eprintid: 17626
rev_number: 2
eprint_status: archive
userid: 1
dir: disk0/00/01/76/26
datestamp: 2023-12-19 03:23:59
lastmod: 2023-12-19 03:23:59
status_changed: 2023-12-19 03:08:23
type: article
metadata_visibility: show
creators_name: Bin, O.K.
creators_name: Hooi, Y.K.
creators_name: Kadir, S.J.A.
creators_name: Fujita, H.
creators_name: Rosli, L.H.
title: Enhanced Symbol Recognition based on Advanced Data Augmentation for Engineering Diagrams
ispublished: pub
keywords: Pattern recognition, Convolution neural network; Cyclegan; Data augmentation; Engineering diagrams; Engineering drawing; Learning dataset; Piping and instrument diagram; Symbol recognition; Symbol spotting, Convolution
note: cited By 0
abstract: Symbol recognition has generated research interest for image analytics of engineering diagrams. Techniques including structural, syntactic, statistical, Convolution Neural Network (CNN) were studied to identify gaps of research. Despite popularity, CNN requires huge learning dataset, which often involves costly procurement. To address this, combination between CycleGAN and CNN is proposed. CycleGAN generates more learning dataset synthetically, thus yielding opportunity to improve accuracy of symbol recognition. In the domain of for engineering symbols, standard CNN model is developed and used in experimental testing. Different ratios of training dataset were tested in multiple experiments using Piping and Instrument Diagram (P&IDs) drawings. Result of highest accuracy for symbol recognition is up to 92.85 against baseline and other method. The results determined that gradual reduction of training samples, the effectiveness of recognition accuracy performance after using proposed method was remained substantially stable. © 2022. International Journal of Advanced Computer Science and Applications. All Rights Reserved.
date: 2022
publisher: Science and Information Organization
official_url: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85131400870&doi=10.14569%2fIJACSA.2022.0130563&partnerID=40&md5=e6e273ecba3174ee42957a0f0f44cd4e
id_number: 10.14569/IJACSA.2022.0130563
full_text_status: none
publication: International Journal of Advanced Computer Science and Applications
volume: 13
number: 5
pagerange: 537-546
refereed: TRUE
issn: 2158107X
citation:   Bin, O.K. and Hooi, Y.K. and Kadir, S.J.A. and Fujita, H. and Rosli, L.H.  (2022) Enhanced Symbol Recognition based on Advanced Data Augmentation for Engineering Diagrams.  International Journal of Advanced Computer Science and Applications, 13 (5).  pp. 537-546.  ISSN 2158107X