Melola Discovery

← Melola Discovery

Journal No SINTA rank

A Comparative Study of Deep Learning Architectures for Content-Based Image Retrieval on a Stone Texture Dataset

Scientific Journal of Informatics · Vol. 13 No. 3 (2026) · pp. 673–684

DOI
10.15294/sji.v13i3.51297
ISSN
2407-7658, 2460-0040
Published
2026-08-20
Publisher
Universitas Negeri Semarang
Language
eng

Abstract

Purpose: This study addresses a persistent gap in geological content-based image retrieval (CBIR): the reliance of existing sequence-based and hybrid deep learning models on static, two-dimensional rock texture representations that they were not originally designed to handle. Methods: Four architectures were evaluated on a unified 9,853-image, ten-category Stone-2D dataset, partitioned into training and validation subsets at an 8:2 ratio (7,882/1,971 images), with FAISS employed for similarity search over 128-dimensional embeddings. Findings: On an identical held-out test set, all four architectures achieved comparable performance (82.4%–85.2% accuracy; macro F1 0.826–0.860), with a standard 2D CNN achieving the highest accuracy (85.2%) and the custom pseudo-3D MineralNet architecture achieving the highest macro F1 (0.860). No architecture dominated across both metrics. Novelty: These results indicate that, once preprocessing is aligned with the native dimensionality of the image data, architectural complexity provides limited additional benefit for static geological texture classification — challenging the common assumption that specialized or sequential architectures are necessary for this task. A working CBIR retrieval pipeline was validated with representative query examples, demonstrating semantically consistent nearest-neighbour retrieval across distinct rock textures.

Keywords

Collaboration network of the authors

6 authors · 11 co-authorship links

Hover or tap a name to highlight its co-authors. Double-click to open the author's page.

  • Informatics and Business Institute Darmajaya
  • Different organisations
  • Same organisation
  • Affiliation unknown
  • Thicker line = more joint articles · bigger circle = more articles
Show as table
Co-authorOrganisationJoint articles
Rio KurniawanInformatics and Business Institute Darmajaya1
Dona YuliawatiInformatics and Business Institute Darmajaya1
Bayu NugrohoInformatics and Business Institute Darmajaya1
Sri KarnilaInformatics and Business Institute Darmajaya1

Affiliations come from OpenAlex; journal OAI-PMH metadata carries none. Authors without an OpenAlex ID are matched by name, which can merge or split people.

Metadata source: OAI-PMH (metadata jurnal) https://journal.unnes.ac.id/journals/sji/oai, first harvested 6 Nov 2026, last seen 8 Nov 2026.
Original metadata (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">A Comparative Study of Deep Learning Architectures for Content-Based Image Retrieval on a Stone Texture Dataset</dc:title>
	<dc:creator xml:lang="en">Taufiqurrohman, Maulana</dc:creator>
	<dc:creator xml:lang="en">Suhendro Irianto</dc:creator>
	<dc:subject xml:lang="en">CBIR</dc:subject>
	<dc:subject xml:lang="en">Deep learning</dc:subject>
	<dc:subject xml:lang="en">Convolutional neural network</dc:subject>
	<dc:subject xml:lang="en">Image retrieval</dc:subject>
	<dc:subject xml:lang="en">Stone texture</dc:subject>
	<dc:description xml:lang="en">Purpose: This study addresses a persistent gap in geological content-based image retrieval (CBIR): the reliance of existing sequence-based and hybrid deep learning models on static, two-dimensional rock texture representations that they were not originally designed to handle.
Methods: Four architectures were evaluated on a unified 9,853-image, ten-category Stone-2D dataset, partitioned into training and validation subsets at an 8:2 ratio (7,882/1,971 images), with FAISS employed for similarity search over 128-dimensional embeddings.
Findings: On an identical held-out test set, all four architectures achieved comparable performance (82.4%–85.2% accuracy; macro F1 0.826–0.860), with a standard 2D CNN achieving the highest accuracy (85.2%) and the custom pseudo-3D MineralNet architecture achieving the highest macro F1 (0.860). No architecture dominated across both metrics.
Novelty: These results indicate that, once preprocessing is aligned with the native dimensionality of the image data, architectural complexity provides limited additional benefit for static geological texture classification — challenging the common assumption that specialized or sequential architectures are necessary for this task. A working CBIR retrieval pipeline was validated with representative query examples, demonstrating semantically consistent nearest-neighbour retrieval across distinct rock textures.</dc:description>
	<dc:publisher xml:lang="en">Universitas Negeri Semarang</dc:publisher>
	<dc:date>2026-08-20</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">Peer-reviewed Article</dc:type>
	<dc:format>application/pdf</dc:format>
	<dc:identifier>https://journal.unnes.ac.id/journals/sji/article/view/51297</dc:identifier>
	<dc:identifier>10.15294/sji.v13i3.51297</dc:identifier>
	<dc:source xml:lang="en">Scientific Journal of Informatics; Vol. 13 No. 3: Special Issue; 673-684</dc:source>
	<dc:source>2407-7658</dc:source>
	<dc:source>2460-0040</dc:source>
	<dc:source>10.15294/sji.v13i3</dc:source>
	<dc:language>eng</dc:language>
	<dc:relation>https://journal.unnes.ac.id/journals/sji/article/view/51297/11166</dc:relation>
	<dc:rights xml:lang="en">Copyright (c) 2026 Scientific Journal of Informatics</dc:rights>
</oai_dc:dc>