Melola Discovery

← Melola Discovery

Jurnal SINTA 1

A Comparison of CNN-based Image Feature Extractors for Weld Defects Classification

INDONESIAN JOURNAL OF APPLIED PHYSICS · Vol. 14 No. 1 (2024) · hlm. 190–201

DOI
10.13057/ijap.v14i1.72509
ISSN
2477-6416, 2089-0133
Terbit
2024-05-02
Penerbit
Department of Physics, Sebelas Maret University
Bahasa
eng

Abstrak

Classification of the types of weld defects is one of the stages of evaluating radiographic images, which is an essential step in controlling the quality of welded joints in materials. By automating the weld defects classification based on deep learning and the CNN architecture, it is possible to overcome the limitations of visually or manually evaluating radiographic images. Good accuracy in classification models for weld defects requires the availability of sufficient datasets. In reality, however, the radiographic image dataset accessible to the public is limited and imbalanced between classes. Consequently, simple image cropping and augmentation techniques are implemented during the data preparation stage. To construct a weld defect classification model, we proposed to utilize the transfer learning method by employing a pre-trained CNN architecture as a feature extractor, including DenseNet201, InceptionV3, MobileNetV2, NASNetMobile, ResNet50V2, VGG16, VGG19, and Xception, which are linked to a simple classification model based on multilayer perceptron. The test results indicate that the three best classification models were obtained by using the DenseNet201 feature extractor with a test accuracy value of 100%, followed by ResNet50V2 and InceptionV3 with an accuracy of 99.17%. These outcomes are better compared to state-of-the-art classification models with a maximum of six classes of defects. The research findings may assist radiography experts in evaluating radiographic images more accurately and efficiently.

Kata kunci

Jaringan kolaborasi para penulis

10 penulis · 17 tautan kepenulisan bersama · 2 lintas organisasi

Arahkan kursor atau ketuk sebuah nama untuk menyorot rekan penulisnya. Klik dua kali untuk membuka halaman penulis.

  • University of Indonesia
  • National Research and Innovation Agency
  • Afiliasi tidak diketahui
  • Organisasi berbeda
  • Organisasi sama
  • Afiliasi tidak diketahui
  • Garis lebih tebal = lebih banyak artikel bersama · lingkaran lebih besar = lebih banyak artikel
Tampilkan sebagai tabel
Rekan penulisOrganisasiArtikel bersama
Finkan DanitasariUniversity of Indonesia2
Fitria Yunita DewiUniversity of Indonesia1
Harry Kusuma AliwargaAfiliasi tidak diketahui1
Faris Zulkarnain S. Hi. RaufUniversity of Indonesia1
Ilham S PradanaUniversity of Indonesia1
Dimas AliftaAfiliasi tidak diketahui1

Afiliasi berasal dari OpenAlex; metadata OAI-PMH jurnal tidak memuatnya. Penulis tanpa ID OpenAlex dicocokkan berdasarkan nama, sehingga dua orang bisa tergabung atau satu orang bisa terpisah.

Sumber metadata: OAI-PMH (metadata jurnal) https://jurnal.uns.ac.id/ijap/oai, pertama di-harvest 8 Nov 2026, terakhir terlihat 8 Nov 2026.
Metadata asli (oai_dc)
<oai_dc:dc xmlns="http://www.openarchives.org/OAI/2.0/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:oai_dc="http://www.openarchives.org/OAI/2.0/oai_dc/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/oai_dc/  http://www.openarchives.org/OAI/2.0/oai_dc.xsd">
	<dc:title xml:lang="en-US">A Comparison of CNN-based Image Feature Extractors for Weld Defects Classification</dc:title>
	<dc:creator>Purnomo, Tito Wahyu</dc:creator>
	<dc:creator>Ramadhany, Harun Al Rasyid</dc:creator>
	<dc:creator>Jati, Hapsara Hadi Carita</dc:creator>
	<dc:creator>Handoko, Djati</dc:creator>
	<dc:subject xml:lang="en-US">CNN; Transfer Learning; Feature Extraction; Radiographic Image; Weld Defects</dc:subject>
	<dc:description xml:lang="en-US">Classification of the types of weld defects is one of the stages of evaluating radiographic images, which is an essential step in controlling the quality of welded joints in materials. By automating the weld defects classification based on deep learning and the CNN architecture, it is possible to overcome the limitations of visually or manually evaluating radiographic images. Good accuracy in classification models for weld defects requires the availability of sufficient datasets. In reality, however, the radiographic image dataset accessible to the public is limited and imbalanced between classes. Consequently, simple image cropping and augmentation techniques are implemented during the data preparation stage. To construct a weld defect classification model, we proposed to utilize the transfer learning method by employing a pre-trained CNN architecture as a feature extractor, including DenseNet201, InceptionV3, MobileNetV2, NASNetMobile, ResNet50V2, VGG16, VGG19, and Xception, which are linked to a simple classification model based on multilayer perceptron. The test results indicate that the three best classification models were obtained by using the DenseNet201 feature extractor with a test accuracy value of 100%, followed by ResNet50V2 and InceptionV3 with an accuracy of 99.17%. These outcomes are better compared to state-of-the-art classification models with a maximum of six classes of defects. The research findings may assist radiography experts in evaluating radiographic images more accurately and efficiently.</dc:description>
	<dc:publisher xml:lang="en-US">Department of Physics, Sebelas Maret University</dc:publisher>
	<dc:contributor xml:lang="en-US"></dc:contributor>
	<dc:date>2024-05-02</dc:date>
	<dc:type>info:eu-repo/semantics/article</dc:type>
	<dc:type>info:eu-repo/semantics/publishedVersion</dc:type>
	<dc:type xml:lang="en-US">Peer-reviewed Article</dc:type>
	<dc:format>application/pdf</dc:format>
	<dc:identifier>https://jurnal.uns.ac.id/ijap/article/view/72509</dc:identifier>
	<dc:identifier>10.13057/ijap.v14i1.72509</dc:identifier>
	<dc:source xml:lang="en-US">INDONESIAN JOURNAL OF APPLIED PHYSICS; Vol 14, No 1 (2024): April; 190-201</dc:source>
	<dc:source>2477-6416</dc:source>
	<dc:source>2089-0133</dc:source>
	<dc:language>eng</dc:language>
	<dc:relation>https://jurnal.uns.ac.id/ijap/article/view/72509/45241</dc:relation>
	<dc:rights xml:lang="en-US">Copyright (c) 2024 INDONESIAN JOURNAL OF APPLIED PHYSICS</dc:rights>
</oai_dc:dc>