Melola Discovery

← Melola Discovery

Akses terbuka Jurnal SINTA 1

Surface crack edge-aware segmentation using hierarchical transformer encoder and Sobel edges for aerial inspection

Sinergi · Vol. 30 No. 3 (2026) · hlm. 981–994

DOI
10.22441/sinergi.2026.3.022
ISSN
2460-1217, 1410-2331
Terbit
2026-09-29
Penerbit
Universitas Mercu Buana
Bahasa
eng
Lisensi
creativecommons.org/licenses/by-sa/4.0

Abstrak

Structural crack detection remains a key component of structural health monitoring for large-scale civil infrastructure, where manual inspection is costly and hazardous. While aerial imaging enables rapid data acquisition, reliable crack analysis is limited by the difficulty of obtaining high-resolution images at close range without risking UAV damage. To address this, we introduce a UAV equipped with a protective spherical shell that allows safe surface contact. Although CNN-based architectures such as FPN and U-Net are established for crack segmentation, their encoders struggle to model long-range dependencies and fine edge structures, which result in poor delineation of thin, low-contrast, and discontinuous cracks. This limitation remains a persistent research gap in aerial crack analysis. To address this gap, this study investigates two complementary strategies: (i) replacing conventional CNN encoders with hierarchical Transformer backbones to capture global contextual relationships, and (ii) introducing edge-aware input augmentation using luminosity-based grayscale and Sobel gradient maps to optimize boundary sensitivity. We conducted experiments on our CrackUAS dataset, which consists of aerial images collected from infrastructure inspection scenarios. Integrating a hierarchical Transformer encoder with Sobel edge augmentation yields consistent performance gains across architectures. The ablation study concludes that F1-scores increase by approximately 33% for U-Net and 13% for FPN relative to their baseline configurations. These results show that integrating Transformer-based feature encoding with edge-aware input augmentation improves aerial crack segmentation.

Kata kunci

Jaringan kolaborasi para penulis

5 penulis · 10 tautan kepenulisan bersama

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

  • Mindanao State University – Iligan Institute of Technology
  • 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

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://publikasi.mercubuana.ac.id/index.php/sinergi/oai, pertama di-harvest 10 Nov 2026, terakhir terlihat 16 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">Surface crack edge-aware segmentation using hierarchical transformer encoder and Sobel edges for aerial inspection</dc:title>
	<dc:creator>Aleluya, Earl Ryan</dc:creator>
	<dc:creator>Ocaña, Kathleen Rose</dc:creator>
	<dc:creator>Manulat, Preus Prixor</dc:creator>
	<dc:creator>Alagon, Francis Jann</dc:creator>
	<dc:creator>Salaan, Carl John</dc:creator>
	<dc:subject xml:lang="en-US">Crack Detection</dc:subject>
	<dc:subject xml:lang="en-US">Deep Learning</dc:subject>
	<dc:subject xml:lang="en-US">Semantic Segmentation</dc:subject>
	<dc:subject xml:lang="en-US">Structural Health Monitoring</dc:subject>
	<dc:subject xml:lang="en-US">Unmanned Aerial Vehicles</dc:subject>
	<dc:description xml:lang="en-US">Structural crack detection remains a key component of structural health monitoring for large-scale civil infrastructure, where manual inspection is costly and hazardous. While aerial imaging enables rapid data acquisition, reliable crack analysis is limited by the difficulty of obtaining high-resolution images at close range without risking UAV damage. To address this, we introduce a UAV equipped with a protective spherical shell that allows safe surface contact. Although CNN-based architectures such as FPN and U-Net are established for crack segmentation, their encoders struggle to model long-range dependencies and fine edge structures, which result in poor delineation of thin, low-contrast, and discontinuous cracks. This limitation remains a persistent research gap in aerial crack analysis. To address this gap, this study investigates two complementary strategies: (i) replacing conventional CNN encoders with hierarchical Transformer backbones to capture global contextual relationships, and (ii) introducing edge-aware input augmentation using luminosity-based grayscale and Sobel gradient maps to optimize boundary sensitivity. We conducted experiments on our CrackUAS dataset, which consists of aerial images collected from infrastructure inspection scenarios. Integrating a hierarchical Transformer encoder with Sobel edge augmentation yields consistent performance gains across architectures. The ablation study concludes that F1-scores increase by approximately 33% for U-Net and 13% for FPN relative to their baseline configurations. These results show that integrating Transformer-based feature encoding with edge-aware input augmentation improves aerial crack segmentation.</dc:description>
	<dc:publisher xml:lang="en-US">Universitas Mercu Buana</dc:publisher>
	<dc:date>2026-09-29</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://publikasi.mercubuana.ac.id/index.php/sinergi/article/view/36954</dc:identifier>
	<dc:identifier>10.22441/sinergi.2026.3.022</dc:identifier>
	<dc:source xml:lang="en-US">SINERGI; Vol. 30 No. 3 (2026); 981-994</dc:source>
	<dc:source>2460-1217</dc:source>
	<dc:source>1410-2331</dc:source>
	<dc:language>eng</dc:language>
	<dc:relation>https://publikasi.mercubuana.ac.id/index.php/sinergi/article/view/36954/12772</dc:relation>
	<dc:rights xml:lang="en-US">Copyright (c) 2026 SINERGI</dc:rights>
	<dc:rights xml:lang="en-US">https://creativecommons.org/licenses/by-sa/4.0</dc:rights>
</oai_dc:dc>