00:17:12Z Chapter A Comparative Study of Deep Learning Models for Symbol Detection in Technical Drawings Gann, Damaris Faltin, Benedikt König, Markus Computer Vision Technical Drawings Symbol Detection Comparative Study thema EDItEUR::U Computing and Information Technology::UT Computer networking and communications::UTV Virtualization Symbols are a universal way to convey complex information in technical drawings since they can represent a wide range of elements, including components, materials, or relationships, in a concise and space-saving manner. Therefore, to enable a digital and automatic interpretation of pixel-based drawings, accurate detection of symbols is a crucial step. To enhance the efficiency of the digitization process, current research focuses on automating this symbol detection using deep learning models. However, the ever-increasing repertoire of model architectures poses a challenge for researchers and practitioners alike in retaining an overview of the latest advancements and selecting the most suitable model architecture for their respective use cases. To provide guidance, this contribution conducts a comparative study of prevalent and state-of-the-art model architectures for the task of symbol detection in pixel-based construction drawings. Therefore, this study evaluates six different object detection model architectures, including YOLOv5, YOLOv7, YOLOv8, Swin-Transformer, ConvNeXt, and Faster-RCNN. These models are trained and tested on two distinct datasets from the bridge and residential building domains, both representing substantial sub-sectors of the construction industry. Furthermore, the models are evaluated based on five criteria, i.e., detection accuracy, robustness to data scarcity, training time, inference time, and model size. In summary, our comparative study highlights the performance and capabilities of different deep learning models for symbol detection in construction drawings. Through the comprehensive evaluation and practical insights, this research facilitates the advancement of automated symbol detection by showing the strengths and weaknesses of the model architectures, thus providing users with valuable guidance in choosing the most appropriate model for their real-world applications 2024-05-13T00:17:09Z 2024-05-13T00:17:09Z 2024-04-02T15:44:42Z 2023 chapter ONIX_20240402_9791221502893_14 2704-5846 https://library.oapen.org/handle/20.500.12657/89045 9791221502893 https://directory.doabooks.org/handle/20.500.12854/137218 eng Proceedings e report open access image/jpeg n/a https://library.oapen.org/bitstream/20.500.12657/89045/1/9791221502893_87.pdf Firenze University Press 10.36253/979-12-215-0289-3.87 10.36253/979-12-215-0289-3.87 2ec4474d-93b1-4cfa-b313-9c6019b51b1a 9791221502893 10 Florence open access
spellingShingle Computer Vision
Technical Drawings
Symbol Detection
Comparative Study
thema EDItEUR::U Computing and Information Technology::UT Computer networking and communications::UTV Virtualization
Gann, Damaris
Faltin, Benedikt
König, Markus
Chapter A Comparative Study of Deep Learning Models for Symbol Detection in Technical Drawings
title Chapter A Comparative Study of Deep Learning Models for Symbol Detection in Technical Drawings
title_full Chapter A Comparative Study of Deep Learning Models for Symbol Detection in Technical Drawings
title_fullStr Chapter A Comparative Study of Deep Learning Models for Symbol Detection in Technical Drawings
title_full_unstemmed Chapter A Comparative Study of Deep Learning Models for Symbol Detection in Technical Drawings
title_short Chapter A Comparative Study of Deep Learning Models for Symbol Detection in Technical Drawings
title_sort chapter a comparative study of deep learning models for symbol detection in technical drawings
topic Computer Vision
Technical Drawings
Symbol Detection
Comparative Study
thema EDItEUR::U Computing and Information Technology::UT Computer networking and communications::UTV Virtualization
topic_facet Computer Vision
Technical Drawings
Symbol Detection
Comparative Study
thema EDItEUR::U Computing and Information Technology::UT Computer networking and communications::UTV Virtualization
url ONIX_20240402_9791221502893_14
work_keys_str_mv AT ganndamaris chapteracomparativestudyofdeeplearningmodelsforsymboldetectionintechnicaldrawings
AT faltinbenedikt chapteracomparativestudyofdeeplearningmodelsforsymboldetectionintechnicaldrawings
AT konigmarkus chapteracomparativestudyofdeeplearningmodelsforsymboldetectionintechnicaldrawings